National Academies Press: OpenBook

Enhancing Internal Trip Capture Estimation for Mixed-Use Developments (2011)

Chapter: Chapter 3 - Findings and Applications

« Previous: Chapter 2 - Research Approach
Page 30
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 30
Page 31
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 31
Page 32
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 32
Page 33
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 33
Page 34
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 34
Page 35
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 35
Page 36
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 36
Page 37
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 37
Page 38
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 38
Page 39
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 39
Page 40
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 40
Page 41
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 41
Page 42
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 42
Page 43
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 43
Page 44
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 44
Page 45
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 45
Page 46
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 46
Page 47
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 47
Page 48
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 48
Page 49
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 49
Page 50
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 50
Page 51
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 51
Page 52
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 52
Page 53
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 53
Page 54
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 54
Page 55
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 55
Page 56
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 56
Page 57
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 57
Page 58
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 58
Page 59
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 59
Page 60
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 60
Page 61
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 61
Page 62
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 62
Page 63
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 63
Page 64
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 64
Page 65
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 65
Page 66
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 66
Page 67
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 67
Page 68
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 68
Page 69
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 69
Page 70
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 70
Page 71
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 71
Page 72
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 72
Page 73
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 73
Page 74
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 74
Page 75
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 75
Page 76
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 76
Page 77
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 77
Page 78
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 78
Page 79
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 79
Page 80
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 80
Page 81
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 81
Page 82
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 82
Page 83
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 83
Page 84
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 84
Page 85
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 85
Page 86
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 86
Page 87
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 87
Page 88
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 88
Page 89
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 89
Page 90
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 90
Page 91
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 91
Page 92
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 92
Page 93
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 93
Page 94
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 94
Page 95
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 95
Page 96
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 96
Page 97
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 97
Page 98
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 98
Page 99
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 99
Page 100
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 100
Page 101
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 101
Page 102
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 102
Page 103
Suggested Citation:"Chapter 3 - Findings and Applications." National Academies of Sciences, Engineering, and Medicine. 2011. Enhancing Internal Trip Capture Estimation for Mixed-Use Developments. Washington, DC: The National Academies Press. doi: 10.17226/14489.
×
Page 103

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

30 Chapter 1 provides background about MXDs and current practice in estimating internal trip capture for MXDs. It also describes what was available from existing data found from other sources. Chapter 3 describes the findings from the pilot studies and the compilation of usable data into the estimation procedure. Appendix B provides details about the land use classification system. Appendix C describes the data-collection methodology. Appendix D summarizes the experiences and lessons learned when conducting the surveys. Pilot Study Surveys The following are the results for the pilot study surveys conducted. Mockingbird Station Development Characteristics Mockingbird Station is a midtown mixed-use TOD in Dal- las, Texas, consisting of five primary land use types: residential, retail, office, restaurant, and cinema. Figure 2 shows an aerial photograph of Mockingbird Station. Figure 3 shows the site plan for Mockingbird Station. The site plan is of the second- floor level, but the notes describe what is on each of the levels of each building. Not shown is a parking garage beneath the surface parking area between the two north-south buildings; this garage serves the loft apartments. The parking shown at the north end of the site also extends below the two buildings on the north end of the site. That parking is available to all users. Lower portions of the garage in the west building are also open to any user; upper spaces are reserved for the office building. However, almost no one other than office building occupants or visitors was observed by the survey crew to have used this garage during survey periods. Mockingbird Station is bordered on the east by a Dallas Area Rapid Transit (DART) station and transit center on a light-rail line that splits just north of the station. Mockingbird Station has direct access to the station as well as to the transit center that is served by six bus routes. Five routs are year round; the sixth provides shuttle service to nearby Southern Methodist University (SMU) when school is in session. Bus service head- ways range from 6 to 20 minutes during weekday peak periods, from 20 to 45 minutes during daytime off-peak periods, and 45 to 60 minutes during the late evening hours. Bus service on most of the routes begins before 5 A.M. and continues until about 12:30 A.M. Two routes run slightly shorter schedules. The two light-rail lines have peak-period service ranging between 6 and 10 minutes, with daytime off-peak service ranging between 20 and 30 minutes and evening weekday service at about 30-minute headways. Mockingbird Station is bounded by Mockingbird Lane, a six-lane arterial on the south and US 75, the North Central Expressway, an eight-lane freeway on the west. To the north, Mockingbird Station is bounded by another development containing an office building and a health club. There is no vehicular access between the two developments, but there is a connecting pedestrianway about midway along the bound- ary between the two developments. Beginning about a block west of US 75 is the SMU cam- pus, which has a total enrollment of approximately 11,000 students. SMU students occupy several apartments in the area, although no percentage was available and rental rates were reported to be the highest in the area and beyond bud- gets of most students. Mockingbird Station has vehicular access along only Mockingbird Lane (two driveways) and the northbound frontage road of US 75 (one direct garage access and two additional driveways). Walk access is available from the east and north via conven- tional sidewalks adjacent to the street curbs. From the south and west, walk access requires crossing the very busy Mockingbird Lane intersection with the US 75 frontage roads. Walk access to the east is also available through the DART light-rail station and requires traversing stairs (of an elevation of about one building level) between the west side of the station and Mockingbird C H A P T E R 3 Findings and Applications

Source: Google Earth. Pedestrianway to Premier PREMIER MOCKINGBIRD STATION DART Station Figure 2. Mockingbird Station. Source: Selzer Associates. Figure 3. Mockingbird Station upper-level leasing plan and land uses, 2006.

Station’s ground level. There is no elevation change on the east side of the station. As noted previously, walk access is available to the middle of the development to the north via a walkway, which is also about one level above ground and is reachable by a stairway. An elevator is also available to reach these last two pedestrian connections; it is near the stairway to the DART station. Parking is provided in three garages and surface lots. One garage is reserved for office building use although its visitor spaces can also be used for reaching other Mockingbird Sta- tion destinations. A second garage serves the apartments. The third garage is for general use. The second and third garages are actually a single garage that has been partitioned into two facilities by a fence. During peak onsite activity periods, surface parking is usu- ally fully occupied and drivers circulate hoping to find a space close to the desired destination, but convenient garage parking was observed to be always available. There was no noticeable traffic congestion at any access point during the field surveys. The only congestion occurred occasionally in the parking lots due to excess circulation by drivers seeking a parking space. Access between the DART station and Mockingbird Sta- tion is very convenient. Walking distance between the station and the most distant building entrance is about 700 ft. Only the stairway is judged to present any challenge. Walking within Mockingbird Station is very easy and con- venient. Although few sidewalks are much more than 10-ft wide, there are no obstacles except where three restaurants have set up outdoor tables and left fairly narrow walking widths. However, those constraints did not present deterring bottlenecks. No special provisions have been made for bicy- cle access. Figure 4 shows an example of sidewalk provisions at the entrance to the apartment building. Table 31 shows the occupied development in Mockingbird Station. The combined retail and restaurant space and the apartments are more than 90% occupied, and the office space is about 80% occupied. The development appears to be mature and has been in operation long enough to be experiencing initial turnover of tenants that are not correctly positioned in the local market. About one-third of Mockingbird Station’s occupied floor space is residential, and another third is retail. More than 20% is office with the remainder split between the restaurants and the cinema. The residential is high-end rental. Mocking- bird Station has no major retailers. All have 15,000 sq ft or less. The retail is primarily specialty women’s apparel. The restaurants represent a range of middle- to upper-priced sit- down and convenience offerings, including an ice cream shop and a specialty coffee shop. 32 Source: Texas Transportation Institute Figure 4. Entrance to Mockingbird Station lofts. Land Use Occupied Development Units Largest Residential 191 DU, 192,940 sq ft 84% one-bedroom Retail 156,100 sq ft Two specialty apparel stores of 15,000 sq ft only stores over 10,000 sq ft Office 114,600 sq ft All in one building Restaurant 28,900 sq ft Largest about 8,800 sq ft Cinema 31,500 sq ft, 8 screens Parking 1,528 spaces Table 31. Mockingbird Station development.

Travel Survey The survey of travel characteristics focusing on internal trip capture was conducted on Tuesday afternoon through Thurs- day morning, May 9–11, 2006. The primary objective was to quantify the percentage of internal trip capture during week- day peak periods in a manner that would support the proposed methodology to estimate internal capture using component land use quantities and reflect mode of original access and the degree of internal connectivity. The survey was designed to be adaptable to a variety of mixed-use areas. Mockingbird Station was the first site sur- veyed, and a second site was proposed with somewhat differ- ent characteristics. At the time, permission for a second site had not been secured, so specific survey requirements for that site were not known; however, it was known that while there was a standard survey method to be used, some customizing might be needed to fit other sites. The essential requirement was to produce comparable data for each survey site. For Mockingbird Station, the following travel data were collected for peak periods between 6:30 A.M. and 10 A.M. and between 4 P.M. and 7 P.M.: • Multimodal cordon count covering all access points; • Counts of people entering and exiting doors of each build- ing or business being surveyed during a particular period; • Exit interviews of people as they departed selected doors; and • Interviews of people leaving the DART rail station and tran- sit center (customer survey to respond to local conditions). The exit interviews were the primary information source. The counts were used to factor interview results. The DART sta- tion interviews were used to provide a more complete indica- tion of who was using transit. All interviews were conducted recognizing that the results would be a sample of all people exit- ing during a time period. Over the complete duration of the sur- vey, interviews were conducted at all entrances that were open during the survey periods (a few secondary entrances were kept locked by businesses). Survey supervisors selected the entrances to be surveyed during each period, and interviews were con- ducted at those entrances for complete periods. In some cases, the business activity was low so interviewers were assigned to cover multiple entrances and to intercept and interview any exiting patron they could. During the A.M. peak period, the only businesses open during the full period were the office building and a Star- bucks coffee shop. One other business opened at 9 A.M. while the remainder opened at 10 A.M. (restaurants at 11 A.M.). During the P.M. peak period, all businesses were open for the complete survey period. As a result, all entrances could be fully covered during the A.M. peak, but P.M. interviews cov- ered some entrances one day and the remainder the second day, although interviews were conducted at some locations both days. Interviewers and counters were trained for several hours prior to the first afternoon’s surveys. Each was observed dur- ing the first hour in the field (i.e., an hour before the actual data were going to be used) and adjustments were made as needed. Supervision continued throughout the survey period. In a few cases, interviewers were moved to locations that were more active or better suited the interviewer’s particular skills (e.g., more mobile to cover several entrances). In another few cases, interviewers were reassigned to perform counts to optimize results. Interview forms were reviewed during each shift and then checked more completely at the end of each shift. Any errors or missing data were checked with the interviewer either by phone and/or prior to starting the next shift. Incomplete and erroneous interviews were not used. The same process was used for the counts although those checks were much more straightforward. Survey personnel who did not perform ade- quately were released and not used again. The survey crew was initially overstaffed with the expectation that some would be released, so there was no need to add new personnel and repeat the training. Survey Results Most of the findings are based on 761 completed exit inter- views conducted during two morning and two afternoon peak periods. Of these approximately 30% were obtained during the A.M. peak and 70% during the P.M. peak. The completed and usable interviews covered an average of 33% of people exiting buildings during the A.M. peak period and about 11% during the P.M. peak period. Table 32 shows the numbers of interviews completed and usable for each peak period and land use cate- gory. Most interviews yielded one usable trip made during one of the peak periods; some interviews yielded two trips. The A.M. interview percentage was higher than the P.M. percentage because A.M. activity was lower and a similar number of inter- viewers were available near each interview location. The A.M. population also included more regulars and fewer occasional visitors, which resulted in interviewees who were more com- fortable with being interviewed in the morning. The interview forms included questions not only about the exit trip, but also about the trip made to the location just being departed (see Appendix C for forms closely resembling the forms used in this pilot survey). If the inbound trip to the sur- vey location occurred during the survey period, it could be used as part of the survey database if the information was sufficiently complete. Most inbound trips preceding exiting trips occurred before the survey period or lacked complete information. Table 33 shows the number of usable trips that were derived from the usable interviews (a usable interview was defined as 33

one that contained at least one fully reported trip). For the A.M. peak, total usable surveyed trips accounted for about 36% of all counted exit movements. For the P.M. peak period, about 13% of the counted trips are represented with usable interview information. Table 33 information provides the basis for factoring the survey data to represent all peak-period trips made. That expansion is needed to permit an estimate of the number of internal trips. The results reported herein are based on factor- ing to reflect sampling at each building entrance; factoring was performed by land use for each peak period. The survey results were summarized for the A.M. and P.M. peak periods. Table 34 shows a different summary of completed inter- views, exiting people, and usable trips derived from the inter- views. Respondents were asked about not only the trips that they were in the midst of making as they exited from an estab- lishment, but also the trip they had previously made to that same place. The total of the reported trips, if made during one of the two peak periods, are shown as usable trips in Table 34. Some of the reported inbound trips occurred outside the peak periods, but for many of those trips, the respondent was unable or unwilling to provide enough complete information to make the inbound trip usable. Finally, some otherwise complete interviews were not usable because the inbound trip reported was not actually the immediately previous trip—for example, some respondents thought they were being asked for the first trip of the day onsite or to the site and not the immediately previous trip to the establishment they were just leaving. Many of those trips were made outside the peak peri- ods. First trips of the day from the onsite apartments did not have a previous trip that day. The interviews reported in Table 34 differ slightly from interviews reported in prior tables because the interviews reported in Table 34 are associated with the land use for which an exit trip is reported. Hence, if an interview that was reported in Table 32 has a valid entering trip but not a valid exiting trip 34 A.M. Peak Period P.M. Peak Period Land Use Interviews1 Exit Movements Percent Interviewed Interviews 1 Exit Movements Percent Interviewed Office 49 130 38% 78 275 28% Retail1 — — — 285 2,311 12% Restaurant 146 395 37% 104 1,560 7% Residential 43 188 23% 34 218 16% Cinema2 — — — 22 220 10% Hotel3 — — — — — — Total 238 713 33% 523 4,584 11% 1 Number of interviews conducted with travelers exiting doors of a particular land use that contained at least one usable trip. 2 Retail and cinema not open during morning peak period. 3 No onsite hotel at Mockingbird Station. Table 32. Peak-period interviews, exit movements, and percent interviewed—Mockingbird Station. A.M. Peak Period P.M. Peak Period Land Use Usable Trips1 Exit Movements Percent Usable Usable Trips1 Exit Movements Percent Usable Office 59 130 45% 85 275 31% Retail2 — — — 307 2,311 13% Restaurant 147 395 37% 108 1,560 7% Residential 51 188 27% 49 218 23% Cinema2 — — — 24 220 11% Hotel3 — — — — — — Total 257 713 36% 573 4,584 13% 1 Must include specific origin location, location of destination, and land use of destination if internal; this total includes reported exiting and entering trips made this period.. 2 Retail and cinema not open during morning peak period. 3 No onsite hotel at Mockingbird Station. Table 33. Peak-period usable trips, exit movements, and percent usable—Mockingbird Station.

(e.g., incomplete information), that interview is reported in Table 32 for the land use where the interview occurred, but reported for the trip origin land use in Table 34. Table 34 points out one final lesson learned from the sur- vey procedures used in the pilot studies. To obtain accurate inbound trip information while conducting exit surveys, it is necessary to increase the amount of interview practice for each interviewer (i.e., mock interviews with trainers). Inter- viewers recorded too many incomplete interviews and incor- rect previous trips. Some interviewers also failed to ask or record responses for all of the questions about the inbound trip, resulting in more incomplete inbound trip information. However, since only trips that occurred during the two peak periods were of interest and since some of the respondents’ inbound trips occurred outside the two peak periods, it was expected that inbound trips would be fewer than outbound trips that are directly surveyed. The only way to obtain simi- lar samples of inbound trips is to interview people as they enter an establishment—something management declined to approve at all three pilot study sites. Table 34 shows that a few A.M. interviews were completed at retail outlets. These were primarily employees and deliver- ies. However, the project panel agreed that the number of interviews was too small to provide a representative sample and that the results would not be representative of retail stores that might be open during the A.M. peak period (generally convenience retail or grocery or drug stores), so those data are not reported in other tables. Table 35 shows for the A.M. peak period the total number of people exiting from each land use. People could exit in one of two ways: (1) from a door of the establishment to the side- walk in front of the establishment or (2) from the establish- ment directly to a parking garage via an internal access way and then drive out of the garage and off the site without an opportunity to be interviewed. A sample of the first group was interviewed. None of the second group was interviewed because they immediately became external trips and could be directly categorized in that manner. Table 35 also has a column labeled un-surveyed locations. That column does not apply for Mockingbird Station, but does apply to two other pilot survey sites. Numbers in that column represent the numbers of people counted exiting establishments where no interviews were taken. Table 35 shows that a number of people exiting the office building and loft apartments did so by going internally to their garage parking space and then driving out of Mockingbird Sta- tion. As mentioned above, all of these trips were classified as external trips; drivers and passengers did not need to be inter- viewed to get the needed information since one garage exits only outside the development’s boundary, and the driving dis- tance from the other garage to internal locations is longer than walking. The right column shows the percentage of all exiting trips represented by survey information—either a completed interview or a count of vehicles and occupants exiting the site from garages with internal access. The interviewed and direct garage trips accounted for about half of all exiting trips. Table 36 shows similar information but for the P.M. peak period. The direct exits from the site establishments through the garages accounted for a much smaller percentage of the total trips. The resulting surveyed percentage of total trips is about 22% in total, but ranges between 7 and 73% by land use. The survey samples for Mockingbird Station and all other surveyed developments were factored in the same manner. Interviews were expanded to represent the door counts by 35 Morning Peak Period Afternoon Peak Period Land Use Interviews4 Exit Movements Percent Interviewed Usable Trips5 Interviews 4 Exit Movements Percent Interviewed Usable Trips5 Office 44 130 34% 50 68 275 25% 70 Retail1 8 18 50% 11 292 2,311 13% 368 Restaurant 146 395 37% 165 85 1,560 5% 105 Residential 33 188 18% 33 28 218 13% 30 Cinema2 — — — — 22 220 10% 22 Hotel3 — — — — — — — — Total1 231 731 32% 259a 495 4,584 11% 595b 1 Retail trips subsequently removed from further analysis since all stores closed during this period. 2 No interviews attempted at cinema during the morning peak period since cinema was closed. 3 No onsite hotel at Mockingbird Station. 4 Number of interviews conducted with travelers exiting doors of a particular land use that contained at least one usable trip. 5 Must include specific origin location, location of destination, and land use of destination if internal; this total includes reported exiting and entering trips made this period. a Includes 2 movements counted at establishments where too few interviews were completed for valid sample. b Excludes 22 movements counted at establishments where too few interviews were completed for valid sample. Table 34. Peak-Period interviews, exit movements, percent interviewed, and usable trips—Mockingbird Station.

land use. Where door counts were not available for all estab- lishments within a land use classification, development units were used as a basis for expanding door counts to cover all floor space of a classification. Direct movements to and from inside buildings to external locations were handled through direct counts. The complete discussion is contained in Appendix G. Table 37 shows data for entering trips that resembles the contents of Tables 35 and 36. A sample of persons entering from the DART rail station and transit center was inter- viewed; numbers of those persons are shown by the destina- tion land use. The first three columns under each time period accounted for people who entered through either the estab- lishments’ outside doors (and are represented by expanded interviews at those doors or at the DART station) or an inter- nal access from a parking garage. The last column shows the remaining people who were counted upon entry but are not represented in the first three columns of the table. All these were considered to be from external origins since they did not have an internal trip origin represented in an interview. These 36 Exit Movements Land Use Surveyed Trips3 Doors Unsurveyed Locations4 Garage Direct5 Total Percent Surveyed6 Office 59 130 — 34 164 57% Retail1 — — — — — — Restaurant 147 395 — — 395 37% Residential 51 188 — 216 404 66% Cinema1 — — — — — — Hotel2 — — — — — — Total 257 713 — 250 963 53% 1 Retail and cinema did not actively generate trips during the morning peak period. 2 No onsite hotel at Mockingbird Station. 3 Number of usable trip origins at each land use recorded from traveler interviews. 4 Includes locations where no interviews were attempted (prorated by sq ft) and locations where door counts were made, but no usable trip origins were recorded on interviews. 5 Person-trips observed exiting onsite parking garages, assumed to be traveling directly to an external location. external street system. 6 Includes those trips described in usable interviews or direct exits from a parking garage to the Table 35. Morning peak-period surveyed trips, exit movements, and percent surveyed—Mockingbird Station. Exit Movements Land Use Surveyed Trips3 Doors Unsurveyed Locations4 Garage Direct5 Total Percent Surveyed6 Office 85 275 — 416 691 73% Retail1 307 2,311 — — 2,311 13% Restaurant 108 1,560 — — 1,560 7% Residential 49 218 — 144 362 53% Cinema1 24 220 — — 220 11% Hotel2 — — — — — — Total 573 4,584 — 560 5,144 22% 1 Retail and cinema did not actively generate trips during the morning peak period. 2 No onsite hotel at Mockingbird Station. 3 Number of usable trip origins at each land use recorded from traveler interviews. 4 Includes locations where no interviews were attempted (prorated by sq ft) and locations where door counts were made, but no usable trip origins were recorded on interviews. 5 Person-trips observed exiting onsite parking garages, assumed to be traveling directly to an external location. external street system. 6 Includes those trips described in usable interviews or direct exits from a parking garage to the Table 36. Afternoon peak-period surveyed trips, exit movements, and percent surveyed—Mockingbird Station.

trips account for approximately one-third of the A.M. peak- period entering trips and about 40% of the P.M. peak-period entering trips. Table 38 shows the mode split of person trips to Mocking- bird Station during the A.M. peak period. Personal vehicles (drivers and passengers) account for about three-quarters of the person trips to and about 70% from Mockingbird Station during the A.M. peak period. The A.M. peak-hour exiting per- centages by personal vehicle are about 5% lower than during the A.M. peak period. Table 39 shows similar information for the P.M. peak period. Transit is a major mode of access for Mockingbird Station. About 15% of inbound and 11% of outbound A.M. peak- period trips use DART rail or bus transit. The peak hour per- centages are slightly higher. During the P.M. peak period, transit accounts for about 13% of inbound and 19% of out- bound trips. Peak hour percentages are approximately simi- lar. The larger outbound percentage reflects employees who came by transit in the morning in addition to the evening vis- itors who come and leave by transit. Transit accounts for a significant amount of the trips during both peak periods, attributable at least in part to the proximity of the DART light-rail station and bus transfer center adjacent to Mocking- bird Station. Walk trips also account for more than might be expected in a midtown area, with 5 to 15% walking to or from Mocking- bird Station. Although Mockingbird Station is close to SMU, bicycle trips were negligible as were trips by motorcycle. There are limited street crossings of US 75. All are heavily used by traffic and there are no bike lanes. SMU students tend to be more affluent. All these factors may explain the low bicycle share of peak period trips to and from Mockingbird Station. 37 Morning Peak Period Afternoon Peak Period Land Use Survey3 Garage Direct4 Transit Direct5 Balance 6 Total Survey3 GarageDirect4 Transit Direct5 Balance 6 Total Office 101 382 91 110 684 69 126 12 56 263 Retail1 — — — — — 787 — 129 256 1,172 Restaurant 167 — 29 196 392 380 — 170 1,051 1,601 Residential 12 48 5 138 203 161 236 18 34 449 Cinema1 — — — — — 79 — 106 171 356 Hotel2 — — — — — — — — — — Total 280 430 125 444 1,279 1,476 362 435 1,568 3,841 1 Retail and cinema did not actively generate trips during the morning peak period. 2 No onsite hotel at Mockingbird Station. 3 Trip destinations recorded from exit interviews, expanded as described. 4 Person-trips observed entering onsite parking garages, assumed to be traveling directly from an external location. 5 Trips entering onsite land uses from external locations recorded on transit interviews. 6 Balance of person-trips entering onsite land uses; assumed to originate externally. Table 37. Peak-period person-trips entering land uses—Mockingbird Station. Peak Period (7:00 A.M.–10:00 A.M.) Peak Hour (7:45 A.M.–8:45 A.M.) Trips Percent3 Trips Percent Travel Mode Entering Exiting Entering Exiting Entering Exiting Entering Exiting Personal Vehicle1 670 361 76% 70% 280 129 77% 65% Motorcycle 0 0 0% 0% 0 0 0% 0% Delivery Vehicle 39 17 4% 3% 12 3 3% 2% Transit2 128 57 15% 11% 58 28 16% 14% Walk 42 79 5% 15% 15 38 4% 19% Bicycle 0 0 0% 0% 0 0 0% 0% Total All Modes 879 514 100% 100% 365 198 100% 100% 1 Personal vehicle occupancies (entering/exiting): peak period 1.08/1.11; peak hour 1.09/1.11. 2 Transit trips include light rail and bus. 3Percentage totals may not sum to 100 due to rounding. Table 38. A.M. peak-period and peak-hour cordon person-trip count and mode split—Mockingbird Station.

Vehicle occupancies were higher during the P.M. peak (more than 1.2) than for the A.M. peak (about 1.1). This is attribut- able to people going shopping or to restaurants or the cinema. Table 40 shows (1) the number and percent of internal person trips each peak period and (2) the total person trips generated by each land use type and those that are internal to Mocking- bird Station. For example, during the A.M. peak period, 64% of trips leaving the office building are destined for internal desti- nations. Similarly, 15% of the A.M. inbound trips come from origins within Mockingbird Station. Note that the only uses active during the A.M. peak were the apartments, the office building, a coffee shop, and a mobile phone store that opened at 9 A.M. All other businesses opened at 10 A.M., although a few employees and delivery people entered before that time. Hence, most of Mockingbird Sta- tion was inactive during the A.M. peak period. Table 40 shows that for the A.M. peak, about 22% of the inbound and 31% of the outbound trips were internal, excluding trips between similar uses (e.g., from retail to retail). Internal trips between similar uses have been excluded (from both internal and total trips) because they are not included in trip generation estimates used for TIS, which are based on trips entering and leaving a site. The office building has about 64% of its trips destined for internal destinations. Nearly all of those were to a coffee shop located less than 300 ft from the office building. The office building did not contain a snack shop, so a strong linkage developed with the coffee shop. This may not always be the case with other types of restaurants. Note that most office building trips during the A.M. peak are inbound; only about 17% of the trips are outbound, so the high percentage of internal trips does not reflect a high number of internal trips in this case. 38 Peak Period (4:00 P.M.–7:00 P.M.) Peak Hour (5:00 P.M.–6:00 P.M.) Trips Percent Trips Percent Travel Mode Entering Exiting Entering Exiting Entering Exiting Entering Exiting Personal Vehicle1 1,292 1,208 76% 74% 456 443 77% 73% Motorcycle 12 5 1% <1% 2 1 <1% <1% Delivery Vehicle 24 21 1% 1% 8 4 1% 1% Transit2 225 301 13% 19% 71 131 12% 21% Walk 153 83 9% 5% 55 32 9% 5% Bicycle 4 6 <1% <1% 0 0 0% 0% Total All Modes 1,710 1,624 100% 100% 592 611 100% 100% 1 Personal vehicle occupancies (entering/exiting): peak period 1.25/1.22; peak hour 1.26/1.21. 2 Transit trips include light rail and bus. Table 39. P.M. peak-period and peak-hour cordon person-trip count— Mockingbird Station. A.M. Peak Period P.M. Peak Period Entering Exiting Entering Exiting Land Use Trips PercentInternal Trips Percent Internal Trips Percent Internal Trips Percent Internal Office 684 15% 142 64% 263 26% 669 15% Retail1 — — — — 1,172 67% 1,284 32% Restaurant 392 43% 371 28% 1,601 22% 1,519 46% Residential 203 5% 388 22% 449 36% 361 43% Cinema1 — — — — 356 22% 220 50% Hotel2 — — — — — — — — Total All Trips 1,279 22% 901 31% 3,841 38% 4,053 36% 1 Retail and cinema not open during morning peak period. 2 No onsite hotel at Mockingbird Station. Table 40. Peak period person-trips and percent internal trip capture by land use—Mockingbird Station.

The P.M. peak period internal trip capture percentages are somewhat higher, with about 38% of the inbound and 36% of the outbound trips being internal. Table 41 shows the A.M. peak period internal trip capture for outbound trips by land use. Since Mockingbird Station is fully and conveniently walkable, there are virtually no driving trips (although a few people were observed driving a few hundred feet from one end of a parking area to the other). Since no internal transit is provided, there are no internal trips by transit. Table 42 shows the same information for the P.M. peak period. For both A.M. and P.M. peak periods, it appears that there are a few stronger linkages between land use pairs and several modest linkages. However, note that the inter- nal trip capture percentages are a result of inherent inter- action between given land use pairs as well as the quantities and proximities of each. This is discussed elsewhere in this report. Table 43 shows the distribution of internal origins resulting from inbound trips. For example, of trips inbound to the office building, 1% come from the onsite residential units. This shows that while the coffee shop has a strong interaction with the office building, it also has a stronger interaction with the residential apartments that result in several trips to the coffee shop. However, this table shows that the residents then proceed from the coffee shop to off-site destinations; few return home. Table 44 shows similar data for the P.M. peak period. As with the interactions shown in Table 42, there are a few strong relationships and a number of minor relationships. The cases and relationships are discussed in a subsequent chapter. Table 45 shows the percent of trips made into and out of Mockingbird Station buildings that are internal for each mode of travel. Only a small percentage of vehicle driver trips are internal. Not surprisingly, a very high percentage of walk trips are internal. 39 Internal Destination Land Use Destination Summary Origin Land Use Office Retail2 Restaurant Residential Cinema2 Hotel3 Internal External Total TotalTrips Office —1 — 63%a 1% — — 64% 36% 100% 142 Retail2 — —1 — — — — — — — — Restaurant 25% — —1 3 — — 28 72 100 371 Residential 2 — 20 —1 — — 22 78 100 388 Cinema2 — — — — —1 — — — — — Hotel3 — — — — — —1 — — — — All Origins 11% — 19% 1% — — 31% 69% 100% 901 1 Internal trips within a land use are not included in internal trip capture methodology. 2 Retail and cinema not open during morning peak period. 3 No onsite hotel at Mockingbird Station. a Chain specialty coffee shop close to office building. Table 41. Percent distribution of internal trip destinations for trips exiting Mockingbird Station buildings—A.M. peak period. Internal Destination Land Use Destination Summary Origin Land Use Office Retail Restaurant Residential Cinema Hotel2 Internal External Total TotalTrips Office —1 9% 4% 2% 0 — 15% 85% 100% 669 Retail 1% —1 20 7 4% — 32 68 100 1,284 Restaurant 3 38 —1 3 2 — 46 54 100 1,519 Residential 1 31 11 —1 0 — 43 57 100 361 Cinema 0 17 25 8 —1 — 50 50 100 220 Hotel2 — — — — — —1 — — — — All Origins 2% 19% 9% 4% 2% — 36% 64% 100% 4,053 1 Internal trips within a land use are not included in internal trip capture methodology. 2 No onsite hotel at Mockingbird Station. Table 42. Percent distribution of internal trip destinations for trips exiting Mockingbird Station buildings—P.M. peak period.

40 Internal Origin Land Use Origin Summary Destination Land Use Office Retail2 Restaurant Residential Cinema2 Hotel3 Internal External Total TotalTrips Office —1 — 14% 1% — — 15% 85% 100% 684 Retail2 — —1 — — — — — — — — Restaurant 23% — —1 20 — — 43 57 100 392 Residential 0 — 5 —1 — — 5 95 100 203 Cinema2 — — — — —1 — — — — — Hotel3 — — — — — —1 — — — — All Destinations 7% — 8% 7% — — 22% 78% 100% 1,279 1 Internal trips within a land use are not included in internal trip capture methodology. 2 Retail and cinema not open during morning peak period. 3 No onsite hotel at Mockingbird Station. Table 43. Percent distribution of internal trip origins for trips entering Mockingbird Station buildings—A.M. peak period. Internal Origin Land Use Origin Summary Destination Land Use Office Retail Restaurant Residential Cinema Hotel2 Internal External Total TotalTrips Office —1 5% 19% 2% 0 — 26% 74% 100% 263 Retail 5% —1 50 9 3% — 67 33 100% 1,172 Restaurant 1 16 —1 2 3 — 22 78 100% 1,601 Residential 3 19 10 —1 4 — 36 64 100% 449 Cinema 1 14 7 0 —1 — 22 78 100% 356 Hotel2 — — — — — —1 — — — — All Destinations 2% 11% 18% 4% 3% — 38% 62% 100% 3,841 1 Internal trips within a land use are not included in internal trip capture methodology. 2 No onsite hotel at Mockingbird Station. Table 44. Percent distribution of internal trip origins for trips entering Mockingbird Station buildings—P.M. peak period. A.M. Peak Period P.M. Peak Period1 Entering Exiting Entering Exiting Mode of Travel Trips PercentInternal Trips Percent Internal Trips Percent Internal Trips Percent Internal Automobile Driver 857 2 526 3 1,941 4 1,815 5 Automobile Passenger 32 0 35 0 72 0 212 0 Taxi/Car Service 0 0 0 0 0 0 0 0 Transit (Bus) 0 0 12 0 0 0 88 0 Transit (Light Rail) 125 0 22 0 435 0 413 0 Walk/Bicycle 266 100 307 87 1,367 100 1,500 91 1 Travel mode not reported for 26 entering and 25 exiting trips. Table 45. Peak-period person-trips and percent internal trip capture by mode of travel—Mockingbird Station.

Table 46 shows different information. The table shows internal trips by the original mode of access to Mockingbird Station—for example, the first row of the table shows that for the A.M. peak period, of the exiting people who came to Mockingbird Station as vehicle drivers, 31% of them went to internal destinations. The purpose of this table is to deter- mine whether people arriving by different modes have differ- ent internal trip-making tendencies. The cells that have larger numbers of trips provide the most useful comparisons. The cells that contain more than 300 trips have consistent inter- nal trip percentages ranging from 20% to 37% internal trips, but even though most numbers of total trips are small, those who arrive by bus or walking/bicycling are much more prone to making more internal trips, perhaps due to fewer options. The apparent tendency of rail transit riders to make fewer internal trips—at least during peak periods—may reflect that many of them use transit to commute to work and are less likely to make internal peak-period trips during peak periods (e.g., some of those trips may be made during midday). The small numbers of people who walk, bike, or ride buses to Mockingbird Station make further analysis speculative. The other two MXDs surveyed for this project had fewer transit rid- ers, so they do not provide significant insight into this question. Table 47 attempts to explore whether having a personal vehicle available for trips affected internal travel tendencies. One might assume that a person with no vehicle available would have fewer options to make off-site (external) trips, so they would make more internal trips. Accounting for effects of the sample sizes shown, the table does not support that assumption: internal trip capture does not exceed that for people with access to drive personal vehicles. Findings from surveys of two other MXDs—Atlantic Sta- tion and Legacy Town Center—follow. Analyses of relation- ships between the findings and causal factors are described in subsequent chapters. Atlantic Station Development Characteristics As with Mockingbird Station, Atlantic Station is a midtown redevelopment/infill project, but it is substantially larger and is spread over several blocks rather than being on a single block. Figure 5 shows an illustrative site plan of Atlantic Station when it is fully complete. A dashed outer boundary line shows the outer limits of the portions that had been completed and 41 A.M. Peak Period1 P.M. Peak Period2 Entering Exiting Entering Exiting Mode of Access Trips PercentInternal Trips Percent Internal Trips Percent Internal Trips Percent Internal Vehicle Driver 1,058 20% 697 31% 2,847 35% 2,694 37% Vehicle Passenger 32 0% 34 0% 130 45% 354 23% Taxi/Car Service 0 0% 0 0% 0 0% 0 0% Transit (Bus) 7 100% 7 100% 60 100% 118 51% Transit (Light Rail) 131 5% 23 26% 586 26% 654 23% Walk/Bicycle 10 100% 13 77% 71 100% 157 45% 1 Access mode not reported for 41 entering trips and 177 exiting trips. 2 Access mode not reported for 79 entering trips and 244 exiting trips. Table 46. Peak-period person-trips and percent internal trip capture by mode of access—Mockingbird Station. A.M. Peak Period1 P.M. Peak Period2 Entering Exiting Entering Exiting Vehicle Access Trips PercentInternal Trips Percent Internal Trips Percent Internal Trips Percent Internal Yes (Vehicle Driver) 1,098 20% 720 30% 265 29% 462 51% Yes (Non-Vehicle Driver) 118 14% 31 52% 318 33% 571 59% No Vehicle Access 18 28% 16 31% 3,094 35% 2,795 32% 1 Automobile access not reported for 45 entering trips and 134 exiting trips. 2 Automobile access not reported for 164 entering trips and 225 exiting trips. Table 47. Peak-period person-trips and percent internal trip capture by vehicle access—Mockingbird Station.

occupied at the time surveys for this project were completed. The area inside the inner dashed boundary line has not been developed although some of the parking to serve that devel- opment has been completed. Although complete, the survey conducted for this project did not include the IKEA store on the far west end because it was viewed as a non-integrated, free-standing component of Atlantic Station oriented away from the remainder of the development. Figure 6 shows the street names and sectors of Atlantic Station. Atlantic Station consists of three adjacent sectors: the District, the Commons, and the Village. The District. The District is clearly the heart and most active part of Atlantic Station. It is the densest and has six interactive land uses. The mostly commercial District is on a grid of blocks extending from the south side of 14th Street to 20th Street and from Fowler Street on the east to State Street on the west. Virtually the entire area north of 17th Street has three levels of parking below ground. All parking is contigu- ous although parking for a few buildings has been partitioned. The garages are designed so the streets on the surface are dupli- cated underground for ease of navigation and comprehension. There is also short-term metered curb parking on most blocks of the surface level. All spaces are pay parking although busi- nesses have the option of validating parking for specific durations. Visitor parking is free for the first 2 hours, $2 up to 3 hours, $3 up to 4 hours, $5 up to 5 hours, then increas- ing $3 per hour to the daily maximum of $14 for more than 7 hours. Employees park free on the lowest garage level dur- ing their work hours. At the time of the survey, garage park- ing supply far exceeded demand. Surface curb parking was generally fully occupied during normal business hours. Land uses in the District consist of residential, retail, office, restaurant, hotel, and cinema. Table 48 shows the number of development units of each type. The largest retailers are Dillard’s (department store), Publix (grocery store), and a shoe store. The retailers provide a range of products similar to what can be found in a regional mall. Restaurants range from specialty coffee shops to high-end shops. Residential units open at the time of the survey are in one high-rise build- ing at the corner of 17th and State Streets and along both sides of 16th Street (townhouses). The office space is all in one high-rise building at 17th and Market Streets, while the restaurants and retail are distributed across most of the District. The entire area north of 17th Street is conveniently walkable due to general compactness; short block lengths (about 150 to 300 ft in most cases); and an attractive walking environ- 42 Source: www.atlanticstation.com/images/SitePlan_large.jpg. Image is used by permission: ©2010 Atlantic Station, LLC. All rights reserved. Figure 5. 2006 Atlantic Station site plan at buildout.

ment. There is a grade between 16th and 17th Streets that may discourage some from walking, but walking is viable for most people. The Commons. This area includes two multistory apart- ment complexes along the north (Park District) and south (Art Foundry) sides of 17th Street, which has a wide median in the middle of the area. Resident parking is beneath the residential units, with visitor parking along 17th Street. The walking environment is typical for urban areas. Typical sidewalks are provided along 17th Street. There are no special provisions other than banners that integrate The Commons with The District. Other than banners and a few signs, the two areas could easily pass for being totally disassociated. The Village. This western sector contains only one build- ing: an IKEA furniture store. The IKEA building faces away from 17th Street and the remainder of Atlantic Station. It has its own parking and, as with the Commons, no strong con- nection to the District. Although walking between IKEA and the Commons is convenient by conventional sidewalk, little pedestrian activity was observed. Access Vehicular access to Atlantic Station and the District is con- centrated on 17th Street from both east and west. I-75 and I-85 merge just to the north of 17th Street so Atlantic Station has good regional access by motor vehicle. An interchange 43 Source: www.atlanticstation.com/site_parking.php. Image is used by permission: ©2010 Atlantic Station, LLC. All rights reserved. Figure 6. Schematic map of Atlantic Station, 2006. Land Use Occupied Development Units Largest Residential 798 DU • District: 190 apartments at 17th and State; 55 townhomes south of 16th Street • Commons: 553 apartments Retail 434,500 sq ft • Department store 227,000 sq ft • Grocery store 30,300 sq ft • Shoe store 27,000 sq ft • Only stores over 12,000 sq ft Office 550,600 sq ft • Almost all in one building Restaurant 64,600 sq ft • Sports bar/restaurant 19,100 sq ft • Only restaurant over 10,000 sq ft Hotel 101 rooms • One hotel Cinema 87,000 sq ft, 16 screens, 6,000 seats • One cinema Table 48. Atlantic Station Development (all units within the District except as noted).

exists serving 14th and 17th Streets. Access to regional arte- rial streets is via 17th Street. There is some additional access to and from the south via local streets between Fowler and State. Access to underground parking of The District is from Fowler on the east, 16th Street on the south, State Street on the west, and 20th Street on the north. There are no ramps between the three levels; all access is to a single level. Some of the largest buildings are garage-accessible only from one or two of the three levels; however, there are stairs, elevators, and an escalator providing access between all garage levels and the street level. Stairs are spaced closely so that stairway access is quite convenient. Vehicular access to The Commons and IKEA is primarily via 17th Street although 16th Street also provides access. Atlantic Station has two types of transit access as Figure 7 shows. Atlantic Station provides a dedicated free shuttle between the District and the Metropolitan Atlanta Regional Transit Authority (MARTA) Art Center rail station that is about 1⁄2 mile to the east of the District. The shuttle uses 17th Street but loops through the District. The shuttle operates on 5- to 10-minute headways, between 5 A.M. and 1 A.M., covering MARTA’s rail system hours of 5 A.M. to 1 A.M. MARTA also provides conventional bus service along 17th Street and along 14th Street. The 17th Street service (Route 23) also connects with the Art Center Station on the east. Going west and south from Atlantic Station, this route extends past Georgia Institute of Technology (Georgia Tech) and to down- town. It runs on 30- to 40-minute headways during the day, operating between about 5:30 A.M. and midnight. The 14th Street service runs between the Arts Center Station and the MARTA rail west end station, providing service to the Geor- gia Tech campus and an additional rail station. This route (Route 98) operates on 14th Street east of State Street and has approximately 40-minute headways all day. Two additional local routes run north–south on Northside Drive just west of the IKEA store. Those routes operate on 60-minute headways between about 6 A.M. and 10 P.M. Travel Survey The travel survey was conducted in the same manner as for Mockingbird Station. Surveys were conducted on Tuesday afternoon July 11, 2006, through Thursday morning July 13, 2006. The surveys conducted were 44 Source: www.atlanticstation.com/site_parking.php. Image is used by permission: ©2010 Atlantic Station, LLC. All rights reserved. Figure 7. Atlantic Station transit access, 2006.

• Multimodal cordon count covering all access points of the District and the Bezar townhome area plus the parking garages for the Art Foundry and Park District apartment complexes; • Counts of people entering and exiting doors of each building or business being surveyed during a particular period; • Exit interviews of people as they departed selected doors; • Pedestrian intercept interviews at one sidewalk location between the District and the Commons to catch those not included in the interviews in the Commons apartment com- plexes (custom survey to respond to local conditions); and • Interviews of people using the Atlantic Station shuttle (cus- tom survey to respond to local conditions). MARTA bus patrons were not interviewed specifically because there were very few observed. The Atlantic Station shuttle seemed to be functioning as an almost complete sub- stitute for conventional transit access. Interviews were conducted in a manner similar to that used for Mockingbird Station, with one exception: the large num- ber of businesses and entrances precluded all being covered. All large businesses were covered as was a sample of smaller ones. Those not surveyed (including a small number that declined permission) were accounted for by including expansion factors using applicable development units (e.g., sq ft). The research team was told by onsite management that occupants of the two residential developments in the Commons were similar and provided permissions for only one complex. Dwelling units were used to apply survey results to cover those units. Survey Results Most of the findings are based on 822 usable interviews conducted near doorways to Atlantic Station establishments during two morning and two afternoon peak periods. Of these approximately 27% were obtained during the A.M. peak and 73% during the P.M. peak. Approximately 45% of exiting people were interviewed in the A.M. peak period while about 15% were interviewed in the much more active P.M. peak. The results described in this report are based on expansion factors applied to usable interviews based on sampling rates for each land use and time period as well as the businesses surveyed. The expansion factor process is explained elsewhere in this report. Table 49 shows the numbers of completed and usable inter- views by peak period and land use category. Interviews were completed for about 30% of people exiting at street level (and the grocery store garage) during the A.M. peak and 15% during the P.M. peak. Interviews were not attempted at building entrances within the garage because it was felt that nearly all people using those few entrances would be either leaving Atlantic Station or returning to street level through the same buildings. In the first case, the trips would be external and eas- ily categorized as such. In the second case, the people would be candidates for interviews as they departed from the building at street level. In addition, a few retail buildings were not covered with interviews due to the available survey crew and simi- larity of tenants to those being surveyed. Survey results were expanded to cover unsurveyed buildings in accordance with development units. Finally, not all residential buildings were surveyed. Local management staff indicated that profiles of the residents were similar to those in buildings being surveyed. Cor- don counts were used to factor the residential survey results. As with the Mockingbird Station interviews, some Atlantic Station interviews yielded more than one trip. All exiting trips were obtained. Some people interviewed also provided com- plete and usable information about their inbound trips to the interview location. Table 50 shows the total numbers of 45 A.M. Peak Period P.M. Peak Period Land Use Interviews1 Building Exit Movements Percent Interviewed Interviews 1 Building Exit Movements Percent Interviewed Office 15 93 16% 15 84 18% Retail 29 153 19% 266 2,138 12% Restaurant 24 29 83% 184 918 20% Residential 157 523 30% 66 305 22% Cinema2 — — — 31 282 11% Hotel 21 36 58% 14 95 15% Total 246 834 30% 576 3,822 15% 1 Number of interviews conducted with travelers exiting doors of a particular land use that contained at least one usable trip. 2 Cinema not open during morning peak period. Table 49. Peak-period interviews, counted building exit movements, and percent interviewed—Atlantic Station.

usable interviews available based on their points of origin. Interviews during the P.M. peak period yielded some second usable trips; none were derived from the A.M. interviews. The results reported herein are based on factoring to reflect sampling at each building entrance; factoring was performed separately for each peak period. That process is described else- where in this report. The survey results were summarized for the A.M. and P.M. peak periods. Table 51 shows a summary of completed interviews, exit- ing people, and usable trips derived from the interviews. The total of the reported trips, if made during one of the two peak periods, is shown as usable trips in Table 50. Some of the reported inbound trips occurred outside the peak periods; however, for many of those trips, the respondent was unable or unwilling to provide enough complete information to make the inbound trip usable. Finally, some otherwise com- plete interviews were not usable because the inbound trip reported was not actually the immediately previous trip. Many of those trips were made outside the peak periods. First trips of the day from the onsite apartments did not have a previous trip that day. The interviews reported in Table 51 differ slightly from inter- views reported in prior tables. This is because the interviews reported in Table 50 are associated with the land use for which an exit trip is reported whereas previous tables reported by where the interview occurred. Table 51 points out the same lesson learned from the sur- vey procedures as did the Mockingbird Station surveys: to 46 A.M. Peak Period P.M. Peak Period Land Use Usable Trips1 Building Exit Movements Percent Usable Usable Trips 1 Counted Exit Movements Percent Usable Office 13 93 14% 26 84 31% Retail 29 153 19% 313 2,138 15% Restaurant 26 29 90% 189 918 21% Residential 141 523 27% 56 305 18% Cinema2 — — — 38 282 14% Hotel 37 36 103%a 43 95 45% Total 246 834 30% 665 3,822 17% 1 Must include specific origin location, location of destination, and land use of destination if internal; this total includes reported exiting and entering trips made this period. 2 Cinema not open during morning peak period. a More trips reported from both exit and entering trip responses than movements counted. Table 50. Peak-period usable trips, counted building exit movements, and percent usable—Atlantic Station. Morning Peak Period Afternoon Peak Period Land Use Interviews2 Exit Movements Percent Interviewed Usable Trips3 Interviews 2 Exit Movements Percent Interviewed Usable Trips3 Office 13 93 14% 13 15 84 18% 15 Retail 29 153 19% 30 243 2,138 11% 311 Restaurant 26 29 90% 28 167 918 18% 212 Residential 141 229a 62% 141 79 115b 69% 79 Cinema1 — — — — 32 282 11% 32 Hotel 34 36 94% 34 16 95 17% 16 Total 243 540 45% 246c 552 3,632 15% 665 1 No interviews attempted at cinema during the morning peak period since cinema closed during this period. 2 Number of interviews conducted with travelers exiting doors of a particular land use that contained at least one usable trip. 3 Must include specific origin location, location of destination, and land use of destination if internal; this total includes reported exiting and entering trips made this period. a Excludes 294 movements counted at two residential developments where no interviews were completed. b Excludes 190 movements counted at two residential developments where no interviews were completed. c Includes 3 movements counted at establishments where no or too few interviews were completed for valid sample. Table 51. Peak-period interviews, exit movements, percent interviewed, and usable trips—Atlantic Station.

obtain accurate inbound trip information while conducting exit surveys, it is necessary to increase the amount of interview practice for each interviewer (mock interviews with trainers). Interviewers recorded too many incomplete interviews and incorrect previous trips. Most interviewers also failed to ask or record responses for all of the questions about the inbound trip, resulting in more incomplete inbound trip information. Table 51 shows that, unlike Mockingbird Station, A.M. inter- views were completed at retail outlets. These were almost all at the onsite grocery store. Table 52 shows, for the morning peak period, surveyed trips (usable from interviews) by origin land use as well as the number of people exiting doors for each land use. The third column represents exit movements from establishments where no completed interviews occurred. These trips were estimated based on square footage for the specific land use. The fourth column contains the number of drivers plus pas- sengers who exited Atlantic Station from parking garages after reaching the garages via direct internal access from establish- ments above. Trips in this fourth column were assumed to all be external since they involved trips downstairs into the below ground garages and a drive along the perimeter or beyond to another location. In almost all cases, a walk trip would take less time except between the District and the apartment com- plexes to the west or townhouses to the south. About 46% of all trips made from survey locations were represented by an interview or direct external trips. Table 53 displays similar information for the P.M. peak period. For this period, about 31% of the total trips are rep- resented by interviews or direct external trips. Unsurveyed locations, which were judged by the research team to have char- acteristics similar to other establishments of the same land uses, represent less than 15% of the exiting trips made. The direct garage trips to the external street system accounted for about 20% of the total trips, a little less than for the A.M. peak period. Table 54 contains somewhat similar information for the entering trips for both peak periods. As with the similar table for Mockingbird Station, this table shows the several sources for information on trips made. Trips represented by exit surveys are shown in the first column of numbers. Trips made direct from internally accessed parking garages to external locations are shown in the second column. Trips made using the free Atlantic Station shuttle and represented by interviews conducted on the shuttle are shown in the third column of numbers. The balance column represents the difference between the total number of counted (or prorated by sq ft) persons entering the establish- ments (through public doorway or from external points to the garage sections with private internal access) and the trips repre- sented in the prior three columns. All trips in the fourth column of numbers were assumed to be external since they had no reported internal source for trips. Table 55 shows the mode split of person trips to and from Atlantic Station during the A.M. peak period and peak hour. Personal vehicles account for about 80% of the inbound trips and slightly more than 70% of the outbound trips during both periods, indicating little difference between the periods. Similarly, peak-period and peak-hour mode shares differed little for other modes. Transit, including the free Atlantic Station shuttle, accounted for 8% of the inbound A.M. peak period trips and 9% of the inbound A.M. peak-hour trips. Outbound percentages were smaller. Virtually all transit trips used the free shuttle; MARTA bus service attracted almost no 47 Exit Movements Land Use Surveyed Trips2 Doors Unsurveyed Locations3 Garage Direct4 Total Percent Surveyed5 Office 13 93 — 66 159 50% Retail 29 153 8 136 297 56% Restaurant 26 29 — — 29 90% Residential 141 523 — 68 591 35% Cinema1 — — — — — — Hotel 37 36 — — 36 100% Total 246 834 8 270 1,112 46% 1 Cinema did not actively generate trips during the morning peak period. 2 Number of usable trip origins at each land use recorded from traveler interviews. 3 Includes locations where no interviews were attempted (prorated by sq ft) and locations where door counts were made but no usable trip origins were recorded on interviews. external location. 4 Person-trips observed exiting onsite parking garages, assumed to be traveling directly to an the external street system. 5 Includes those trips described in usable interviews or direct exits from a parking garage to Table 52. Morning peak-period surveyed trips, exit movements, and percent surveyed—Atlantic Station.

48 Exit Movements Land Use Surveyed Trips2 Doors Unsurveyed Locations3 Garage Direct4 Total Percent Surveyed5 Office 26 84 — 585 669 91% Retail 313 2,138 532 418 3,088 24% Restaurant 189 918 115 — 1,033 18% Residential 56 305 — 50 355 30% Cinema1 38 282 — — 282 13% Hotel 43 95 — — 95 45% Total 665 3,822 647 1,053 5,522 31% 1 Cinema did not actively generate trips during the morning peak period. 2 Number of usable trip origins at each land use recorded from traveler interviews. 3 Includes locations where no interviews were attempted (prorated by sq ft) and locations where door counts were made but no usable trip origins were recorded on interviews. external location. 4 Person-trips observed exiting onsite parking garages, assumed to be traveling directly to an external street system. 5 Includes those trips described in usable interviews or direct exits from a parking garage to the Table 53. Afternoon peak-period surveyed trips, exit movements, and percent surveyed—Atlantic Station. Morning Peak Period Afternoon Peak Period Land Use Survey2 Garage Direct3 Transit Direct4 Balance 5 Total Survey2 GarageDirect3 Transit Direct4 Balance 5 Total Office 86 829 41 41 997 55 69 17 (17)a 124 Retail 114 35 24 17 190 1,769 411 66 406 2,652 Restaurant 26 — 31 (23)a 34 542 — 48 694 1,284 Residential 0 8 6 186 200 313 90 14 131 548 Cinema1 — — — — — 165 — 39 111 315 Hotel 1 — 7 17 25 88 — 7 0 95 Total 227 872 109 238 1,446 2,932 570 191 1,325 5,018 1 Cinema did not actively generate trips during the morning peak period. 2 Trip destinations recorded from exit interviews, expanded as described. 3 Person-trips observed entering onsite parking garages, assumed to be traveling directly from an external location. 4 Trips entering onsite land uses from external locations recorded on transit interviews. 5 Balance of person-trips entering onsite land uses; assumed to originate externally. a See Appendix C for more information. Table 54. Peak-period person-trips entering land uses—Atlantic Station. Peak Period (7:00 A.M.–10:00 A.M.) Peak Hour (8:00 A.M.–9:00 A.M.) Trips Percent Trips Percent Travel Mode Entering Exiting Entering Exiting Entering Exiting Entering Exiting Personal Vehicle1 2,378 1,165 79% 71% 972 447 81% 72% Motorcycle 0 0 0% 0% 0 0 0% 0% Delivery Vehicle 172 216 6% 13% 46 70 4% 11% Transit2 244 56 8% 3% 104 34 9% 5% Walk 226 195 7% 12% 78 68 6% 11% Bicycle 6 4 <1% <1% 2 1 <1% <1% Total All Modes 3,026 1,636 100% 100% 1,202 620 100% 100% 1 Personal vehicle occupancies (entering/exiting): peak period 1.08/1.13; peak hour 1.05/1.12. 2 Transit trips include circulating shuttle and bus. Table 55. A.M. peak-period and peak-hour person-trip cordon count— Atlantic Station.

49 Peak Period (4:00 P.M.–7:00 P.M.) Peak Hour (5:00 P.M.–6:00 P.M.) Trips Percent Trips Percent Travel Mode Entering Exiting Entering Exiting Entering Exiting Entering Exiting Personal Vehicle1 3,727 3,423 87% 88% 1,382 1,242 89% 86% Motorcycle 0 0 0% 0% 0 0 0% 0% Delivery Vehicle 50 43 1% 1% 14 18 1% 1% Transit2 195 243 5% 6% 72 103 5% 7% Walk 300 184 7% 5% 86 72 6% 5% Bicycle 15 4 <1% <1% 4 2 <1% <1% Total All Modes 4,287 3,897 100% 100% 1,558 1,437 100% 100% 1 Personal vehicle occupancies (entering/exiting): peak period 1.40/1.27; peak hour 1.37/1.22. 2 Transit trips include circulating shuttle and bus. Table 56. P.M. peak-period and peak-hour person-trip cordon count— Atlantic Station. A.M. Peak Period P.M. Peak Period Entering Exiting Entering Exiting Land Use Trips PercentInternal Trips Percent Internal Trips Percent Internal Trips Percent Internal Office 990 8% 152 33% 124 45% 668 9% Retail1 135 44% 136 42% 1,431 38% 1,867 39% Restaurant 34 77% 29 48% 1,218 39% 967 60% Residential 200 0% 591 2% 543 57% 350 13% Cinema2 — — — — 315 52% 281 42% Hotel 25 4% 36 95% 95 92% 94 86% Total All Trips 1,384 12% 944 17% 3,726 44% 4,227 38% 1 Retail open during A.M. peak period was primarily grocery store. 2 Cinema not open to customers during morning peak period. Table 57. Peak-period person-trips and percent internal trip capture by land use—Atlantic Station. Atlantic Station trips, possibly due to the fare difference, more frequent shuttle service, and the shuttle loop throughout the District, which the MARTA service does not provide. Table 56 shows similar summaries for the P.M. peak period and peak hour. Personal vehicles account for more of the travel during the P.M. peaks, accounting for 87–88% of all peak period trips and 86–89% of peak-hour trips. As for the A.M. peaks, there is little difference in mode splits between the peak period and peak hour. Deliveries account for far fewer trips during the P.M.. The transit mode splits are simi- lar in total but more balanced between inbound and out- bound trips. Inbound walk trips are similar for both A.M. and P.M. peaks, but outbound walk trips make up a larger percent- age of A.M. peak trips than for the P.M.. Note, however, that the A.M. inbound and outbound walk volumes are fairly similar. From these two tables it is apparent that motor vehicles are the primary mode of travel for Atlantic Station, but transit and walking also play a role. As was found for Mockingbird Station, the P.M. peak-period vehicle occupancies are signifi- cantly higher than those for the A.M. peak; this is attributable to people going shopping or to restaurants or the cinema in groups of two or more during the P.M. peak. Almost none of those businesses are open during the A.M. peak. Table 57 shows the total A.M. and P.M. peak-period entering and exiting trips by land use category plus the percentage of those that were internal. Overall A.M. peak-period internal trip capture was about 12% for inbound and 17% outbound. For the P.M. peak period, the inbound and outbound internal cap- ture percentages total about 44 and 38 percent, respectively. As the table shows, internal capture varies significantly by land use as it did for Mockingbird Station. The A.M. retail percentage reflects activity at the full-service grocery store, which serves both residential and office patrons during that period. The open restaurant was a coffee shop, similar to Mockingbird

Station. The hotel appeared to be heavily oriented to serving the onsite office building, which was the only nearby office build- ing although others exist about 1⁄2 mile away. For the P.M. peak period, the internal percentage of entering trips was consistent across most uses, other than the hotel, which again appeared to be very internally oriented. There was more variation in trips exiting Atlantic Station buildings during the P.M. peak period, ranging between 9 and 86%. The office low percentage is reflec- tive of commuters going home, or at least off-site, after work. The low residential percentage is a little surprising, but the longer distance to other uses may be influential. Table 58 shows the percent distribution of trips from each origin land use to other land uses within Atlantic Station as well as to external destinations. As was explained for Mockingbird Station, trips between similar land uses are not included because they would not be counted as external trips for single-use developments. Some land use pairs have little interchange; others have extensive interaction, as described in the preceding paragraph. Table 59 shows similar information for the P.M. peak period. A few land use pairs account for most of the internal trips, sim- ilar to Mockingbird Station; however, the specific pairs are not the same. This is a result of the residential units being farther removed from the non-residential uses and different balances of land uses. During the P.M. peak, the heaviest percentages of interaction are retail-restaurant (both directions) and from retail to restaurant and residential, cinema to retail, and from hotel to retail and restaurant. Table 60 shows the A.M. peak period percentage distribu- tion of inbound trips to each destination land use from each origin land use. The highest inbound internal capture per- centages are retail from office and restaurant from office and retail. Many of those trips may result from trips for morning coffee or picking up a lunch on the way to work. Table 61 shows similar information for the P.M. peak period. The largest internal capture percentages are to office, restaurant, cinema, and hotel from retail and to restaurant from retail, cin- ema, and hotel. These two tables demonstrate that there is a 50 Destination Land Use Summary Origin Land Use Office Retail2 Restaurant Residential Cinema3 Hotel Internal External Total Total Trips Office —1 28 5 0 — 0 33 67 100 152 Retail2 29 —1 13 0 — 0 42 58 100 136 Restaurant 31 14 —1 0 — 3 48 52 100 29 Residential 1 1 0 —1 — 0 2 98 100 591 Cinema3 — — — — —1 — — — — — Hotel 75 14 6 0 — —1 95 5 100 36 All Origins 8 6 3 0 — 0 17 83 100 944 1 Internal trips within a land use are not included in internal trip capture methodology. 2 Retail open during A.M. peak period was primarily grocery store. 3 Cinema not open to customer during morning peak period. Table 58. Percent distribution of internal trip destinations for exiting trips— Atlantic Station, A.M. peak period. Destination Land Use Summary Origin Land Use Office Retail Restaurant Residential Cinema Hotel Internal External Total TotalTrips Office —1 6 3 0 0 0 9 91 100 668 Retail 2 —1 19 13 4 1 39 61 100 1,867 Restaurant 1 41 —1 3 8 7 60 40 100 967 Residential 0 9 3 —1 0 1 13 87 100 350 Cinema 2 21 11 8 —1 0 42 58 100 281 Hotel 0 16 68 2 0 —1 86 14 100 94 All Origins 1 13 11 7 4 2 38 62 100 4,227 1 Internal trips within a land use are not included in internal trip capture methodology. Table 59. Percent distribution of internal trip destinations for exiting trips— Atlantic Station, P.M. peak period.

different internal capture rate by direction, similar to that for Mockingbird Station. This should be no surprise as some of the peak period activities are very directional (e.g., commuting to work in A.M., from work in P.M.). Table 62 shows the internal person trips and percentages for both peak periods by mode of travel for the reported trip. Dur- ing the A.M. peak, there are a few internal vehicle driver and vehicle passenger trips. All walk and bicycle trips are internal. However, no taxi or transit trips are internal. Unlike Mocking- bird Station, Atlantic Station is larger and encourages use of personal vehicles for some trips; the maximum internal trip length is about 0.6 miles. In the P.M. peak period, the internal capture percentages are significant for trips by all modes. Some combination of after-work shopping, dinner, or cinema may be the cause of higher vehicle use for internal trips. Visitors who are less familiar with specific locations or distances may view the Atlantic Station shuttle as more convenient for internal trips. Table 63 shows similar information for Atlantic Station, but by original mode of access. The mode shown is that used for the first trip to Atlantic Station, not for the trip being reported; therefore, if a person arrived early in the morning driving a vehicle but is interviewed during a walk trip, the mode of access is vehicle driver. Hence, for those who originally entered Atlantic Station by driving a personal vehicle, during the A.M. peak period 6% of the inbound trips were from internal origins and 26% of the outbound trips from Atlantic Station build- ings were to internal destinations. During the A.M. peak period, there were few enough non-vehicle driver trips that the differ- ences in internal trip percentages may be nearly meaningless. However, during the P.M. peak period, those with a personal vehicle as their access mode have a lower percentage of inter- nal trips than most other modes. People who originally arrived by walk/bike and circulator modes tend to have higher inter- nal trip percentages. Table 64 reports internal capture percentages by whether or not the trip-maker had access to a vehicle for the trip. The 51 Origin Land Use Summary Destination Land Use Office Retail2 Restaurant Residential Cinema3 Hotel Internal External Total Total Trips Office —1 4 1 0 — 3 8 92 100 990 Retail2 32 —1 3 5 — 4 44 56 100 135 Restaurant 21 50 —1 0 — 6 77 23 100 34 Residential 0 0 0 —1 — 0 0 100 100 200 Cinema3 — — — — —1 — — — — — Hotel 0 0 4 0 — —1 4 96 100 25 All Destinations 4 4 1 1 — 2 12 88 100 1,384 1 Internal trips within a land use are not included in internal trip capture methodology. 2 Retail open during A.M. peak period was primarily grocery store. 3 Cinema not open to customer during morning peak period. Table 60. Percent distribution of internal trip origins for entering trips— Atlantic Station, A.M. peak period. Origin Land Use Summary Destination Land Use Office Retail Restaurant Residential Cinema Hotel Internal External Total TotalTrips Office —1 31 8 0 6 0 45 55 100 124 Retail 3 —1 28 2 4 1 38 62 100 1,431 Restaurant 2 29 —1 1 2 5 39 61 100 1,218 Residential 1 46 6 —1 4 0 57 43 100 543 Cinema 1 26 25 0 —1 0 52 48 100 315 Hotel 0 17 71 5 0 —1 92 8 100 95 All Destinations 2 20 16 1 3 2 44 56 100 3,726 1 Internal trips within a land use are not included in internal trip capture methodology. Table 61. Percent distribution of internal trip origins for entering trips— Atlantic Station, P.M. peak period.

52 A.M. Peak Period1 P.M. Peak Period2 Entering Exiting Entering Exiting Mode of Travel Trips PercentInternal Trips Percent Internal Trips Percent Internal Trips Percent Internal Vehicle Driver 1,110 4% 761 6% 2,125 18% 2,336 17% Vehicle Passenger 72 10% 26 27% 215 27% 358 16% Taxi/Car Service 0 0% 0 0% 18a 100% 43 42% Transit (Bus) 0 0% 17 0% 15a 100% 101 15% Transit (Circulating Shuttle) 86 0% 9 0% 325 41% 406 62% Walk/Bicycle 116 100% 129 90% 1,026 100% 1,158 89% 1 Travel mode not reported for 2 exiting trips. 2 Travel mode not reported for 2 entering trips and 25 exiting trips. a Limited sample; possible erroneous response. Table 62. Peak period person-trips and percent internal trip capture by mode of travel—Atlantic Station. A.M. Peak Period1 P.M. Peak Period2 Entering Exiting Entering Exiting Mode of Access Trips PercentInternal Trips Percent Internal Trips Percent Internal Trips Percent Internal Vehicle Driver 1,141 6% 283 26% 2,552 32% 2,645 31% Vehicle Passenger 70 7% 31 16% 277 43% 409 29% Taxi/Car Service 1 100% 1 100% 22 100% 22 100% Transit (Bus) 36 100% 56 64% 40 100% 152 26% Transit (Circulating Shuttle) 89 3% 4 75% 468 59% 331 84% Walk/Bicycle 11 100% 18 61% 86 100% 129 68% 1 Access mode not reported for 36 entering trips and 551 exiting trips. 2 Access mode not reported for 281 entering trips and 539 exiting trips. Table 63. Peak-period person-trips and percent internal trip capture by mode of access—Atlantic Station. A.M. Peak Period1 P.M. Peak Period2 Entering Exiting Entering Exiting Automobile Access Trips PercentInternal Trips Percent Internal Trips Percent Internal Trips Percent Internal Yes (Vehicle Driver) 1,206 6% 302 24% 2,710 30% 2,857 29% Yes (Non-Vehicle Driver) 58 24% 14 100% 71 49% 78 45% No Vehicle Access 54 13% 20 35% 169 21% 107 34% 1 Automobile access not reported for 66 entering trips and 608 exiting trips. 2 Automobile access not reported for 776 entering trips and 1,185 exiting trips. Table 64. Peak period person-trips and percent internal trip capture by vehicle access—Atlantic Station.

contain retail, restaurant, office, and/or residential space). Residential development consists of owner-occupied town- homes and rental apartments. There is one major full-service hotel and one five-screen cinema that shows mainly artistic movies. Table 65 contains the development program. Addi- tional phases of Legacy Town Center are being developed to the north of Legacy Drive. Figure 8 illustrates the site plan of Legacy Town Center. The site is about 1,600 ft by 2,000 ft. The site is well connected by streets, with block lengths ranging between 300 and 600 ft. Each land use tends to be concentrated in a section of Legacy Town Center—for example, the apartments are in the southeastern 53 Land Use Occupied Development Units Residential 1,300 apartments; 60 townhomes Retail 196,264 sq ft Office 310,764 sq ft Restaurant 69,318 sq ft Cinema 27,125 sq ft; 5 screens; 1,019 seats Parking 6,070 parking spaces Table 65. Legacy Town Center development components. Source: The Shops at Legacy L.P. Dallas North Tollway Tennyson Parkway Legacy Drive Parkwood Boulevard Figure 8. Legacy Town Center illustrative site plan. limited response seems to indicate that vehicle availability does not consistently influence the amount of internal trip making at Atlantic Station. Since most is conveniently walk- able, the personal vehicles are not crucial to move around within Atlantic Station. Legacy Town Center Development Characteristics The third MXD surveyed was Legacy Town Center in Plano, Texas, which is a northern suburb in the Dallas–Ft. Worth area. Plano is located about 20 miles north of downtown Dallas. Development is almost solid to the south and east. Much of the area north and west of Plano is in active development. Phase 1 of Legacy Town Center is completely developed. It is a multiple-block, single-development site bounded on all four sides by major roadways. The site is just over 70 acres. It consists of office, retail, restaurant, cinema, hotel, and residen- tial land uses. Most retail and restaurant buildings are single story. Tenants tend to be specialty retail and restaurants rang- ing from better fast food to very exclusive. Most office space is in six-to-eight story buildings although a small amount is in second and third stories of two mixed-use buildings (which

region although, again, few are thought by management to live within Legacy Town Center. Access to Legacy Town Center is provided by the Dallas North Tollway (which extends to downtown Dallas) and State Highway 121 (which extends across the region from northeast to central Ft. Worth), which is less than 1 mile north of Legacy Town Center. Legacy Drive is a regional east–west arterial. Tennyson Parkway and Parkwood Boulevard are minor arte- rials about 5 miles long. Transit service to Legacy Town Center is limited: there is one DART bus route (452) that serves Legacy Town Center in a directional loop along both its north and south boundaries. Service operates between 6 A.M. and 11:30 P.M. Headways are 30 minutes during peak periods and 60 minutes during off- peak periods. There are multiple stops on both Legacy Drive and Tennyson Parkway. No use of this route by travelers to and from Legacy Town Center was observed either during survey periods or occasional additional observations. The Marriott Hotel offers its patrons free shuttle service to and from desti- nations in the area. Small shuttle vehicles provide this service. Hotel patrons do use this service. Travel Survey Travel surveys virtually identical to those conducted at Mockingbird Station and Atlantic Station were conducted at Legacy Town Center. Surveys were conducted Tuesday after- noon through Thursday morning, May 22–24, 2007. Due to rainy weather, surveys were also conducted on the following Thursday afternoon and Tuesday morning. One addition was employed for this survey: inbound office building interviews were conducted during the morning peak period. Survey Results Table 66 shows the number of exit movements and inter- views during the 6:30–10 A.M. morning and 3:00–7:00 P.M. 54 Source: Texas Transportation Institute Figure 9. Legacy Town Center retail, restaurant, and cinema buildings along main retail street. Source: Texas Transportation Institute Figure 10. Pedestrian-friendly environment in main retail and restaurant area. section, retail and restaurants are mixed but in the north por- tion, and townhomes are in a small portion of the east side. However, the connectivity provides a high degree of linkages between all blocks and land uses. Almost all blocks have land- scaped walkways making them attractive to walk along and often are shaded. Figures 9 through 11 illustrate the character of Legacy Town Center. Legacy Town Center is actually part of a large business park development, which could evolve upon buildout into a major suburban activity center. Much of the area is occupied by free-standing corporate headquarters buildings, but there are many other types of commercial and residential development close by including hotels, regional shopping centers, and residential complexes. Observed activity demonstrated that Legacy Town Center has become a center of activity in the area, especially after work and in the evenings. Several restaurants were very busy, even during the week. The developer reported that most Legacy Town Center residents work within 5 miles of Legacy Town Center although a few work in Legacy Town Center. Most Legacy Town Center employees also live in the north Dallas Source: Texas Transportation Institute Figure 11. Apartment building on south side of central park with office building in background.

buildings for which no interviews were conducted). In all, usable trips constituted 18% of the morning people exiting surveyed buildings and 14% in the afternoon. The minimum sample rate for the morning was 12% and afternoon was 8%. Table 68 shows a summary of completed interviews, exiting people, and usable trips derived from the interviews. Table 67 shows the total of the reported trips as usable trips, if they were made during one of the two peak periods. Unusable trips included inbound trips that occurred outside the peak periods or trips for which the respondent was unable or unwilling to provide enough complete information to make the inbound trip usable. Finally, some otherwise complete interviews were not usable because the inbound trip reported was not actually the immediately previous trip; many of those trips were made outside the peak periods. First trips of the day from the onsite apartments did not have a previous trip that day. The interviews reported in Table 68 differ slightly from interviews reported in prior tables because the interviews 55 Morning Peak Period Afternoon Peak Period Land Use Interviews1 Exit Movements Percent Interviewed Interviews 1 Exit Movements Percent Interviewed Office 9 77 12% 80 362 22% Retail 24 91 26% 59 595 12% Restaurant 99 453 22% 74 913 11% Residential 146 628 23% 80 592 19% Cinema2 — — — 48 108 49% Hotel 49 181 27% 50 299 17% Total 327 1,430 23% 391 2,869 16% 1 Number of interviews conducted with travelers exiting doors of a particular land use that contained at least one usable trip. 2 Cinema not open during morning peak period. Table 66. Peak-period interviews, exit movements, and percent interviewed—Legacy Town Center. Morning Peak Period Afternoon Peak Period Land Use Usable Trips1 Exit Movements Percent Usable Usable Trips1 Exit Movements Percent Usable Office 9 73 12% 74 362 20% Retail 25 108 23% 62 595 10% Restaurant 100 551 18% 77 913 8% Residential 148 710 21% 96 592 16% Cinema2 — — — 49 108 45% Hotel 54 400 14% 50 299 17% Total 336 1,842 18% 408 2,869 14% 1 Must include specific origin location, location of destination, and land use of destination if internal; the total includes total reported outbound and inbound trips made this period. 2 Cinema not open during morning peak period. Table 67. Peak-period usable trips, exit movements, and percent usable—Legacy Town Center. afternoon peak periods. Due to the size of Legacy Town Cen- ter and number of buildings and businesses, interviews could not be conducted at all buildings and businesses. Samples of each land use were selected based on discussions with the developer and property managers—for example, two of the three office buildings were surveyed and six of the eight major block faces of retail were surveyed. One furniture store declined to be included in the survey. Overall, 23% of people exiting surveyed buildings were interviewed during the A.M. peak and 16% for the P.M. peak. No land use interview rate was less than 10%. The cinema was closed during the A.M. peak as were almost all non-convenience retail businesses and most restaurants. The only restaurants open were a specialty coffee shop and a bakery/coffee/light breakfast restaurant. Table 67 shows the number of usable trips derived from the interviews. These are compared with the counted exit movements during the interview periods (including some

reported in Table 66 are associated with the land use for which an exit trip is reported, whereas previous tables reported where the interview occurred. A.M. interviews were completed at the open retail outlets—in this case, a convenience retail store and a dry cleaner—throughout the morning peak period. A United Parcel Service (UPS) store opened at 9 A.M. Table 69 shows the number of trips exiting Legacy Town Center establishments that had exit trips described in inter- views. This table also shows the number of persons counted exiting at locations where interviews were conducted plus the prorated number estimated to have exited at locations where counts and interviews were not conducted. These locations were judged by the researchers to be represented by similar establishments that were surveyed except one case in which the proprietor declined to permit any interviewing. That location was included in the proration by square footage. Legacy Town Center had no direct internal access to parking garages where the interviewers could not intercept exiting people. Hence, unlike Mockingbird Station and Atlantic Station, Legacy Town Center trip characteristics are based entirely on the exit inter- views. Legacy Town Center also has far more separate estab- lishments than either of the other two developments surveyed 56 Morning Peak Period Afternoon Peak Period Land Use Interviews2 Exit Movements Percent Interviewed Usable Trips3 Interviews 2 Exit Movements Percent Interviewed Usable Trips3 Office 9 73a 12% 9 80 312b 26% 74 Retail 24 108 22% 25 59 536c 11% 62 Restaurant 99 551 18% 100 74 913 8% 77 Residential 146 710 21% 148 80 592 14% 96 Cinema1 — — — — 48 108 44% 49 Hotel 49 400 12% 54 50 299 17% 50 Total 327 1,842 18% 336 391 2,760 14% 408 1 No interviews attempted at cinema during the morning peak period since cinema was closed. 2 Number of interviews conducted with travelers exiting doors of a particular land use that contained at least one usable trip. 3 Must include specific origin location, location of destination, and land use of destination if internal; the total includes total reported outbound and inbound trips made this period. a Excludes 4 movements counted at establishments where no or too few interviews were completed for valid sample. b Excludes 50 movements counted at establishments where no or too few interviews were completed for valid sample. c Excludes 59 movements counted at establishments where no or too few interviews were completed for valid sample. Table 68. Peak-period interviews, exit movements, percent interviewed, and usable trips—Legacy Town Center. Exit Movements Land Use Surveyed Trips2 Doors Unsurveyed Locations3 Garage Direct4 Total Percent Surveyed5 Office 9 77 21 — 98 9% Retail 25 108 — — 108 21% Restaurant 100 551 — — 551 18% Residential 148 710 953 — 1,663 9% Cinema1 — — — — — — Hotel 54 400 — — 400 14% Total 336 1,846 974 — 2,820 12% 1 Cinema did not actively generate trips during the morning peak period. 2 Number of usable trip origins at each land use recorded from traveler interviews. 3 Includes locations where no interviews were attempted (prorated by sq ft) and locations where door counts were made but no usable trip origins were recorded on interviews. 4 Person-trips observed exiting onsite parking garages, assumed to be traveling directly to an external location. 5 Includes those trips described in usable interviews or direct exits from a parking garage to the external street system. Table 69. Morning peak-period surveyed trips, exit movements, and percent surveyed—Legacy Town Center.

in the pilot surveys, so a smaller portion of the establishments could be surveyed with the available resources. Approximately 2⁄3 of all trips were covered by direct interview sampling; the other 1⁄3 was included by proration. For the morning peak period, approximately 12% of all exiting trips are represented by surveyed trips. Table 70 displays similar information for the P.M. peak period. There were about 60% more exiting trips in the P.M. peak period than during the A.M. peak period. The interviewed trips represent a sample of approximately 9% of all exiting trips. Table 71 shows for each peak period the sources of enter- ing trip information for trips entering Legacy Town Center establishments. For the morning peak period, interviews rep- resent about 18% of the trips and the balance was assumed to all be external. For the evening peak period, about 38% of the entering trips are represented by interviews and the remain- ing 62% considered all external. Table 72 shows results from the morning cordon count. As might be expected for a suburban development with lim- ited transit service, almost all trips to and from Legacy Town Center were by motor vehicle. Transit, shuttle, walking, and bicycling combined accounted for about 4 to 6% of the A.M. and P.M. peak person trips. Almost none were by bicycle or public transit although some were by hotel shuttle van. Per- sonal vehicle occupancy rates were about 1.07 inbound and 1.12 outbound for the A.M. peak period. 57 Exit Movements Land Use Surveyed Trips2 Doors Unsurveyed Locations3 Garage Direct4 Total Percent Surveyed5 Office 74 362 155 — 517 14% Retail 62 595 266 — 861 7% Restaurant 77 913 491 — 1,404 5% Residential 96 592 794 — 1,386 7% Cinema1 49 108 — — 108 45% Hotel 50 299 — — 299 17% Total 408 2,869 1,706 — 4,575 9% 1 Cinema did not actively generate trips during the morning peak period. 2 Number of usable trip origins at each land use recorded from traveler interviews. 3 Includes locations where no interviews were attempted (prorated by sq ft) and locations where door counts were made but no usable trip origins were recorded on interviews. 4 Person-trips observed exiting onsite parking garages, assumed to be traveling directly to an external location. 5 Includes those trips described in usable interviews or direct exits from a parking garage to the external street system. Table 70. Afternoon peak-period surveyed trips, exit movements, and percent surveyed—Legacy Town Center. Morning Peak Period Afternoon Peak Period Land Use Survey2 Garage Direct3 Transit Direct4 Balance 5 Total Survey2 GarageDirect3 Transit Direct4 Balance 5 Total Office 121 — — 476 597 89 — — 7 96 Retail 30 — — 89 119 316 — — 507 823 Restaurant 156 — — 437 593 787 — — 1,217 2,004 Residential 79 — — 593 672 592 — — 924 1,516 Cinema1 — — — — — 71 — — 150 221 Hotel 6 — — 181 187 115 — — 200 315 Total 392 — — 1,776 2,168 1,970 — — 3,005 4,975 1 Cinema did not actively generate trips during the morning peak period. 2 Trip destinations recorded from exit interviews, expanded as described. 3 Person-trips observed entering onsite parking garages, assumed to be traveling directly from an external location. 4 Trips entering onsite land uses from external locations recorded on transit interviews. 5 Balance of person-trips entering onsite land uses; assumed to originate externally. Table 71. Peak-period person-trips entering land uses—Legacy Town Center.

Table 73 shows similar data for the P.M. peak period. As with the A.M. peak period, the P.M. shows that nearly all trips to and from Legacy Town Center are by personal vehicle. Transit, bicycle, and walk modes in total compose a slightly lower per- centage of trips in the P.M. than the A.M.. Vehicle occupancies were significantly higher during the P.M. peak period, possibly due to the increased percentages of trips to and from retail, restaurants, and entertainment businesses that are open dur- ing the P.M. peak period but not during the A.M. peak period. Table 74 shows the A.M. and P.M. peak-period internal trip capture percentages as reported in the interviews. The A.M. peak-period internal capture was about 15% for entering trips and 11% for exiting trips. During the P.M. peak period, the internal capture percentages were higher at 33% for entering trips and 37% for exiting trips. These summaries include only trips between different land uses; trips between the same land use are not included to remain consistent with the trip gen- eration methodology used by the ITE. As might be expected, the highest A.M. internal capture rates are for retail (largely convenience). Office, residential, and hotel generated the lowest percentages of internal trips. Residential trips to internal destinations were primarily to convenience retail or the coffee shop. During the P.M. peak period, interaction between retail, restaurant, cinema, and hotel was demonstrated. Many onsite residents also traveled to these destinations. P.M. internal trip capture percentages were consistent for most land uses with between 30% and 43%. Although there were few trips destined for the office buildings, a high percentage were from internal origins; how- ever, very few of those leaving office space at Legacy Town Center made trips to other onsite destinations. Trips leaving retail also had a high percentage of internal capture, with 58 Peak Period (7:00 A.M.–10:00 A.M.) Peak Hour (7:30 A.M.–8:30 A.M.) Trips Percent Trips Percent Travel Mode Entering Exiting Entering Exiting Entering Exiting Entering Exiting Personal Vehicle1 1,767 1,745 93% 91% 770 725 94% 93% Motorcycle 0 1 0% <1% 0 0 0% 0% Delivery Vehicle 77 66 4% 3% 17 12 2% 2% Transit2 15 76 1% 4% 11 29 1% 4% Walk 49 31 3% 2% 21 13 3% 2% Bicycle 0 0 0% 0% 0 0 0% 0% Total All Modes 1,908 1,919 100% 100% 819 779 100% 100% 1 Personal vehicle occupancies (entering/exiting): 1.07/1.12. 2 Transit trips include bus and hotel shuttle. Table 72. Morning peak-period and peak-hour person-trip cordon count— Legacy Town Center. Peak Period (4:00 P.M.–7:00 P.M.) Peak Hour (5:00 P.M.–6:00 P.M.) Trips Percent Trips Percent Travel Mode Entering Exiting Entering Exiting Entering Exiting Entering Exiting Personal Vehicle1 3,192 2,832 94% 95% 1,107 1,066 93% 95% Motorcycle 5 5 <1% <1% 3 1 <1% <1% Delivery Vehicle 61 57 2% 2% 20 22 2% 2% Transit2 39 13 1% <1% 22 6 2% 1% Walk 100 60 3% 2% 35 27 3% 2% Bicycle 0 0 0% 0% 0 0 0% 0% Total All Modes 3,397 2,967 100% 100% 1,187 1,122 100% 100% 1 Personal vehicle occupancies (entering/exiting): peak period 1.23/1.16. 2 Transit trips include bus and hotel shuttle. Table 73. Afternoon peak-period and peak-hour person-trip cordon count— Legacy Town Center.

leisure shoppers remaining for dinner or a movie or going home to their residence onsite. Table 75 shows the percentages of internal capture by land use for exiting A.M. peak period trips—that is, trips leaving those land uses. These percentages are based on the inter- views. This table shows the degree of interaction between the various land uses. The greatest synergies during the A.M. peak period are from retail (i.e., convenience retail) to office and residential. As with Mockingbird Station and Atlantic Sta- tion, there is some interchange from restaurant (i.e., the cof- fee shop) to office, although at Legacy Town Center the spe- cialty coffee shop is most of the way across the development from the major office buildings. Table 76 shows similar data for the P.M. peak period. Exiting trips destined to other internal destinations are most frequent from retail to restaurant and residential; from restaurant to res- idential; and from residential, cinema, and hotel to restaurant. This reflects what is expected for an area that has significant amounts of synergy between complementary land uses. Table 77 shows the internal trip capture percentages for entering trips by interchange between land uses. These per- centages are shown as the percentage of total entering trips from individual land uses in Legacy Town Center. This table is for trips entering the various Legacy Town Center land uses. Internal capture percentages are highest entering retail (i.e., convenience retail) and restaurant from onsite residential. Table 78 shows P.M. peak-period trip capture percentages for entering trips by interchange between land uses. With a higher total internal capture, the P.M. peak period also exhibits higher percentages of internal trips on individual interchanges with other land uses. The highest percentage of internal trip capture for entering trips was observed for trips entering office buildings from onsite residential and restaurant; how- ever, inbound trips to office are very small in total numbers. 59 Morning Peak Period Afternoon Peak Period Entering Exiting Entering Exiting Land Use Trips PercentInternal Trips Percent Internal Trips Percent Internal Trips Percent Internal Office 560 15% 61 8% 90 93% 511 3% Retail 119 25% 109 37% 728 30% 766 61% Restaurant 593 26% 550 16% 1,833 33% 1,233 39% Residential 631 6% 1,622 9% 1,352 32% 1,222 34% Cinema1 — — — — 221 32% 108 43% Hotel 187 3% 400 9% 315 36% 299 38% Total All Trips 2,090 15% 2,742 11% 4,539 33% 4,139 37% 1 Cinema not open during morning peak period. Table 74. Peak-period person-trips and percent internal trip capture by land use—Legacy Town Center. Destination Land Use Summary Origin Land Use Office Retail Restaurant Residential Cinema2 Hotel Internal External Total TotalTrips Office —1 0 8 0 — 0 8 92 100 61 Retail 17 —1 6 14 — 0 37 63 100 109 Restaurant 9 2 —1 4 — 1 16 84 100 550 Residential 1 1 7 —1 — 0 9 91 100 1,622 Cinema2 — — — — —1 — — — — — Hotel 0 0 9 0 — —1 9 91 100 400 All Origins 3 1 6 1 — 0 11 89 100 2,742 1 Internal trips within a land use are not included in internal trip capture methodology. 2 Cinema not open during morning peak period. Table 75. Percent distribution of internal trip destinations for exiting trips— Legacy Town Center, morning peak period.

60 Destination Land Use Summary Origin Land Use Office Retail Restaurant Residential Cinema Hotel Internal External Total TotalTrips Office —1 0 1 2 0 0 3 97 100 511 Retail 1 —1 29 26 0 5 61 39 100 766 Restaurant 2 10 —1 18 6 3 39 61 100 1,233 Residential 4 6 21 —1 0 3 34 66 100 1,222 Cinema 0 8 31 2 —1 2 43 57 100 108 Hotel 0 5 33 0 0 —1 38 62 100 299 All Origins 2 5 15 10 2 3 37 63 100 4,139 1 Internal trips within a land use are not included in internal trip capture methodology. Table 76. Percent distribution of internal trip destinations for exiting trips— Legacy Town Center, afternoon peak period. Origin Land Use Summary Destination Land Use Office Retail Restaurant Residential Cinema2 Hotel Internal External Total TotalTrips Office —1 3 9 3 — 0 15 85 100 560 Retail 0 —1 8 17 — 0 25 74 100 119 Restaurant 1 1 —1 18 — 6 26 74 100 593 Residential 0 2 4 —1 — 0 6 94 100 631 Cinema2 — — — — —1 — — — — — Hotel 0 0 3 0 — —1 3 97 100 187 All Destinations 0 2 4 7 — 2 15 85 100 2,090 1 Internal trips within a land use are not included in internal trip capture methodology. 2 Cinema not open during morning peak period. Table 77. Percent distribution of internal trip origins for entering trips— Legacy Town Center, morning peak period. Origin Land Use Summary Destination Land Use Office Retail Restaurant Residential Cinema Hotel Internal External Total TotalTrips Office —1 6 30 57 0 0 93 7 100 90 Retail 0 —1 17 10 1 2 30 70 100 728 Restaurant 0 12 —1 14 2 5 33 67 100 1,833 Residential 1 15 16 —1 0 0 32 68 100 1,352 Cinema 0 0 32 0 —1 0 32 68 100 221 Hotel 0 13 10 12 1 —1 36 64 100 315 All Destinations 0 10 10 9 1 3 33 67 100 4,539 1 Internal trips within a land use are not included in internal trip capture methodology. Table 78. Percent distribution of internal trip origins for entering trips— Legacy Town Center, afternoon peak period.

More significant were trips entering the cinema and retail from restaurants; trips entering restaurants from residential and retail; trips entering residential from retail and restaurants; and trips entering the hotel from retail, restaurants, and res- idential. Inbound trips to restaurant and residential make up the great majority of the total inbound trips to Legacy Town Center land uses. Table 79 shows the percentages of internal trip capture by mode of travel for each entering and exiting trip. This table shows data for trips for which mode of travel was reported. Unlike Mockingbird Station where there were almost no inter- nal driving trips, Legacy Town Center has 8% to 10% of per- sonal driving trips that are internal in the A.M. peak period and 16% to 20% in the P.M. peak period. By contrast, all inbound and nearly all outbound walk and bike trips remained internal to Legacy Town Center. Table 80 shows similar data, but these are for trips made by people based on their original mode of access to Legacy Town Center. These results are limited to those who correctly reported mode of access; a few travelers reported mode of access to the area rather than to Legacy Town Center. During the A.M. peak period, nearly all trips were made by people who arrived at Legacy Town Center as a personal vehicle driver. During the P.M. peak period, there are more trips made by people who used modes of access other than driving. Of those nondrivers (who presumably did not have a vehicle available to drive the next trip unless they were onsite residents), virtu- ally all of the trips entering Legacy Town Center land uses were internal trips. However, for trips exiting the Legacy Town Center land uses, about 25% of the original vehicle passen- gers were going to other internal land uses and about 50% of the taxi/car service passengers were destined internally; this compares with 31% for people who originally arrived onsite by driving a personal vehicle. Given the size of the samples and internal trip capture percentages, no conclusion can be drawn from these data as to whether mode of access affects internal capture rates. Table 81 shows similar data, but this is based on availability 61 Morning Peak Period Afternoon Peak Period1 Entering Exiting Entering Exiting Mode of Travel Trips PercentInternal Trips Percent Internal Trips Percent Internal Trips Percent Internal Vehicle Driver 1,968 10% 2,337 8% 3,569 16% 2,875 20% Vehicle Passenger 6 100% 102 6% 20 100% 177 11% Taxi/Car Service 0 0% 15 0% 0 0% 16 0% Transit (Bus) 0 0% 16 0% 0 0% 0 0% Transit (Hotel Shuttle) 0 0% 140 0% 5 100% 5 100% Walk/Bicycle 118 100% 136 87% 943 100% 1,069 89% 1 Travel mode not reported for 2 entering trips and 7 exiting trips. Table 79. Peak-period person trips and percent internal trip capture by mode of travel—Legacy Town Center. Morning Peak Period1 Afternoon Peak Period2 Entering Exiting Entering Exiting Mode of Access Trips PercentInternal Trips Percent Internal Trips Percent Internal Trips Percent Internal Vehicle Driver 1,892 6% 670 17% 3,862 22% 2,772 31% Vehicle Passenger 0 0% 11 0% 40 100% 159 25% Taxi/Car Service 0 0% 0 0% 23 100% 45 51% Transit (Bus) 6 100% 6 100% 0 0% 0 0% Transit (Hotel Shuttle) 0 0% 0 0% 5 100% 5 100% Walk/Bicycle 9 100% 9 100% 0 0% 0 0% 1 Access mode not reported for 189 entering trips and 2,052 exiting trips. 2 Access mode not reported for 609 entering trips and 1,158 exiting trips. Table 80. Peak-period person-trips and percent internal trip capture by mode of access—Legacy Town Center.

of a personal vehicle for travel. It appears that the internal capture rate for those who did not have a personal vehicle available is higher than for those who could have driven. This is logical since once one travels to a destination, it makes sense to combine trips at the destination area. Florida Survey Data FDOT sponsored two studies of MXDs during the early 1990s. The two studies each covered three developments. While the objectives were similar to those for this project, procedural details were quite different for one of the studies. For the Florida study, the resulting internal capture data do not have as much specificity about internal trip-making. The two studies used different questionnaires and, there- fore, collected different data. One questionnaire provided data by individual trip, and the resulting data were usable for the current project. The other aggregated internal trips, so the data were not usable. For the usable study, data were collected for midday and P.M. peak periods. No data were collected for the A.M. peak period. Three MXDs were included. All three are located in Broward and Palm Beach counties (i.e., the east coast of Florida in the Fort Lauderdale–Palm Beach area). Data for the usable study were not available in original form, so the relevant portion was re-keyed from copies of for- matted printouts of the original data so that they could be analyzed. Some survey trip records were not totally clear and a few ran off the available pages, so there could be minor in- accuracies in a few records; however, this was not judged to compromise the overall value of the data for the purposes of this project. Data for the three Florida developments were collected from mid-morning until 6:15 P.M. Only data matching the data col- lection periods for the NCHRP Project 8-51 pilot studies were used. Hence, the Florida data used covered the P.M. peak period (3:30 P.M. to 6:15 P.M. compared with 3:30 P.M. to 7:00 P.M. for the NCHRP Project 8-51 pilot studies). One other characteristic of the data was different from the pilot study data. Interviews for the pilot studies were pri- marily exit interviews conducted as people departed from specific businesses or other uses and were expanded based on counts of people exiting the same doors. The Florida inter- views were conducted at locations within the developments, some of which were in front of entrances and some of which were along busy walkways. The three developments sur- veyed were • Country Isles, • Village Commons, and • Boca del Mar. Three sites were surveyed in 1993 as part of a study that produced a report titled FDOT Trip Characteristics Study of Multi-Use Developments (18). Each site not only had multiple uses, but also had different parts of the development separated from each other in distance, connectivity, or both. Whereas Mockingbird Station, Atlantic Station, and Legacy Town Center are essentially fully integrated and well connected, the Florida sites were more conventionally arranged in pods. Inter- action between the pods requires crossing parking lots and some separations that are beyond reasonable walking distances. Origin-destination interviews were conducted at each of the study sites. The interviews were structured to collect three different types of information about each site: macro trip- making characteristics, micro trip-making characteristics, and trip length. The characteristics for each type of informa- tion are as follows: 1. Macro trip-making characteristics pertain to the charac- teristics of a trip to and from the site. These characteristics were used in the FDOT research project to categorize trips as captured, primary, diverted, or secondary. 2. Micro trip-making characteristics pertain only to the part of the trip within the site. This information was used in the FDOT research project to determine the number of inter- nally captured trips, the number of the trip stops within the site, and the interaction between land uses. Through examination of individual survey records, researchers for 62 Morning Peak Period1 Afternoon Peak Period2 Entering Exiting Entering Exiting Vehicle Access Trips PercentInternal Trips Percent Internal Trips Percent Internal Trips Percent Internal Yes (Vehicle Driver) 1,892 6% 670 17% 3,862 22% 2,772 31% Yes (Non-Vehicle Driver) 0 0% 11 0% 13 100% 121 11% No Vehicle Access 9 100% 9 100% 52 100% 90 58% 1 Automobile access not reported for 189 entering trips and 2,052 exiting trips. 2 Automobile access not reported for 612 entering trips and 1,156 exiting trips. Table 81. Peak-period person-trips and percent internal trip capture by automobile access—Legacy Town Center.

NCHRP Project 8-51 were able to identify next-stop loca- tions and to categorize them as internal or external, and, if internal, the specific land use or site tenant. 3. The length of the trip made to and from the site was used in the FDOT research project to calculate the percentages of trips originating or ending at various distances from the site. This was collected in the Florida study for use in impact fee analyses. Pedestrian count data were collected at most locations where origin-destination surveys were conducted. The purpose of the pedestrian data was to develop survey sample rates. The number of pedestrians entering and exiting each business was recorded. The areas of each site were sectioned off to establish areas of responsibility for each pedestrian counter so that the counts represented a complete, and not overlapping, count of persons entering or leaving the site. Three different origin-destination survey forms were used. The office and retail/services forms were very similar, each containing 14 questions. The residential survey forms were divided into two different categories: one for incoming sur- veys and one for outgoing surveys. Appendix E includes copies of the forms. Country Isles The Country Isles mixed-use site is located in an area of west Broward County known as Weston. Figure 12 illustrates the general location of the site. Its commercial area is bounded by I-75, SW 14th Street, Weston Road, and Dykes Road. The res- idential component is directly across Weston Road from the Country Isles Shopping Center. The Country Isles mixed-use site covers approximately 61 acres, of which 46 are commercial and 15 are residential. The Country Isles site was surveyed on June 30, 1993. Origin- destination surveys were conducted at 18 different locations within the site. Based on site observations, there appeared to be ample parking. There was no charge for parking anywhere within the site. There was no fixed-route transit service to the site. Site Composition Country Isles consists of three major development areas: • Fairlake at Weston, a multi-family residential area; • Country Isles Shopping Center; and • Indian Trace Shopping Center. Figure 13 shows the layout of Country Isles. Total com- mercial building square footage was 252,681, with about 70% retail (175,697 gross sq ft); 25% office (64,234 gross sq ft); and 5% daycare. The total number of dwelling units was 368. Fig- ure 13 shows the tenant types and locations. Proximity of Commercial Competition At the time of data collection, the Country Isles develop- ment was the primary shopping center site serving the Weston and Bonaventure areas. The closest competing shopping cen- ter was Westgate Square, located approximately 2 miles away. Both sites had a supermarket, drug store, restaurants, banks, and small retail land uses; however, the Country Isles develop- ment was larger and more centrally located within the Weston community. It also offered a wider variety of land uses includ- ing medical and professional offices, a movie theater, daycare, and a convenience store. Finally, in the opinion of the original FDOT report authors, the general appearance of the Country Isles site (e.g., landscaping, site entrance) was more appealing than that of Westgate Square. Site Components The descriptions that follow are grouped according to how land uses are aggregated for the data collection and data analysis. Country Isles Shopping Center. The Country Isles Shop- ping Center was the primary retail center of this mixed-use site. Its 33 businesses included • A supermarket and drug store; • Five restaurants, including pizza, bagel, Italian, and Chinese; • Numerous retail stores, including ice cream, party goods, video rental, shoes, liquor, children’s clothing, framing, bicycles, florist, hardware, cards; and • Several services such as medical offices, insurance agents, banks, shoe repair, a hair salon, a dry cleaner, a weight clinic, real estate agencies, an eye center, and a travel agency. The Country Isles Shopping Center was 99,651 gross sq ft in size. Its largest tenant was a supermarket composing about 40% of the space. Restaurant use was 11%, bank space was 5%, and miscellaneous office space was 6%. The convenience retail, service units, and the supermarket composed about 78% of the shopping center space. The shopping center was approximately 90% occupied at the time of data collection. The shopping center had approximately 459 parking spaces, all in a surface lot. Indian Trace Shopping Center. The Indian Trace Shop- ping Center included various restaurant, retail service, small office, and movie theater land uses. Because Indian Trace faces the back of the Country Isles Shopping Center, the most direct 63

64 Source: FDOT Trip Characteristics Study of Multi-Use Developments, Tindale-Oliver & Associates, final report, December 1993, p. II-4 Figure 12. Country Isles as depicted in the FDOT report (18).

Source: FDOT Trip Characteristics Study of Multi-Use Developments, Tindale-Oliver & Associates, final report, December 1993, p. II-6 65 Figure 13. Country Isles land use details as depicted in the FDOT report (18). path between their primary parking areas was approximately 1,400 ft. Indian Trace tenants included • Four restaurants, ranging from pizza to yogurt to deli; • Several services including mail shipping, a dance studio, a dry cleaner, a karate school, health foods, an animal clinic, insurance, real estate, and a hair salon; • Several small offices including a church office and medical offices; and • A movie theater. The Indian Trace Shopping Center was 68,400 gross sq ft in size. Its largest tenant was a movie theater composing about 38% of the space. Restaurant use was 8% and miscellaneous office space was 7%. Convenience Stores. The Country Isles site also had two convenience stores, both with gasoline pumps. The first was located near the Country Isles shopping center and was 924 gross sq ft in size. It had 12 vehicle-fueling positions. The second was located near the Indian Trace shopping center and was 2,946 gross sq ft, with 8 vehicle-fueling posi- tions. Together, the two convenience stores were 3,870 sq ft (less than 2% of the total commercial square footage at Coun- try Isles). Of the three sites surveyed as part of the FDOT inter- nal capture research project and the three sites surveyed as part of NCHRP Project 8-51, Country Isles was the only mixed-use site with onsite convenience stores with gasoline pumps. Fast-Food Restaurant. There was a stand-alone, fast- food restaurant located in the northern most corner of the

site. Its building was 3,776 gross sq ft (less than 2% of the total commercial square footage at Country Isles). During the P.M. peak period for the NCHRP Project 8-51 analysis window, only two useable interviews were conducted. Therefore, no trips from fast-food restaurant were assumed to be internal to the site, but trips to the fast-food restaurant from other Country Isles uses were recorded and reported in the follow- ing sections. Office Buildings. The Country Isles site had three stand- alone office buildings, totaling 64,234 gross sq ft. There was a three-story office building complex located just to the south of the fast-food restaurant. The complex had 26,000 sq ft and included a bank with drive-through facilities, a real estate agency, some medical office space, and some general office space. About one-third of the space was the bank. The build- ing had approximately 118 parking spaces. Persons leaving this office building were not interviewed. In the west central part of the site, there was a three-story office building complex. The building was 10,000 sq ft and included professional offices and a bank with drive-through facilities. About one-third of the space was occupied by the bank. The building had approximately 46 parking spaces and was located approximately 300 ft from the center of the Country Isles Shopping Center. In the southern part of the site, there was a three-story office building complex. The building was 28,234 sq ft and included a bank with drive-through facilities, an insurance agency, and professional office space. About 15% of the building space was occupied by the bank. The building had approximately 113 parking spaces and was located approximately 300 ft from the center of the Country Isles Shopping Center. Fairlake at Weston. Fairlake at Weston is a residential, multi-family apartment development with 368 units. Its occu- pancy level at the time of this study was estimated at 90%. The approximate center of Fairlake was located 1,200 ft from the supermarket at the Country Isles Shopping Center. Daycare Center. The Country Isles site had a daycare cen- ter located near the northern edge of the Indian Trace Shop- ping Center, approximately 1,700 ft from the supermarket at the Country Isles Shopping Center and 2,600 ft from the center of the Fairlake residential development. The daycare center was 12,750 gross sq ft. Of the three sites surveyed as part of the FDOT internal capture research project and the three sites surveyed as part of NCHRP Project 8-51, Country Isles was the only one with a daycare center. Data Collection Origin-destination interviews were conducted at 18 sta- tions throughout Country Isles. Different expansion factors were developed for each site (i.e., residential, office, and com- mercial land use categories) based on pedestrian counts, vehi- cle counts, and vehicle-occupancy counts. Table 82 lists the numbers of useable surveys collected at each land use. Also interviewed were 13 (or 5%) of the 269 inbound motorists at the Fairlake at Weston residential site. Analysis of Internal Capture Table 83 summarizes the overall internal capture found at the individual Country Isles land uses. The data shown in the second column represent the percentage of trips from the origin land use that are internally captured within the study site. Data in the right column show the same for trips to the destination land use. To more fully understand these overall internal capture rates for each land use, it was necessary to investigate internal capture rates for pairs of land uses. The following presents these data. Table 84 presents the distribution of trip destinations for trips exiting each of the surveyed Country Isles land uses. Sep- arate sets of values are listed for the Country Isles Shopping Center, for the Indian Trace Shopping Center, and for the combined trips exiting both shopping centers. The distribu- tion is as follows: • Of trips leaving the onsite office buildings, 25% had an internal retail destination—20% at the shopping centers and 5% at either of the two gasoline/convenience stores. 66 Land Use Exit Movements Usable Interviews Percent Interviewed Office 573 45 8% Retail 1,644 123 7% Gasoline/Convenience 466 65 14% Residential 173 44 25% Daycare 396 73 18% Total 3,252 350 11% Table 82. P.M. peak-period useable surveys and sample rate— Country Isles.

• Of the trips leaving the Country Isles Shopping Center, 12% were destined to a non-shopping center internal use; 29% of the trips leaving the Indian Trace Shopping Center were destined for the same. • When combined, the two shopping centers sent 1% of their outbound trips to onsite office buildings; 7% to the onsite gasoline/convenience stores; 2% to the onsite, free- standing fast-food restaurant; and 4% to the onsite resi- dential area. • Of the trips leaving the onsite gas/convenience stores, 4% were destined to onsite retail; the remainder travel to exter- nal destinations. • Of trips leaving onsite residential, 36% were traveling to onsite destinations—25% to the shopping centers, 9% to the gasoline/convenience stores, and 2% to the fast-food restaurant. • Of trips leaving the onsite daycare center, 17% were trav- eling to onsite destinations—15% to the shopping cen- ters and 1% each to the gasoline/convenience stores and to residential. Table 85 shows the distribution of trip origins for trips entering each of the surveyed Country Isles land uses. Sepa- rate sets of values are listed for the Country Isles Shopping Center, for the Indian Trace Shopping Center, and for the combined trips entering both shopping centers. The distribu- tion is as follows: • Of trips entering the onsite office buildings, 2% had an internal origin, all from the shopping centers (as opposed to 25% of the exiting trips that are internal as shown in Table 84). • Of the trips entering the Country Isles Shopping Center, 13% arrived from internal use (same as for exiting); also arriving from an internal use were 34% of the trips entering the Indian Trace Shopping Center (greater than the per- centage exiting). When combined, the two shopping cen- ters received 7% of their inbound trips from onsite office buildings, 1% from the onsite gasoline/convenience stores, 3% from the onsite residential area, and 4% from the onsite daycare center. 67 Land Use Percent Internal Capture as Origin Land Use Percent Internal Capture as Destination Land Use Office 25% 2% Shopping Center 20% for Country Isles 44% for Indian Trace 26% overall 20% for Country Isles 57% for Indian Trace 28% overall Gasoline/Convenience 4% 36% Fast-Food Restaurant Not available 30% Residential 36% 25% Daycare 18% 0% Total 22% 24% Table 83. P.M. peak-period percent internal capture by land use— Country Isles. Percent Internal Trips by Destination Land Use1 Total Origin Land Use Office ShoppingCenter Gas/ Conv FF Rest Residential Day Care Total Internal Percent External Office — 20 5 0 0 0 25 75% 100 Country Isles Shopping Center 0 — 9 3 0 0 12 88% 100 Indian Trace Shopping Center 2 — 5 3 19 0 29 71% 100 Both Shopping Centers 1 — 8 2 5 0 16 84% 100 Gasoline/Convenience 0 4 — 0 0 0 4 96% 100 Residential 0 25 9 2 — 0 36 64% 100 Daycare 0 15 1 0 1 — 17 83% 100 1 Calculated to exclude trips within the same land use. Table 84. P.M. peak-period percent distribution of internal trip destinations for exiting trips—Country Isles.

• Of the trips entering the onsite gasoline/convenience stores, 36% came from onsite uses—6% from the office, 25% from the shopping centers, 3% from residential, and 1% from the daycare center. • Of the trips that traveled to the onsite, free-standing, fast- food restaurant, 30% came from onsite uses—27% from the shopping centers and 3% from residential. • Of the trips entering onsite residential, 25% traveled from an onsite origin—23% from the shopping centers and 2% from the daycare. • Of the trips entering the onsite daycare center, 100% trav- eled from outside Country Isles. Village Commons The Village Commons site is located within the southwest- ern limits of the City of West Palm Beach in Palm Beach County. The overall mixed-use site straddles Village Boulevard, immediately northwest of Palm Beach Lakes Boulevard. The Village Commons mixed-use site encompasses approximately 72 acres, of which 54 are commercial and 18 are residential. Fig- ure 14 shows the general layout of Village Commons. Figure 15 shows the types and locations of tenant land uses. Village Commons was surveyed on July 14, 1993. Origin- destination surveys were conducted at 14 different locations within the site. There was no charge for parking anywhere within the site. Additionally, based on field observations, there was an adequate parking supply to service all land uses, with the possible exception of the health spa. There was no formal fixed-route transit service provided to the Village Commons site. Site Composition The Village Commons site has four major development areas: the Village Commons Shopping Center, the Brandywine Center, various office buildings located throughout the site, and the Pointe multi-family residential community. Total commercial square footage in Village Commons was 524,350 with 34% retail (179,840 sq ft), 57% office (297,581 sq ft), and 9% health spa. The total number of dwelling units was 317. Proximity of Commercial Competition There were numerous office, restaurant, hotel, and retail land uses (including a regional mall) that were proximate to the Village Commons site. Site Components The descriptions that follow are grouped according to how land uses are aggregated for the data collection and data analysis. Village Commons Shopping Center. The Village Com- mons Shopping Center included the following: • A supermarket and drug store; • Eight restaurants ranging from natural foods to Japanese to bagels; • Retail stores including computers, clothing, video, fram- ing, gifts, shoes, cards, jewelry, maternity, sporting goods, consignment, ice cream, paint, cell phones, liquor, and flowers; and 68 Percent Internal Trips by Destination Land Use1 Shopping Centers Origin Land Use Office Country Isles Indian Trace Subtotal Gas/ Conv FF Rest Residential Daycare Office — 6 17 8 6 0 0 0 Shopping Center 2 — — — 25 27 23 0 Gas/Convenience 0 1 5 1 — 0 0 0 Residential 0 3 3 3 3 3 — 0 Daycare 0 3 9 4 1 0 2 — Total Internal 2 13 34 17 36 30 25 0 External 98 87 66 83 64 70 75 100 Total 100 100 100 100 100 100 100 100 1 Calculated to exclude trips within the same land use; totals shown may not equal sums due to rounding. Table 85. P.M. peak-period percent distribution of internal trip origins for entering trips—Country Isles.

69 Source: FDOT Trip Characteristics Study of Multi-Use Developments, Tindale-Oliver & Associates, final report, December 1993, p. II-11 Figure 14. Village Commons site layout as depicted in the FDOT report (18).

70 Source: FDOT Trip Characteristics Study of Multi-Use Developments, Tindale-Oliver & Associates, final report, December 1993, p. II-14 Figure 15. Village Commons tenant details as depicted in the FDOT report (18).

• Service establishments such as banks, mail shipping, dry cleaning, a leasing office, a weight clinic, a hair salon, a tanning salon, real estate, a travel agency, an animal clinic, insurance, and a mortgage company. The shopping center was 170,740 gross sq ft in size. Its largest tenant was a supermarket encompassing 23% of the overall center space. There were a significant number of restau- rants in the shopping center, composing 19% of the center space. Banks composed 6% and office space composed 4%. Despite being a large component of the overall shopping cen- ter, restaurants were not separated in the data analysis because the pedestrian count data did not separate between restaurant and non-restaurant volumes, thereby eliminating the possibil- ity of developing reasonably accurate survey expansion factors for restaurant and non-restaurant trips. Brandywine Center: Sit-Down Restaurant. The Brandy- wine Center contained four buildings: a sit-down chain res- taurant, a bank, and two office buildings. The restaurant in Brandywine Center was 9,100 gross sq ft. The bank and office buildings are included in the next section. Office Buildings. Village Commons had a total of nearly 300,000 sq ft in office buildings. The two three-story office buildings in Brandywine Center totaled 122,870 sq ft (of which 4% was in a bank) and were located approximately 1,500 ft from the Village Commons Shopping Center. Although side- walks and crosswalks were available for pedestrian use, the fact that Village Boulevard was a four-lane divided roadway possibly would discourage pedestrian movement between the Brandywine and Village Commons centers. An office building was located at the northeast corner of Harvard Circle, approximately 1,800 ft from the Village Commons Shopping Center. It totaled 96,270 sq ft. The three office buildings located along Columbia Drive totaled 45,524 sq ft and were approximately 1,800 ft from the Village Commons Shopping Center. A multi-story office/bank building located at the southeast corner of Brandywine Road and Village Boulevard bounded by Columbia Drive and Olympic Place totaled 32,917 sq ft (of which 18% was in a bank). It was located approximately 800 ft from the Village Commons Shopping Center. Persons leaving this office building were not interviewed. The Pointe. The Pointe development is a residential multi- family development containing 317 units. Its occupancy was estimated to be approximately 93% on the survey date. The FDOT research project was not allowed to conduct origin- destination surveys at the Pointe. Village Commons residen- tial internal trips were estimated using Village Commons vehicle count data and average residential internal trip rates observed at the other two sites. The approximate center of the Pointe is located 900 ft from the supermarket at Village Com- mons Shopping Center. Health Spa. There is a health spa located in the area bounded by Village Boulevard and Olympic Place. It encom- passes 46,929 sq ft. An origin-destination survey was not conducted at the health spa, and no trips to the health spa were identified during the surveys conducted at other uses at the Village Commons site. Data Collection Origin-destination interviews were conducted at 14 sta- tions throughout Village Commons. Different expansion fac- tors were developed for each site (i.e., residential, office, and commercial land use categories) based on pedestrian counts, vehicle counts, and vehicle-occupancy counts. Table 86 lists the numbers of useable surveys collected at each land use. Analysis of Internal Capture Table 87 summarizes the overall internal capture found at the individual Country Isles land uses. In order to more fully understand these overall internal capture rates for each land use, it was necessary to investigate internal capture rates for pairs of land uses. Those data follow. 71 Land Use Exit Movements Usable Interviews Percent Interviewed Office 718 78 11% Retail 1,216 253 21% Sit-Down Restaurant 167 27 16% Residential 179 Not Interviewed Not Interviewed Total 2280 358 16% Table 86. P.M. peak-period usable surveys and sample rate— Village Commons.

Table 88 presents the distribution of trip destinations for trips exiting each of the surveyed Village Commons land uses: • Of trips leaving the office buildings onsite, 7% have an internal destination—6% to the shopping center and 1% to residential. • Of the trips leaving the Village Commons Shopping Cen- ter, 7% were destined to an internal use, all residential. • Of the trips leaving the onsite sit-down restaurant, 7% were destined to onsite destinations—half to the shopping center and half to residential. • Of trips leaving the onsite residential, 27% were traveling to onsite destinations—25% to the shopping center and 2% to the fast-food restaurant. (The Pointe was not surveyed; these internal trip-making estimates are based on values derived at the other two FDOT research sites and on observed bal- ancing of trips into and out of the Village Commons uses.) Table 89 presents the distribution of trip origins for trips entering each of the surveyed Village Commons land uses: • None of the trips entering the office buildings onsite had an internal origin. • Of the trips entering the Village Commons Shopping Cen- ter, 7% (same as for exiting) arrived from an internal use— 3% each from office and residential and less than 1% from the sit-down restaurant. • Of the trips traveling to the onsite, sit-down restaurant, 4% came from onsite uses, all from the residential. • Of the trips entering onsite residential, 37% came from an onsite origin—30% from the shopping center, 4% from office, and 2% from the sit-down restaurant. Boca Del Mar The Boca Del Mar site is located in southwest Palm Beach County. The mixed-use site is situated in the southwest quad- rant of the intersection of Powerline Road and Palmetto Park Road. The commercial component of the site encompasses 42 acres. Figures 16 and 17 show the layout of Boca del Mar, and Figure 18 shows the tenant land uses and locations for the eastern portion of Boca del Mar. The western portion is all residential. Boca del Mar was surveyed on July 21, 1993. Origin- destination interviews were conducted at 20 different locations within the site. Based on field observations, there appeared to be an ample parking supply to support the site land uses and there was no charge for parking at the site. There was no for- mal fixed-route transit system serving the site. Site Composition Boca Del Mar has six major development components: 1. Garden Shops at Boca; 2. Palms Plaza; 3. A multi-story office building at the northwest corner of the site; 72 Land Use Percent Internal Capture as Origin Land Use Percent Internal Capture as Destination Land Use Office 7% 0% Retail 7% 7% Sit-Down Restaurant 7% 4% Residential 27% 37% Total 9% 9% Table 87. P.M. peak-period percent internal capture by land use—Village Commons. Destination Land Use Origin Land Use Office Retail Restaurant Residential Total Internal External Total Office — 6 0 1 7 93 100 Retail 0 — 0 7 7 93 100 Sit-Down Restaurant 0 4 — 4 7 93 100 Residential 0 25 2 — 27 73 100 Table 88. P.M. peak-period percent distribution of internal trip destinations for exiting trips—Village Commons.

Source: FDOT Trip Characteristics Study of Multi-Use Developments, Tindale-Oliver & Associates, final report, December 1993. 73 Figure 16. Boca del Mar eastern portion site layout as depicted in the FDOT report (18). Destination Land Use Origin Land Use Office Retail Restaurant Residential Office — 3 0 4 Retail 0 — 0 30 Sit-Down Restaurant 0 <1 — 2 Residential 0 3 4 — Total Internal 0 7 4 37 External 100 93 96 63 Total 100 100 100 100 Table 89. P.M. peak-period percent distribution of internal trip origins for entering trips— Village Commons. 4. A multi-story bank and office building at the northeast corner of the site; 5. Camden Court, a multi-family residential complex on the south side of the site; and 6. Various residential areas to the west of the site. Please note that only the Camden Court residential area was considered internal to the mixed-use site for NCHRP Project 8-51 because insufficient surveys and counts were conducted at the “various residential areas” to enable consis- tent treatment of the survey responses. Total commercial square footage in Boca del Mar was 501,254 with 41% retail (207,787 sq ft) and 59% office (293,467 sq ft). There were 1,144 total dwelling units of

which 513 were single family detached, 517 were town- houses, and 114 were apartments. The total number of dwelling units within Camden Court was 190. All land uses located within the site were accessible via a service road system that bisects the site. All vehicular trips between site land uses could be made without having to use the arterial road system adjacent to the site. Proximity of Commercial Competition Immediately to the north of the Boca Del Mar site was a competing shopping center with several similar retail busi- nesses, including a supermarket. However, the site was older than Boca del Mar and, in the opinion of the original FDOT research team, its general appearance (e.g., landscaping, site entrance) was not as appealing. Site Components The descriptions that follow are grouped according to how land uses are aggregated for the data collection and data analysis. Garden Shops at Boca. The Garden Shops at Boca was the retail center of the mixed-use site. Its 52 tenants included the following: • A supermarket and a drugstore; • Six restaurants; 74 Source: FDOT Trip Characteristics Study of Multi-Use Developments, Tindale-Oliver & Associates, final report, December 1993, p. II-20. Figure 17. Boca del Mar western portion site layout as depicted in the FDOT report (18).

• Various retail stores including men’s, women’s, and chil- dren’s clothing; books; gifts; ice cream; jewelry; liquor; lug- gage; cosmetics; lamps; framing; pet supplies; a boutique; and a florist; and • Various services including a medical office, real estate, dry cleaning, eye care, a psychic reader, manicure/facials, photo development, interior design, a travel agency, a hair salon, and a mail shipper. The shopping center was 140,686 gross sq ft in size. Its largest tenant was a supermarket encompassing about 29% of the overall center space. Restaurants composed 10% of the center space, banks composed 5 percent, and office space composed 7%. The Garden Shops were approximately 95% occupied at the time of the survey. Palms Plaza Shopping Center. The Palms Plaza Shop- ping Center was oriented facing away from the Garden Shops at Boca. The most direct path between their primary parking areas was approximately 1,100 ft. Palm Plaza had 27 tenants, including • Four restaurants, ranging from a major sit-down restau- rant chain to Japanese to fast food; • Various retail stores including clothing, baked goods, gifts, computers, jewelry, maternity wear, cards, and eye wear; and • Various services including a travel agency, photo develop- ment, a hair salon, framing, a dry cleaner, a travel agency, a real estate agency, and a bank. 75 Source: FDOT Trip Characteristics Study of Multi-Use Developments, Tindale-Oliver & Associates, final report, December 1993, p. II-25. Figure 18. Boca del Mar east end tenant details as depicted in the FDOT report (18).

The shopping center was 63,070 gross sq ft in size. Restau- rants composed 25% of the center space, banks composed 4 percent, and office space composed 7%. The remaining 64% was the mix of convenience retail and service businesses. The shopping center was approximately 80% occupied at the time of the survey. Stand-Alone Fast-Food Restaurant. The Boca Del Mar Site had one stand-alone, fast-food restaurant. Its size was 4,031 sq ft. Office Buildings. The Boca del Mar had two office build- ings. The multi-story office complex on the west side of the site had surface parking around the building plus secured underground parking. Its size was 114,881 sq ft and it was located approximately 800 ft from the Garden Shops at Boca. There were a variety of professional businesses including some medical offices located within the office complex. This build- ing was surveyed. A 178,586 sq ft, multi-story office center was located on the northeast corner of the Boca del Mar site. Located in the office building are a bank (2% of the total square footage) and professional offices including a number of medical offices. Permission to obtain traffic counts at the access points and origin/destination surveys at this part of the site was not granted by the site property manager. Camden Court. Camden Court was a residential multi- family community located immediately to the south of the retail shopping center. Camden Court included 190 apart- ment units, with an occupancy of 97% on the survey date. The approximate center of Camden Court was located 900 ft from the supermarket at the Garden Shops at Boca. Data Collection Origin-destination interviews were conducted at 20 stations throughout Boca del Mar. The data collection conformed to the methods used for the other Florida sites reported here. Different expansion factors were developed for each site (i.e., residential, office, and commercial land use categories) based on pedestrian counts, vehicle counts, and vehicle- occupancy counts. Table 90 lists the numbers of useable surveys collected at each land use. Analysis of Internal Capture Table 91 summarizes the overall internal capture found at the individual Boca del Mar land uses. To more fully under- stand these overall internal capture rates for each land use, it was necessary to investigate internal capture rates for pairs of land uses. The following includes the data. 76 Land Use Exit Movements Usable Interviews Percent Interviewed1 Office 139 30 22% Retail 1,672 267 16% Fast-Food Restaurant 100 33 33% Residential 108 95 88% Total 2,019 425 21% 1 Also interviewed were 18 (or 11%) of the 168 inbound residential motorists. Table 90. Usable P.M. peak-period surveys and sample rate— Boca del Mar. Land Use Percent Internal Capture as Origin Land Use Percent Internal Capture as Destination Land Use Office 0% 0% Retail 4% for Garden Shops at Boca 7% for Palms Plaza 5% overall 7% for Garden Shops at Boca 1% for Palms Plaza 5% overall Fast-Food Restaurant 24% 3% Residential 44% 35% Total 8% 7% Table 91. P.M. peak-period percent internal capture by land use— Boca del Mar.

Table 92 presents the distribution of trip destinations for trips exiting each of the surveyed Boca del Mar land uses. Sep- arate sets of values are listed for the Garden Shops at Boca, for the Palms Plaza Shopping Center, and for the combined trips exiting both shopping centers: • None of the surveyed trips leaving the office buildings onsite had an internal destination. • Of the trips leaving the Garden Shops, 4% were destined to an internal use; none of the trips leaving Palms Plaza were. • When combined, the two shopping centers sent less than 1% of the surveyed trips to the free-standing, fast-food restaurant and 4% to the onsite residential area. • Of the trips leaving the onsite, free-standing fast-food restau- rant, 24% were destined to onsite destinations, 18% to the shopping centers, and 6% to residential. • Of trips leaving onsite residential, 44% were travelling to onsite destinations—42% to the shopping centers and 2% to the fast-food restaurant. Table 93 presents the distribution of trip origins for trips entering each of the surveyed Boca del Mar land uses. Sepa- rate sets of values are listed for the Garden Shops at Boca, for the Palms Plaza Shopping Center, and for the combined trips entering both shopping centers: • All of the trips entering the office buildings onsite had an external origin. • Of the trips entering the Garden Shops at Boca, 4% arrived from an internal use; 1% of the trips entering the Palms Plaza Shopping Center also arrived from an inter- nal use. • When combined, the two shopping centers received 3% of their inbound trips from onsite uses—1% from fast food, and 2% from residential. • Of the trips traveling to the onsite, free-standing, fast-food restaurant, 3% arrived from onsite uses—1% from the shopping centers and 2% from residential. 77 Percent Internal Trips by Destination Land Use1Origin Land Use Office Shopping Center FF Rest Residential Total Internal External Total Office — 0 0 0 0 100 100 Garden Shops at Boca 0 — <1 4 4 96 100 Palms Plaza Shopping Center 0 — 0 0 0 100 100 Retail – Total 0 — <1 3 4 96 100 Fast-Food Restaurant 0 18 — 6 24 76 100 Residential 0 42 2 — 44 56 100 1 Calculated to exclude trips within the same land use. Table 92. P.M. peak-period percent distribution of internal trip destinations for origin land uses—Boca del Mar. Percent Internal Trips by Destination Land Use1 Origin Land Use Office Retail – Garden Shops Retail – Palms Plaza Retail – Total FF Rest Residential Office — 0 0 0 0 0 Retail 0 — — — 1 32 Fast-Food Restaurant 0 1 0 1 — 4 Residential 0 3 1 2 2 — Total Internal 0 4 1 3 3 35 External 100 96 99 97 97 65 Total 100 100 100 100 100 100 1 Calculated to exclude trips within the same land use. Table 93. P.M. peak-period percent distribution of internal trip origins for destination land uses—Boca del Mar.

• Of the trips entering onsite residential, a total of about 35% arrived from an onsite origin—almost 32% from the shopping centers and nearly 4% from the fast-food restaurant. Comparison of Findings for Pilot Study Sites and Florida Sites Similarities and Differences Among the Developments The three MXDs surveyed in this project’s pilot studies— Mockingbird Station, Atlantic Station, and Legacy Town Center—are similar in terms of the uses they have and the general proximities of their non-residential uses. All three are highly interconnected. All three are very walkable in their central areas (where commercial uses are located). All three have specialty retail and a range of restaurants from specialty coffee shops to high-end restaurants. All three have a cinema. Most commercial and retail businesses in each of the three developments are small; a few would be considered medium- sized. All have a variety of restaurants. Only Atlantic Station has a large retailer—a national chain department store. However, there are differences. Mockingbird Station is very compact. Driving between internal destinations is an incon- venience compared with walking. The maximum walking dis- tance is about 700 ft. There is a rail transit station next to and directly connected to the development and transit is used as a significant mode of access; that station is also served by six bus routes. The apartment building sits in the middle of and on top of the central commercial building. Mocking- bird Station has no hotel. Atlantic Station’s main residential area extends away from the commercial area and is up to 3,400 ft away. There is a major grocery store there. There is also the retail department store, the only one among the three developments. Transit service that is used by Atlantic Station patrons and residents is via a dedicated shuttle to a nearby MARTA rail station about a mile away. One MARTA bus route serves the area conve- niently, and almost no use was made of it by persons going to and from Atlantic Station. While Mockingbird Station and Atlantic Station both have midtown locations and were redevelopment sites in the mid- dle of fully developed areas, Legacy Town Center is an outer suburban development within a rapidly developing area. At the time of the survey, the area surrounding Legacy Town Center is fully or almost fully developed. Others are in var- ious stages of partial development. Overall the area within about 2 miles is roughly two-thirds developed. Transit is virtually unused and has little presence although a hotel shut- tle does provide service for its patrons to nearby destinations. While Legacy Town Center is well connected internally, its land uses tend to be more concentrated into specific areas of the site. Hence, while the three study sites are truly integrated MXDs, they are not a truly homogenous trio of samples. This is sim- ilar to most land use categories included in the ITE trip gener- ation database, although these three developments are more similar than those included in many ITE land use categories. It is also important to note that the three developments represent a range of typical conditions in which MXDs are developed. The three Florida sites—Village Commons, Country Isles, and Boca del Mar—are less compact than the sites surveyed in this project. The Florida developments are structured sets of development pods separated by parking lots or streets; they are less well connected, less compact, and also have fewer interact- ing uses than the three developments surveyed in this project. The six developments together could be considered repre- sentative of the range of types of MXDs in the range of 1⁄2 to 3 million gross sq ft of development. They are much larger and more diverse than a corner development that might consist of an office building that includes retail and restau- rant uses. On the other extreme, the six developments are not as fully self-contained as a downtown or even a major suburban activity center; hence, use of the data from these developments should not be considered applicable to either very small MXDs or downtowns without having data that confirm similarities. Findings from this project were compared with those con- tained in NCHRP Report 323: Travel Characteristics at Large- Scale Suburban Activity Centers (26), based on limited data in that report on internal capture. As noted previously, suburban activity centers—probably due to their size and greater mix- use uses and choices—have a broader range of internal capture percentages. While the surveys conducted obtained slightly dif- ferent data, internal capture for segments of suburban activity center populations ranged between 6% and 68% with averages among activity centers surveyed ranging from 14% to 58% for specific population types (e.g., office employees). Internal Trip-Making Tables 94 through 97 summarize the internal capture per- centages found for the three developments surveyed as part of this project plus the three Florida sites (P.M. data available only). The tables show internal capture percentages for the origin ends of trips as well as for the destination ends, similar to what was shown in the findings for each development. As is shown in previous tables, there are no values for trips between the same land uses because ITE trip generation rates already reflect trips within the same land use on the same site. Bold italicized percentages are the highest for each land use pair combination. 78

Table 94 shows and compares the distributions of internal trip destinations for exiting trips during the A.M. peak period. Data are available for only the three sites surveyed for this project. It is helpful to compare the internal capture percent- ages by land use pair. Table 94 shows a range of internal capture percentages among the three developments for many of the land use pairs—for example, for trips from office to restaurant, the per- centages found were 5%, 8%, and 63%, respectively. The per- centages are a product of a number of factors, including the amounts of office and restaurant space exchanging interacting trips; the proximity and quality of connections between the interacting land uses (data available for those two factors); similar off-site opportunities; and the relative attractiveness of the destination as that type of land use (data not available for the last two considerations). For example, the 63% office to restaurant at Mockingbird Station results in part due to a 200-ft walking distance to a popular specialty coffee shop. The relationships of internal capture percentage to trip end constraints and proximity are examined later in this chapter. 79 Destination Land Use Origin Land Use MXD Site Office Retail Restaurant Residential Cinema Hotel Atlantic Station 28 5 0 — 0 Legacy Town Center 0 8 0 — 0 Mockingbird Station — 63 1 — — Boca Del Mar — — — — — Country Isles — — — — — Office Village Commons — — — — — Atlantic Station 29 13 0 — 0 Legacy Town Center 17 6 14 — 0 Mockingbird Station — — — — — Boca Del Mar — — — — — Country Isles — — — — — Retail Village Commons — — — — — Atlantic Station 31 14 0 — 3 Legacy Town Center 9 2 4 — 1 Mockingbird Station 25 — 3 — — Boca Del Mar — — — — — Country Isles — — — — — Restaurant Village Commons — — — — — Atlantic Station 1 1 0 — 0 Legacy Town Center 1 1 7 — 0 Mockingbird Station 2 — 20 — — Boca Del Mar — — — — — Country Isles — — — — — Residential Village Commons — — — — — Atlantic Station — — — — — Legacy Town Center — — — — — Mockingbird Station — — — — — Boca Del Mar — — — — — Country Isles — — — — — Cinema Village Commons — — — — — Atlantic Station 75 14 6 0 — Legacy Town Center 0 0 9 0 — Mockingbird Station — — — — — Boca Del Mar — — — — — Country Isles — — — — — Hotel Village Commons — — — — — Table 94. Unconstrained internal capture rates for exiting trips, all sites—A.M. peak period.

The highest percentages found for each land use pair in Table 94 result from actual survey findings. The fact that lower percentages occurred elsewhere means only that the conditions—mainly balance between origin and destination land use demands for the trips between them plus the prox- imity (or other factors for which data are not available)—were not as ideal. These highest percentages represent the most unconstrained interchanges surveyed among the six sites— that is, the prevailing conditions reflect the best match result- ing in the most interactions between the two land uses from among the developments surveyed. Of the cells in Table 94 where morning peak period data for all three developments are available, only two cells show one percentage substantially higher than the other two: office-to- restaurant and residential-to-restaurant, both at Mockingbird Station. Both of these involve a popular coffee shop very close to the apartment building entrance and the office building entrance. It is natural that a high percentage of those types of trips would go to the adjacent onsite coffee shop during the A.M. peak. Table 95 shows similar comparisons for the P.M. peak- period exiting trips. Data are available for all six developments. 80 Destination Land Use Origin Land Use MXD Site Office Retail Restaurant Residential Cinema Hotel Atlantic Station 6 3 0 0 0 Legacy Town Center 0 1 2 0 0 Mockingbird Station 9 4 2 0 — Boca Del Mar 0 0 0 — — Country Isles 20 0 0 — — Office Village Commons 6 0 1 — — Atlantic Station 2 19 13 4 1 Legacy Town Center 1 29 26 0 5 Mockingbird Station 1 20 7 4 — Boca Del Mar 0 0 3 — — Country Isles 1 2 5 — — Retail Village Commons 0 0 7 — — Atlantic Station 1 41 3 8 7 Legacy Town Center 2 10 18 6 3 Mockingbird Station 3 38 3 2 — Boca Del Mar 0 18 6 — — Country Isles — — — — — Restaurant Village Commons 0 4 4 — — Atlantic Station 0 9 3 0 1 Legacy Town Center 4 6 21 0 3 Mockingbird Station 1 31 11 0 — Boca Del Mar 0 42 2 — — Country Isles 0 25 2 — — Residential Village Commons 0 25 2 — — Atlantic Station 2 21 11 8 0 Legacy Town Center 0 8 31 2 2 Mockingbird Station 0 17 25 8 — Boca Del Mar — — — — — Country Isles — — — — — Cinema Village Commons — — — — — Atlantic Station 0 16 68 2 0 Legacy Town Center 0 5 33 0 0 Mockingbird Station — — — — — Boca Del Mar — — — — — Country Isles — — — — — Hotel Village Commons — — — — — Table 95. Unconstrained internal capture rates for exiting trips, all sites—P.M. peak period.

Because there are more percentages, there are fewer cells where one value far exceeds all others when at least three val- ues are given. Table 96 shows a similar comparison for entering trips during the morning peak period. No cell containing three percentages has a single value far exceeding the others. The higher percentages are for interchanges that typically involve few trips. The high percentages in this table involve trips enter- ing Atlantic Station’s coffee shop from the grocery and to the grocery from the office building. Both of these types of trips would be for convenience and would possibly go to the clos- est location available, although more specialized needs might require trips to/from external locations. Table 97 shows internal capture percentages for entering trips during the P.M. peak period. Many cells show consis- tent percentages or a range with values spread throughout. A few cells show three or more percentages and a single value much higher than others. These are to office from retail (Atlantic Station); from residential to office (Legacy Town Center); and from restaurant to retail (Mockingbird 81 Origin Land Use Destination Land Use MXD Site Office Retail Restaurant Residential Cinema Hotel Atlantic Station 4 1 0 — 3 Legacy Town Center 3 9 3 — 0 Mockingbird Station — 14 1 — — Boca Del Mar — — — — — Country Isles — — — — — Office Village Commons — — — — — Atlantic Station 32 3 5 — 4 Legacy Town Center 0 8 17 — 0 Mockingbird Station — — — — — Boca Del Mar — — — — — Country Isles — — — — — Retail Village Commons — — — — — Atlantic Station 21 50 0 — 6 Legacy Town Center 1 1 18 — 6 Mockingbird Station 23 — 20 — — Boca Del Mar — — — — — Country Isles — — — — — Restaurant Village Commons — — — — — Atlantic Station 0 0 0 — 0 Legacy Town Center 0 2 4 — 0 Mockingbird Station 0 — 5 — — Boca Del Mar — — — — — Country Isles — — — — — Residential Village Commons — — — — — Atlantic Station — — — — — Legacy Town Center — — — — — Mockingbird Station — — — — — Boca Del Mar — — — — — Country Isles — — — — — Cinema Village Commons — — — — — Atlantic Station 0 0 4 0 — Legacy Town Center 0 0 3 0 — Mockingbird Station — — — — — Boca Del Mar — — — — — Country Isles — — — — — Hotel Village Commons — — — — — Table 96. Unconstrained internal capture rates for entering trips, all sites— A.M. peak period.

Station). Atlantic Station has almost twice as much retail as any of the other six developments; it has the only depart- ment store and that store is immediately next to the office building. That may explain the relatively higher portion of P.M. trips entering office from retail. Very few trips enter office buildings during the P.M. peak period. For those that do, it is not surprising that most trips from residential would begin close by in Legacy Town Center. More distant trips from home would be expected to wait until the next day. Because of the compactness of Mockingbird Station, many people were seen leaving restaurants during happy hour and strolling along the fronts of stores and entering a few to look at what was being sold. Atlantic Station and Legacy Town Center are somewhat less compact although they offer a similar opportunity. The Legacy Town Center area had more off-site shopping opportunities nearby than did Mockingbird Station or Atlantic Station. Hence, the dif- ferences shown in Table 97 appear logical. The three Florida developments are far less compact and would be expected to have less of this activity. 82 Origin Land Use Destination Land Use MXD Site Office Retail Restaurant Residential Cinema Hotel Atlantic Station 31 8 0 6 0 Legacy Town Center 6 30 57 0 0 Mockingbird Station 5 19 2 0 — Boca Del Mar 0 0 0 — — Country Isles 2 — 0 — — Office Village Commons 0 0 0 — — Atlantic Station 3 28 2 4 1 Legacy Town Center 0 17 10 1 2 Mockingbird Station 5 50 9 3 — Boca Del Mar 0 1 2 — — Country Isles 8 0 3 — — Retail Village Commons 3 1 3 — — Atlantic Station 2 29 1 2 5 Legacy Town Center 0 12 14 2 5 Mockingbird Station 1 16 2 3 — Boca Del Mar 0 1 2 — — Country Isles 0 27 3 — — Restaurant Village Commons 0 0 4 — — Atlantic Station 1 46 6 4 0 Legacy Town Center 1 15 16 0 0 Mockingbird Station 3 19 10 4 — Boca Del Mar 0 32 4 — — Country Isles 0 23 — — — Residential Village Commons 4 30 2 — — Atlantic Station 1 26 25 0 0 Legacy Town Center 0 0 32 0 0 Mockingbird Station 1 14 7 0 — Boca Del Mar — — — — — Country Isles — — — — — Cinema Village Commons — — — — — Atlantic Station 0 17 71 5 0 Legacy Town Center 0 13 10 12 1 Mockingbird Station — — — — — Boca Del Mar — — — — — Country Isles — — — — — Hotel Village Commons — — — — — Table 97. Unconstrained internal capture rates for entering trips, all sites— P.M. peak period.

for example, for Mockingbird Station, 63% of the trips from office goes to internal restaurants, while at Atlantic Station and Legacy Town Center, the corresponding trips amounted to 5% and 8%, respectively. At Mockingbird Station there was a specialty coffee restaurant less than 200 ft from the office building. No other restaurants were open in the devel- opment in the morning. The office building had no internal coffee or snack shop. There was a steady stream of people going between the office building and that restaurant during the morning peak period. The other two developments each had similar restaurants; however, they had roughly 3 to 5 times the office space, they were several blocks away, and Atlantic Station’s office building had a coffee stand in its lobby dur- ing the morning peak. It is no surprise that the Mockingbird Station capture rate is much higher than that for the other two developments. Table 98 contains a comparison of land use development unit ratios and internal capture rates for land use pairs in Tables 94 through 97 for which one internal capture percent- age is much higher than the other two. Table 98 displays rel- ative constraints on trips in the form of the ratio of develop- ment units, which is somewhat of a surrogate for total trips made. For exiting trips, the higher the ratio of origin devel- opment units to development units, the constraint is greater (i.e., the fewer the development units at the destination end). For entering trips, the lower the ratio, the constraint is greater (i.e., the fewer the development units) at the origin end of the trip. Hence, a high internal capture percentage for exiting trips could be expected where there is a low development unit ratio and proximity. For example, for P.M. peak-period trips from retail to residential at Legacy Town Center, 196,000 sq ft of retail were feeding trips to 1,360 units of residential (144 sq ft of retail per dwelling unit) at an average distance of 1,240 ft, resulting in 26% internal trip capture. On the other hand, for the same interchange, Country Isles had 109,000 sq ft of retail feeding 368 residential units (296 sq ft of retail per residential unit) at an average distance of 1,525 ft, resulting in 4% inter- nal capture. The origin end trips were more constrained at the destination end at Country Isles (about half as many units receiving trips). In addition, the separation was greater, further constraining trips from retail to residential. Such is the case for most of the examples shown in Table 98. Except for P.M. peak-period entering trips, constraints imposed by development unit ratios and greater proximity distances result in the lower internal capture percentages. Conclusions As Table 98 shows, lower ratios and higher proximity tend to result in higher capture rates. The highest percentages of internal capture are associated with lesser levels of con- straint and higher proximity. Only additional data will confirm 83 Figure 19. Example of unconstrained demand and balancing of internal trip interchange. Land Use A Land Use B 75 50 35 25 35 25 Figure 20. Continuation of Figure 19 example showing external trips resulting from internal constraints. Land Use A Land Use B 75 50 35 25 35 2540 025 0 Internal capture percentages in the previous tables are sim- ilar for some land use pairs but not for others. One reason for this is the balance between land uses. If two land uses are bal- anced for the purpose of trip generation interaction, Land Use A would want to send as many people to Land Use B as Land Use B would want to receive from Land Use A. However, consider a case where Land Use A wants to send 75 trips to Land Use B, but Land Use B only wants to receive 35 trips from Land Use A (see Figure 19). Land Use B will receive all the Land Use A trips it wants: it can be considered to be unconstrained. There are more than enough Land Use A trips to satisfy Land Use B demand; however, Land Use A demand to send trips to Land Use B is constrained because Land Use B will accept only half of the trips Land Use A wants to send. Examining the opposite direction, Land Use B wants to send 25 trips to Land Use A, and Land Use A wants to receive 50 trips from Land Use B. Because all of the Land Use B trips can be accepted by Land Use A, Land Use B is uncon- strained in that direction, but Land Use A is constrained. When internal trips are constrained, they cannot occur, and the travel demand must be satisfied externally. Figure 20 shows what happens in these two examples. Land Use A wants to send 75 trips to Land Use B. Only 35 of those trips can go to Land Use B internal to the development, so the other 40 trips have to seek Land Use B externally. This is based on the assumption that people make trips for a purpose (e.g., eat lunch), and if that purpose cannot be satisfied internally where it is most convenient, the trip maker will have to find someplace to eat externally. Therefore, returning to Table 94, the (major) differences between the internal trip capture percentages that appear in this table are attributable in many cases to the balance, lack of constraints, or other factors that exist for some zone pairs—

84 Land Use Development Units2 Origin Destination Site 1 Origin Destination Ratio 3 Proximity (ft)4 Internal Capture (%) Comments Morning Peak-Period Exiting Trips (from Table 94) AS 551 1.6 344 1,000 5% Proximity similar to LTC; most constrained at destination. LTC 311 5.1 61 1,200 8% Proximity similar to AS; slightly more constrained than MS. Office Restaurant MS 115 1.5 77 200 63% Closest proximity; close to least constrained at destination. AS 798 1.6 499 2,300 0% Farthest separation; most constrained at destination. LTC 1,360 5.1 267 1,470 7% Moderate proximity; somewhat constrained at destination. Residen- tial Restaurant MS 191 1.5 127 100 20% Closest proximity; least constrained at destination. Afternoon Peak-Period Exiting Trips (from Table 95) AS 551 435 1.29 660 6% About middle for both proximity and constraint at destination. LTC 311 196 1.59 975 0% Third longest separation; third most constrained at destination. MS 115 156 0.74 320 9% Close to least constrained at destination; closest proximity. BDM 316 151 2.09 1,125 0% Second longest separation; second most constrained at destination. CI 75 109 0.69 775 9% Least constrained at destination; third closest proximity. Office Retail VC 315 121 2.60 1,600 6% Most constrained at destination; longest separation. AS 435 798 0.55 2,280 13% Second most constrained at destination; longest separation. LTC 196 1,360 0.14 1,240 26% Least constrained at destination; third longest proximity. MS 156 191 0.82 170 7% Most constrained at destination; closest proximity. BDM 151 1,144 0.13 825 3% Least constrained at destination; second longest separation. CI 109 368 0.30 1,525 4% Moderate constraint and proximity. Retail Residential VC 121 317 0.38 900 7% Most constrained at destination; moderate proximity. AS 65 798 .081 2,360 3% Third least constrained at destination; longest separation. LTC 69 1,360 .051 1,325 18% Third least constrained at destination; third longest separation. MS 29 191 .152 200 3% Most constrained at destination; closest proximity. BDM 34 1,144 .030 1,100 6% Least constrained at destination; fourth longest separation. CI 21 368 .057 1,600 — Second longest separation; second most constrained at destination. Restau- rant Residential VC 42 317 .132 600 4% Second most constrained at destination; second closest. proximity. Morning Peak-Period Entering Trips (from Table 96)—no instances of one internal capture percentage much higher than at least two others Afternoon Peak-Period Entering Trips (from Table 97)—no instances of one internal capture percentage much higher than at least two others AS 435 551 0.79 895 31% Third least constrained at origin; second closest proximity. LTC 196 311 0.63 975 6% Third most constrained at origin; third longest separation. MS 156 115 1.36 150 5% Close to least constrained at origin; closest proximity. BDM 151 316 0.48 1,125 0% Second most constrained at origin; second longest separation. CI 109 75 1.45 775 2% Least constrained at origin. Retail Office VC 121 315 0.38 1,600 0% Most constrained at origin; longest separation. AS 798 551 1.45 3,100 0% Second most constrained at origin; farthest separated. LTC 1,360 311 4.37 900 57% Close to least constrained at origin, second closest proximity. MS 191 115 1.66 225 2% Third most constrained at origin; closest proximity. BDM 1,144 316 3.62 2,000 0% Third least constrained at origin; second farthest separated. CI 368 75 4.91 1,000 0% Least constrained at origin; third closest proximity. Residen- tial Office VC 317 315 1.01 1,750 0% Most constrained at origin; third farthest separated. Table 98. Comparison of internal capture by development unit ratios and proximities for selected land use pairs.

whether other similarly unconstrained and high proximity examples will demonstrate similar internal capture findings. Where the highest internal capture percentage accompanies both the least constrained and highest proximity, the reported internal capture percentage is probably close to the maximum the researchers would expect to find. These percentages could be considered unconstrained internal capture percentages. However, where the highest internal capture percentage for a land use pair and period is associated with either a moderately high constraint and/or a proximity significantly farther than the minimum, the researchers expect that future surveys could find higher internal capture percentages. For now, the highest internal capture percentages reported in this report for each land use pair and time period should be considered the docu- mented unconstrained internal capture percentages and should be used as unconstrained values. In general, the three developments surveyed for this project are more compact, are better connected, and have more com- ponent land uses than do the three Florida developments. Additionally, the three developments surveyed for this proj- ect generally have higher internal capture percentages. This confirms—at least based on the available data—that internal capture can be increased through the use of more interacting land uses, better connectivity, and/or more compactness. Com- pactness or proximity is addressed later in this section. Unconstrained internal capture between individual land uses ranges from a low of none found to highs of over 60%. The comparisons also show a wide range of internal capture rates between land use pairs. This results from a number of factors, the most important and projectable of which (at time of zon- ing) is the balance between land uses within a development. To demonstrate this phenomenon, consider an office building with 20 employees who want to go out for lunch at an onsite restaurant. The restaurant has eight seats. If all employees want to go there, only eight can be seated. The restaurant seat- ing constrains the interaction between the two land uses. Now compare that development to the next similar devel- opment down the street where 25 office employees want to go to a restaurant with 16 seats. At that location, as many as 16 employees can go to that restaurant, so even though the restaurant is again the constraint, the interaction is greater. For the third example, consider that 25 office employees can go to an onsite restaurant with 40 seats. In this example, all 25 employees can be seated. In fact, more could be seated. In this example, the office building is the constraint. Hence, with differing balances of the land uses making up the six surveyed developments, it is understandable that the internal trip capture percentages vary among them. Some of the differences may be explained by the travel distances between trip origins and destinations—that is, proximity. Proximity is addressed in a later section. In addition, there are other factors not quantified in this research that may also affect internal capture such as attractiveness of specific busi- nesses, demographics of trip-makers, and alternative oppor- tunities for similar destinations at nearby developments (i.e., competing opportunities). While these may influence inter- nal trip capture, they may not be known at the time a devel- opment is proposed, so it would be difficult to project those characteristics even if a method of projection was available. Tables 99 through 102 show the highest values from Tables 94 through 97. The values of Tables 99 through 102 show how much internal capture was achieved by the best balances between interacting land uses. In terms of the office/ restaurant example described previously, the values of Tables 99 through 102 demonstrated the most unconstrained individ- ual conditions observed at the six developments. Although it is very possible that MXDs with other balances of development may experience even higher percentages, at 85 Land Use Development Units2 Origin Destination Site 1 Origin Destination Ratio 3 Proximity (ft)4 Internal Capture (%) Comments AS 65 435 0.15 430 28% Most constrained at origin; second closest proximity. LTC 69 196 0.35 500 17% Least constrained at origin. MS 29 156 0.19 300 50% Second most constrained at origin; closest proximity. BDM 34 151 0.23 800 1% Third most constrained at origin; third longest proximity. CI 21 109 0.19 1,200 0% Second most constrained at origin; second longest separation. Restau- rant Retail VC 42 121 0.35 1,100 1% Least constrained at origin; second longest separation. 1 AS = Atlantic Station; LTC = Legacy Town Center; MS = Mockingbird Station; BDM = Boca del Mar; CI = Country Isles; VC = Village Commons. 2 All development units are in gross sq ft except residential, which is in dwelling units; development units shown for restaurant during morning peak are for those restaurants that were open. 3 Origin development units/destination development units. For exiting trips, this constraint at destination end is represented by highest ratio. For entering trips, highest constraint at origin is represented by lowest ratio. 4 Separation between interaction land uses based on average weighted by trips (rounded to closest 100 ft). Table 98. (Continued).

86 Destination Land Use1Origin Land Use Office Retail Restaurant Residential Cinema Hotel Office N/A 28% 63% 1% N/A 0% Retail 29% N/A 13% 14% N/A 0% Restaurant 31% 14% N/A 4% N/A 3% Residential 2% 1% 20% N/A N/A 0% Cinema N/A N/A N/A N/A N/A N/A Hotel 75% 14% 9% 0% N/A N/A 1 Corresponds to ITE Trip Generation Handbook Table 7.1; N/A signifies no data or interchanges within same land use categories that are accounted for within ITE trip generation rates. Table 99. Proposed unconstrained values for percent distribution of internal trip destinations for exiting trips—A.M. peak period. Destination Land Use1Origin Land Use Office Retail Restaurant Residential Cinema Hotel Office N/A 20% 4% 2% 0% 0% Retail 2% N/A 29% 26% 4% 5% Restaurant 3% 41% N/A 18% 8% 7% Residential 4% 42% 21% N/A 0% 3% Cinema 2% 21% 31% 8% N/A 2% Hotel 0% 16% 68% 2% 0% N/A 1 Corresponds to ITE Trip Generation Handbook Table 7.1; N/A signifies no data or interchanges within same land use categories that are accounted for within ITE trip generation rates. Table 100. Proposed unconstrained values for percent distribution of internal trip destinations for exiting trips—P.M. peak period. Destination Land Use1Origin Land Use Office Retail Restaurant Residential Cinema Hotel Office N/A 32% 23% 0% N/A 0% Retail 4% N/A 50% 2% N/A 0% Restaurant 14% 8% N/A 5% N/A 4% Residential 3% 17% 20% N/A N/A 0% Cinema N/A N/A N/A N/A N/A N/A Hotel 3% 4% 6% 0% N/A N/A 1 Corresponds to ITE Trip Generation Handbook Table 7.2; N/A signifies no data or interchanges within same land use categories that are accounted for within ITE trip generation rates. Table 101. Proposed unconstrained values for percent distribution of internal trip origins for entering trips—A.M. peak period. Destination Land Use 1 Origin Land Use Office Retail Restaurant Residential Cinema Hotel Office N/A 8% 2% 4% 1% 0% Retail 31% N/A 29% 46% 26% 17% Restaurant 30% 50% N/A 16% 32% 71% Residential 57% 10% 14% N/A 0% 12% Cinema 6% 4% 3% 4% N/A 1% Hotel 0% 2% 5% 0% 0% N/A 1 Corresponds to ITE Trip Generation Handbook Table 7.2; N/A signifies no data or interchanges within same land use categories that are accounted for within ITE trip generation rates. Table 102. Proposed unconstrained values for percent distribution of internal trip origins for entering trips—P.M. peak period.

this point the researchers had no evidence to verify the pos- sibility of higher percentages. Hence, for the purpose of this research project, the researchers concluded that a conserva- tive approach is to use the values of Tables 99 through 102 in the proposed estimation process developed in this project. At a future time, if subsequent surveys using similar procedures show even higher percentages, those results could be incor- porated into Tables 99 through 102. Proximity Effects Data collected in the pilot study and Florida surveys pro- vided the basis for evaluating proximity effects on internal capture. This analysis was performed to test the hypothesis that travel distance between locations of interacting land uses would affect the degree of interchange between those land uses. The analyses used surveyed interchanges and walking dis- tances between origin and destination. All three pilot study developments had similar pedestrian environments—out- door sidewalks adjacent to buildings, mostly along internal two-lane streets or parking lots. Most sidewalks are land- scaped with trees, although in Mockingbird Station some sidewalks have no trees. In most cases, the sidewalks are at least 10-ft wide in commercial areas. No sidewalk was con- sidered too narrow for people to walk or pass conveniently. Mockingbird Station has an elevator and one main set and two supplemental sets of stairs between the ground and second levels serving the cinema, a few restaurants, and the DART rail station and bus transfer center. Although the elevator and stairways undoubtedly impede some people in Mockingbird Station, so few destinations required using the stairs that they were not considered further. The three Florida sites are more spread out with most of the land uses in pods. Each pod is conveniently walkable within. Many of the pods are not interconnected by sidewalks, but are accessible by walking or driving across parking aisles or lots. However, each of these development pods is clearly designed to encourage internal interaction among land use activities. Data collected and compiled as part of this project provide an indication of the effect of land use proximity on internal capture. It was observed that as distance increases, the level of interaction (i.e., the internal capture) declines. To quantify this relationship, internal capture rates derived from intercept sur- veys were plotted against proximity of pairs of land uses. All land use pairs for the three newly surveyed pilot study sites and the three Florida sites surveyed in the mid-1990s were plotted. To illustrate this concept, the top chart in Figure 21 shows the internal capture observed at the six mixed-use sites for trips from retail/restaurant uses to residential uses. The bot- tom chart shows internal capture observed to residential from retail/restaurant. In the charts, each plot point represents a single mixed-use site. From left to right (i.e., closest to far- thest), the points represent Mockingbird Station, Legacy Town Center, Boca del Mar, Village Commons, Country Isles, and Atlantic Station. A key premise about internal capture is that for a trip from one land use to another at a mixed-use site, one direction of travel must be unconstrained (in terms of internal capture) and the other must be constrained. In some instances, the internal capture rates in both directions of travel are in per- fect balance and are, therefore, both constrained. In Figure 21, the presumed unconstrained direction is des- ignated as a large dot and the presumed constrained direction as a small dot. If a site is constrained in the top chart, it must be unconstrained in the bottom chart; if a site is constrained in the bottom chart, it must be unconstrained in the top chart. Each site must have an unconstrained value in one direction or the other. In addition, the unconstrained internal capture values should exceed the constrained values on each individual chart. In the top chart, internal capture values at the unconstrained sites decrease from around 16% at a proximity of 700 ft to around 5% at a proximity of 2,200 ft. In the bottom chart, internal capture of greater than 50% (at a proximity of 200 ft) decreases to about 20% at a proximity of 1,500 ft. For many land use pairs, the database consists of only three data points representing unconstrained internal capture— two in one direction and one in the other. It is difficult to reach definitive conclusions about the effect of land use proximity on internal capture with so little data. To improve the likeli- hood of defining a reliable relationship between proximity and internal capture, data for various land use pairings with poten- tially common characteristics were grouped and examined. For example, trips to or from retail might have the same proximity- capture characteristics as trips to or from restaurants. As a result of that analysis, two proximity relationships were iden- tified, as Figure 22 shows. Each point in the figure represents a measured uncon- strained internal capture rate for a particular pair of land uses at a single mixed-use site. The x-axis in the figure is the proximity distance. The y-axis is normalized to represent the percent of the highest unconstrained value for the particular land use pair. The square-shaped dots in the figure represent the proxim- ity and internal capture values for all land use pairs with res- idential as the destination, for the origin end of the trip. In other words, these are a combination of the rates • From office to residential, • From retail to residential, • From restaurant to residential, • From hotel to residential, and • From cinema to residential. 87

88 to Residential from Retail/Restaurant 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 0 500 1000 1500 2000 2500 Distance (feet) Unconstrained Constrained from Retail/Restaurant to Residential 0.0% 2.0% 4.0% 6.0% 8.0% 10.0% 12.0% 14.0% 16.0% 18.0% 0 500 1000 1500 2000 2500 Distance (feet) Unconstrained Constrained Figure 21. Example of relationship between internal capture percentage in unconstrained and constrained directions (between residential and retail/ restaurant land use pair).

The best-fit curve for these data points is shown in the fig- ure as Adjustment #1. The R-square for the curve is 0.58. This curve is used in the estimation procedure described later in this chapter to account for land use pair proximity adjustments to unconstrained internal rates. The triangular-shaped dots in the figure represent the prox- imity and internal capture values for all land use pairs with either office or residential as the origin and retail or restaurant as the destination, for both the origin and destination ends of the trip. In other words, these are the rates • From office to retail, • From office to restaurant, • From residential to retail, • From residential to restaurant, • To retail from office, • To restaurant from office, • To retail from residential, and • To restaurant from residential. The best-fit curve for these data points is shown in Figure 22 as Adjustment #2. The curve is actually two straight lines that intersect at a proximity distance of 1,524 ft. The R-square for the less-than-1,524-ft curve is 0.50. The data at longer dis- tances does not track with that equation and was grouped to create a second intersecting line; that line connected from the extreme end points of the upper line to the midpoint between the two points for the longest proximity distance in this data subset. The best-fit curve equations in Figure 22 intersect the x-axis at proximity distances above which there would presumably be no internal capture. However, at the study sites, internal capture was measured between land uses at the extreme lim- its of all six mixed-use sites where data were collected. To account for this assumed synergy between land uses no mat- ter how far apart as long as they are both within the mixed-use center, both proximity adjustment lines in the figure are ter- minated at an arbitrary minimum y-axis value of 0.10 (i.e., at 10% of the unconstrained values). This then leaves a minimal internal capture percentage at long distances. Note that these proximity adjustment relationships repre- sent only a fraction of all potential land use pairs (only 13 prox- imity adjustment factors out of a total of 60 directional internal capture rates for the 6 land uses). For the remaining 47 land use pairs, a definitive relationship between proximity and internal capture rate could not be established with the available data. 89 Figure 22. Relationship between proximity and unconstrained internal capture percentage for P.M. peak-period trips for land use pairs and directions with confirmed proximity effects.

Use of Figure 22 requires information from a site plan show- ing different land uses. During early stages of development planning, it is unlikely that such a layout will always be avail- able; however, a development site will have been defined. Hence, there is a need to be able to estimate travel distances for internal trips based on knowing only site size. There is no end to the variety of potential site layouts for a given parcel of land, but reasonable assumptions can be made. In the case of internal trip capture estimation, it is prudent to err on the conservative side—that is, to under- estimate internal capture rather than to overestimate capture. A few basic assumptions to arrive at a maximum travel dis- tance can be applied. These could include a distance from the property boundary to the building doors and the internal block configuration. Figure 23 shows the site size and average separation dis- tances between interacting land uses for Mockingbird Sta- tion, Atlantic Station, and Legacy Town Center. The straight line relationship may be coincidental since there are differ- ences in the site configurations and layouts of the component land uses. Figure 23 may provide a basis for estimating sepa- ration distances if there is no site plan or conceptual land use plan available when an analysis is performed, but this should be validated and refined in further studies. Procedure for Estimating Internal Capture at a Proposed MXD The estimation procedure developed in this project is essentially an extension and enhancement of the current ITE method documented in the ITE Trip Generation Handbook, 2nd edition (1). The recommended method enriches the ITE method in the following manner: • Adds an A.M. peak-hour period to the existing P.M. peak- hour period; • Adds three land uses—restaurant, hotel, and cinema—to the existing office, retail, and residential uses; • Expands the basis for the A.M. and P.M. peak-hour internal capture factors from three developments in one state to six developments in three states and also broadens the types of MXDs included in the database; and • Adds a proximity adjustment for some land use pairs. Midday and daily periods, which are included in the ITE Trip Generation Handbook, were not addressed since those periods are rarely used in typical TIS and would have increased the data collection cost beyond the available resources. The recommended estimation method consists of the following basic steps: 1. Determine whether the methodology is appropriate for the development to be analyzed. 2. Define the pertinent site and development characteristics. 3. Estimate single-use trip generation for each component land use using ITE or other acceptable source; convert to person trips. 4. Use unconstrained internal capture percentages devel- oped in this project to estimate the number of potential internal trips between each pair of land uses. Include an adjustment for proximity (also developed in this project). 5. Balance internal trips generated at both ends of each inter- acting pair (i.e., internal trips coming from the origin end 90 Figure 23. Relationship between average internal travel distances and site size (Mockingbird Station, Atlantic Station, and Legacy Town Center).

need to be the same as those coming to the destination end); adapt the existing balancing procedure contained in the ITE Trip Generation Handbook (1). 6. Subtract the estimated internal trips from the total trip generation to estimate external trips for the MXD being analyzed; convert to vehicle trips as needed. The user of this estimation methodology is cautioned that each MXD has unique characteristics that influence the extent of internal trip capture. Such characteristics include, but are not limited to the following: • The number and magnitudes of complementary land uses; • The layout of the land uses relative to each other; • Specific businesses, residence types, and other component characteristics within each land use category; • Proximity and connectivity between each pair of land uses; • Design characteristics of the development and its internal transportation system; • Specific characteristics of the development’s access and parking; and • Competing opportunities outside the development. The user is further cautioned that estimates of internal cap- ture for trips between specific pairs of land uses are based on data collected for between one and six surveyed develop- ments. Clearly, additional data on internal capture at existing MXDs would help improve confidence in the accuracy of the internal capture estimates and might result in different inter- nal capture rates. The researchers believe that the successful but limited val- idations conducted for this estimation method do confirm that the results provide accurate approximations of external trip generation for typical MXDs consisting of typical office, retail, restaurant, residential, cinema, and hotel land uses. The researchers also believe these approximations are con- sistent with the accuracy of trip generation estimates for single-use developments as portrayed in such references as Trip Generation, 8th edition (2). The researchers also believe this methodology provides an advancement and improve- ment over a similar method described in Trip Generation Handbook, 2nd edition (1). The estimation procedure is presented step-by-step in the order it would be performed by the analyst: • Step 1: Determine whether the methodology is appropri- ate for your application. • Step 2: Define the pertinent site characteristics. • Step 3: Calculate single-use trip generation for the site components. • Step 4: Estimate the unconstrained internal capture rates for all land use pairs at the site and add adjustments for proximity. • Step 5: Calculate the balanced internal trips between all land use pairs. • Step 6: Calculate the overall internal capture rate for the site. One product of this procedure is an estimate of internal trip capture between pairs of land uses in the development for which internal capture data exist. Users of this estimation procedure are encouraged to carefully and completely read earlier parts of this chapter to understand the background and data supporting this procedure: • The internal capture estimation methodology and its logic, • Descriptions of the six developments from which the data behind the estimation methodology were collected, • Survey findings from the six development sites, and • The following instructions for use of the estimation methodology. If the analyst understands the concept of “internal cap- ture balancing” as described earlier in this chapter, these basic instructions should suffice. At the end of this chapter, addi- tional guidance is provided for the analyst who understands the concepts, but who is unsure of the mechanics of a specific step. Additional guidance is also provided for the analyst who thoroughly understands the basic concept and its data limita- tions, appreciates the uniqueness of each mixed-use site, and is interested in investigating the potential internal capture impacts of the nuances of a particular site. Tables 103 through 106 show an automated spreadsheet tool that can be used to compute internal capture and exter- nal trip generation for MXDs. The entire workbook consists of six separate worksheets in two sets—one for weekday A.M. street peak-hour estimates and one for weekday P.M. street peak-hour estimates. This description covers the A.M. street peak hour only. The six worksheet and tables in which the A.M. sheets are shown are • Table 103 (Worksheet 1): Estimator Input/Output Work- sheet—A.M. Street Peak Hour; • Not shown here (Worksheet 2): Estimator Input/Output Worksheet—P.M. Street Peak Hour; • Table 104 (Worksheet 3): Estimator Intermediate Calcu- lations—A.M. Street Peak Hour; • Not shown here (Worksheet 4): Estimator Intermediate Calculations—P.M. Street Peak Hour; • Table 105 (Worksheet 5): Estimator Updated ITE Trip Gen- eration Handbook Table 7.1 With Proximity Adjustment (1, Ch. 7); and • Table 106 (Worksheet 6): Estimator Updated ITE Trip Generation Handbook Table 7.2 With Proximity Adjust- ment (1, Ch. 7). 91

92 Project Name: Organization: Project Location: Performed By: Scenario Description: Date: Analysis Year: Checked By: Analysis Period: Date: ITE LUCs1 Quantity Units Total Entering Exiting Office 0 Retail 0 Restaurant 0 Cinema/Entertainment 0 Residential 0 Hotel 0 All Other Land Uses2 0 Total 0 0 0 Veh. Occ. % Transit % Non-Motorized Veh. Occ. % Transit % Non-Motorized Office Retail Restaurant Cinema/Entertainment Residential Hotel All Other Land Uses2 Office Retail Restaurant Residential Hotel Office Retail Restaurant Cinema/Entertainment Residential Hotel Office Retail Restaurant Residential Hotel Office 0 0 0 0 Retail 0 0 0 0 Restaurant 0 0 0 0 Cinema/Entertainment 0 0 0 0 0 Residential 0 0 0 0 Hotel 0 0 0 0 Total Entering Exiting Land Use Entering Trips Exiting Trips All Person-Trips Office Internal Capture Percentage Retail Restaurant External Vehicle-Trips3 Cinema/Entertainment External Transit-Trips4 Residential External Non-Motorized Trips4 Hotel 2Total estimate for all other land uses at mixed-use development site-not subject to internal trip capture computations in this estimator 3Vehicle-trips computed using the mode split and vehicle occupancy values provided in Table 2-A 1Land Use Codes (LUCs) from Trip Generation Informational Report , published by the Institute of Transportation Engineers. 4Person-Trips Estimator Tool Developed by the Texas Transportation Institute for NCHRP Project 8-51 Table 5-A: Computations Summary Table 6-A: Internal Trip Capture Percentages by Land Use NCHRP 8-51 Internal Trip Capture Estimator Table 1-A: Base Vehicle-Trip Generation Estimates (Single-Use Site Estimate) 0 0 Cinema/Entertainment Destination (To) Estimated Vehicle-Trips Land Use Origin (From) Origin (From) Destination (To) 0 0 0 Table 4-A: Internal Person-Trip Origin-Destination Matrix AM Street Peak Hour Cinema/Entertainment Development Data (For Information Only ) Table 2-A: Mode Split and Vehicle Occupancy Estimates Land Use Entering Trips Exiting Trips Table 3-A: Average Land Use Interchange Distances (Feet Walking Distance) Table 103. Estimator input/output worksheet—A.M. street peak hour (A.M. sheet 1 of 4).

93 Project Name: Analysis Period: Veh. Occ. Vehicle-Trips Person-Trips Veh. Occ. Vehicle-Trips Person-Trips Office Retail Restaurant Cinema/Entertainment Residential Hotel Office Retail Restaurant Residential Hotel Office 0 0 0 0 Retail 0 0 0 0 Restaurant 0 0 0 0 Cinema/Entertainment 0 0 0 0 0 Residential 0 0 0 0 Hotel 0 0 0 0 Office Retail Restaurant Residential Hotel Office 0 0 0 0 Retail 0 0 0 0 Restaurant 0 0 0 0 Cinema/Entertainment 0 0 0 0 0 Residential 0 0 0 0 Hotel 0 0 0 0 Internal External Total Vehicles1 Transit2 Non-Motorized2 Office Retail Restaurant Cinema/Entertainment Residential Hotel All Other Land Uses3 Internal External Total Vehicles1 Transit2 Non-Motorized2 Office Retail Restaurant Cinema/Entertainment Residential Hotel All Other Land Uses3 0 0 0 External Trips by Mode 0 0 0 3Total estimate for all other land uses at mixed-use development site-not subject to internal trip capture computations in this estimator Person-Trip Estimates Destination Land Use Table 9-A (O): Internal and External Trips Summary (Exiting Trips) Origin Land Use Person-Trip Estimates External Trips by Mode 2Person-Trips 1Vehicle-trips computed using the mode split and vehicle occupancy values provided in Table 2-A Table 7-A (O): Exiting Trips 0 0 Table 8-A (D): Internal Person-Trip Origin-Destination Matrix (Computed at Destination) Origin (From) Destination (To)Cinema/Entertainment 0 0 AM Street Peak Hour Table 9-A (D): Internal and External Trips Summary (Entering Trips) Table 8-A (O): Internal Person-Trip Origin-Destination Matrix (Computed at Origin) Origin (From) Destination (To)Cinema/Entertainment Table 7-A: Conversion of Vehicle-Trip Ends to Person-Trip Ends Land Use Table 7-A (D): Entering Trips Table 104. Estimator intermediate calculations—A.M. street peak hour (A.M. sheet 2 of 4).

AM Peak Hour PM Peak Hour AM PM AM Peak Hour PM Peak Hour To Office 0% 0% 1.000 1.000 To Office 0.0% 0.0% To Retail 28% 20% 1.000 1.000 To Retail 28.0% 20.0% To Restaurant 63% 4% 1.000 1.000 To Restaurant 63.0% 4.0% To Cinema/Entertainment 0% 0% 1.000 1.000 To Cinema/Entertainment 0.0% 0.0% To Residential 1% 2% 1.000 1.000 To Residential 1.0% 2.0% To Hotel 0% 0% 1.000 1.000 To Hotel 0.0% 0.0% To Office 29% 2% 1.000 1.000 To Office 29.0% 2.0% To Retail 0% 0% 1.000 1.000 To Retail 0.0% 0.0% To Restaurant 13% 29% 1.000 1.000 To Restaurant 13.0% 29.0% To Cinema/Entertainment 0% 4% 1.000 1.000 To Cinema/Entertainment 0.0% 4.0% To Residential 14% 26% 1.000 1.000 To Residential 14.0% 26.0% To Hotel 0% 5% 1.000 1.000 To Hotel 0.0% 5.0% To Office 31% 3% 1.000 1.000 To Office 31.0% 3.0% To Retail 14% 41% 1.000 1.000 To Retail 14.0% 41.0% To Restaurant 0% 0% 1.000 1.000 To Restaurant 0.0% 0.0% To Cinema/Entertainment 0% 8% 1.000 1.000 To Cinema/Entertainment 0.0% 8.0% To Residential 4% 18% 1.000 1.000 To Residential 4.0% 18.0% To Hotel 3% 7% 1.000 1.000 To Hotel 3.0% 7.0% To Office 0% 2% 1.000 1.000 To Office 0.0% 2.0% To Retail 0% 21% 1.000 1.000 To Retail 0.0% 21.0% To Restaurant 0% 31% 1.000 1.000 To Restaurant 0.0% 31.0% To Cinema/Entertainment 0% 0% 1.000 1.000 To Cinema/Entertainment 0.0% 0.0% To Residential 0% 8% 1.000 1.000 To Residential 0.0% 8.0% To Hotel 0% 2% 1.000 1.000 To Hotel 0.0% 2.0% To Office 2% 4% 1.000 1.000 To Office 2.0% 4.0% To Retail 1% 42% 1.000 1.000 To Retail 1.0% 42.0% To Restaurant 20% 21% 1.000 1.000 To Restaurant 20.0% 21.0% To Cinema/Entertainment 0% 0% 1.000 1.000 To Cinema/Entertainment 0.0% 0.0% To Residential 0% 0% 1.000 1.000 To Residential 0.0% 0.0% To Hotel 0% 3% 1.000 1.000 To Hotel 0.0% 3.0% To Office 75% 0% 1.000 1.000 To Office 75.0% 0.0% To Retail 14% 16% 1.000 1.000 To Retail 14.0% 16.0% To Restaurant 9% 68% 1.000 1.000 To Restaurant 9.0% 68.0% To Cinema/Entertainment 0% 0% 1.000 1.000 To Cinema/Entertainment 0.0% 0.0% To Residential 0% 2% 1.000 1.000 To Residential 0.0% 2.0% To Hotel 0% 0% 1.000 1.000 To Hotel 0.0% 0.0% Proximity Adjustment Table 7.1 Unconstrained Internal Trip Capture Rates for Trip Origins within a Multi-Use Development From OFFICE From CINEMA/ ENTERTAINMENT From RESIDENTIAL From CINEMA/ ENTERTAINMENT From RESIDENTIAL From RETAIL WeekdayLand Use Pairs From RESTAURANT Table 7.1a Adjusted Internal Trip Capture Rates for Trip Origins within a Multi- Use Development Land Use Pairs Weekday From OFFICE From RETAIL From RESTAURANT From HOTEL From HOTEL Table 105. Estimator updated ITE Trip Generation Handbook (1, p. 93) Table 7.1 with proximity adjustment (sheet 3 of 4).

AM Peak Hour PM Peak Hour AM PM AM Peak Hour PM Peak Hour From Office 0% 0% 1.000 1.000 From Office 0.0% 0.0% From Retail 4% 31% 1.000 1.000 From Retail 4.0% 31.0% From Restaurant 14% 30% 1.000 1.000 From Restaurant 14.0% 30.0% From Cinema/Entertainment 0% 6% 1.000 1.000 From Cinema/Entertainment 0.0% 6.0% From Residential 3% 57% 1.000 1.000 From Residential 3.0% 57.0% From Hotel 3% 0% 1.000 1.000 From Hotel 3.0% 0.0% From Office 32% 8% 1.000 1.000 From Office 32.0% 8.0% From Retail 0% 0% 1.000 1.000 From Retail 0.0% 0.0% From Restaurant 8% 50% 1.000 1.000 From Restaurant 8.0% 50.0% From Cinema/Entertainment 0% 4% 1.000 1.000 From Cinema/Entertainment 0.0% 4.0% From Residential 17% 10% 1.000 1.000 From Residential 17.0% 10.0% From Hotel 4% 2% 1.000 1.000 From Hotel 4.0% 2.0% From Office 23% 2% 1.000 1.000 From Office 23.0% 2.0% From Retail 50% 29% 1.000 1.000 From Retail 50.0% 29.0% From Restaurant 0% 0% 1.000 1.000 From Restaurant 0.0% 0.0% From Cinema/Entertainment 0% 3% 1.000 1.000 From Cinema/Entertainment 0.0% 3.0% From Residential 20% 14% 1.000 1.000 From Residential 20.0% 14.0% From Hotel 6% 5% 1.000 1.000 From Hotel 6.0% 5.0% From Office 0% 1% 1.000 1.000 From Office 0.0% 1.0% From Retail 0% 26% 1.000 1.000 From Retail 0.0% 26.0% From Restaurant 0% 32% 1.000 1.000 From Restaurant 0.0% 32.0% From Cinema/Entertainment 0% 0% 1.000 1.000 From Cinema/Entertainment 0.0% 0.0% From Residential 0% 0% 1.000 1.000 From Residential 0.0% 0.0% From Hotel 0% 0% 1.000 1.000 From Hotel 0.0% 0.0% From Office 0% 4% 1.000 1.000 From Office 0.0% 4.0% From Retail 2% 46% 1.000 1.000 From Retail 2.0% 46.0% From Restaurant 5% 16% 1.000 1.000 From Restaurant 5.0% 16.0% From Cinema/Entertainment 0% 4% 1.000 1.000 From Cinema/Entertainment 0.0% 4.0% From Residential 0% 0% 1.000 1.000 From Residential 0.0% 0.0% From Hotel 0% 0% 1.000 1.000 From Hotel 0.0% 0.0% From Office 0% 0% 1.000 1.000 From Office 0.0% 0.0% From Retail 0% 17% 1.000 1.000 From Retail 0.0% 17.0% From Restaurant 4% 71% 1.000 1.000 From Restaurant 4.0% 71.0% From Cinema/Entertainment 0% 1% 1.000 1.000 From Cinema/Entertainment 0.0% 1.0% From Residential 0% 12% 1.000 1.000 From Residential 0.0% 12.0% From Hotel 0% 0% 1.000 1.000 From Hotel 0.0% 0.0% Proximity Adjustment To HOTEL Table 7.2a Adjusted Internal Trip Capture Rates for Trip Destinations within a Multi-Use Development Land Use Pairs Weekday To OFFICE To RETAIL To RESTAURANT To CINEMA/ ENTERTAINMENT To RESIDENTIAL To RESTAURANT To CINEMA/ ENTERTAINMENT To RESIDENTIAL To HOTEL Table 7.2 Unconstrained Internal Trip Capture Rates for Trip Destinations within a Multi-Use Development To OFFICE To RETAIL WeekdayLand Use Pairs Table 106. Estimator updated ITE Trip Generation Handbook Table 7.2 with proximity adjustment (sheet 4 of 4).

The following description of the computational process has been put into the form of these spreadsheets, and there is an accompanying Excel spreadsheet workbook to auto- mate the computations. It is intended that the spreadsheet be used to perform computations. The spreadsheet workbook is explained for the A.M. street peak hour and consists of the four spreadsheets shown in Tables 103 through 106. The P.M. street peak-hour estimate is prepared identically. The estimator— either automated or manual—was designed to have a cover sheet that contains all input and output of interest, with look- up data and intermediate computations on the subsequent spreadsheets. The description that follows uses the spreadsheet as an example. Step 1: Determine Whether the Methodology Is Appropriate for Your Application The procedure should only be used for estimating internal capture at an MXD that has characteristics resembling the sites from which the internal capture rates have been derived. This step screens/eliminates sites for which the procedure is appropriate. • Development Type: The MXD should be a single, physi- cally and functionally integrated development on a single block or a group of contiguous blocks with three or more revenue-producing uses, with internal pedestrian and vehic- ular connectivity, and with shared parking among some or all uses. The site should have sufficient parking supply to meet demand although the most convenient parking may sometimes fill during peak periods. • Development Location: The MXD should be downtown fringe, general urban, or suburban. It should not be located either within or adjacent to a central business district (CBD). • Development Size: The MXD should have at least 100,000 sq ft of building space within an overall acreage of up to roughly 300 acres. The MXD can be a single site, a block, or a district or neighborhood (with multiple inter- connected or interactive blocks within a defined bound- ary); however, this procedure should not be used for a SAC composed of different adjacent, but not directly con- nected, land uses. • Land Use Mix: The MXD should consist of a combination of at least three of the following uses: retail, restaurant, office, residential, hotel, and cinema. Internal capture for land uses beyond these six should be considered to be zero (unless comparable survey data for other land uses are provided) because there are no supporting data from which to derive an appropriate percentage. In addition, if a sub- stantial portion of the land use at a mixed-use site is outside these six land uses, the reported internal capture rates might not be appropriate. • ITE Trip Generation Database: The MXD should not already be covered in the ITE trip generation database as reported in the latest edition of Trip Generation (2). Cur- rent ITE land use classifications that already account for internal trip-making include shopping center, office park with retail, office building with ground floor retail or onsite cafeteria, and hotel with limited retail and restau- rant space. • Time Period for Analysis: The internal capture rates con- tained in this methodology cover the weekday A.M. and P.M. peak periods for adjacent street traffic. Weekday peak period internal capture rates are not appropriate for estimating weekend internal capture—or weekday midday internal capture—or daily internal capture unless survey data for those periods become available. Step 2: Define the Pertinent Site Characteristics In this step, the following data describing pertinent site characteristics are assembled: • The specific land uses in the mixed-use site in sufficient detail so that vehicle or person trip generation can be esti- mated for each individual land use (described in Steps 2A and 2B); and • Building proximity for each pair of land uses (described in Step 2B). The source of much of this information is a proposed site plan, if one exists. If a site plan is not available, assumptions must be made about general site layout, individual land uses, sharing of parking, and the internal pedestrian circulation system. Step 2A: Identify Land Uses Identify specific land use components of the MXD and assign them into the six classifications—office, retail, restaurant, residential, cinema, and hotel—covered by the estimation procedure. Any component land uses that do not fit into those six classifications or are too unique to be considered normal for a classification should be kept sepa- rate. If in doubt, keep a land use separate from the six listed classifications. Define the land use components in as much detail as pos- sible. The greatest detail will allow for greatest precision in trip-generation estimates. The internal capture relationships quantified in this methodology are provided at the aggregated land use level. It is important to separate the retail and restau- rant uses in this step because they exhibit different internal cap- 96

ture characteristics. If the restaurant component is expected to be only a minor portion of the overall retail component of the MXD (e.g., a traditional shopping center), assume the site has no restaurant component. Enter the development units by land use in sub Table 1-A of Table 103 and the corresponding sub table in the P.M. peak period Worksheet 2 (not shown). ITE land use codes are found in the ITE Trip Generation report (2). The “quantity” is the number of development units. “Units” are the applica- ble development units such as dwelling units or gross sq ft of building floor area. Undefined shopping center space should all be classified as just that—shopping center (ITE land use classification Code 820 or similar applicable classification). No guesses should be made as to how it may break out into cinema, restaurant, and so forth, unless that has already been determined in the development plan. Step 2B: Determine Proximity Determine the walking distance between each pair of inter- acting land uses within the MXD. This component of the esti- mation procedure requires particular consistency in applica- tion. If there is only one building of each land use classification (e.g., one apartment building and one office building), enter the distance between the entrances of each building. If there is a group of buildings or businesses of one land use category in an area, the distance used should be the weighted (by trip gen- eration) average of distances between each pair of buildings of the interacting land uses. For each pair of interacting land uses, determine the actual walking distance along the most direct and reasonable path. Do not use the airline (i.e., shortest direct) distance. For the A.M. street peak hour, there are no proximity adjustments, so the distances are not entered into sub Table 3-A of Table 103; how- ever, proximity distances are to be entered into sub Table 3-P of Worksheet 2 for the P.M. street peak-hour analyses. Step 3: Calculate Single-Use Trip Generation for the Site Components In this step, trip generation is estimated for each land use within the MXD. The procedure accounts for (1) trip- generating characteristics of the specific land uses (described in Step 3A) and (2) vehicle occupancy (described in Step 3B). Mode split is not applied here because it is assumed that the ITE trip generation data, which was almost all collected in suburban areas, is almost totally by motor vehicle. There is typically no or very limited transit and walking for trips to and from development sites. The recommended approach is to work in person trips rather than in vehicle trips, but the analyst can begin from vehi- cle trips and use mode split and vehicle occupancy to generate person trips. If the analyst wishes to work in assumed ITE con- ditions (no adjustments for mode split or vehicle occupancy), then it is workable to perform all calculations in this step (skip- ping Steps 3B and 3C) and all subsequent steps in vehicle trips. In this case, input mode split as 100% vehicle occupancy is 1.00; these will cause the inherent ITE values to be reflected through the process. Step 3A: Estimate Trip Generation Enter vehicle trips in the two right columns of Table 103, sub Table 1-A for the A.M. peak hour and in corresponding sub Table 1-P on Worksheet 2 for the P.M. peak hour. For each land use within the MXD, estimate single-use trip gen- eration individually. Then, sum the individual estimates into the six aggregated classifications: office, retail, restaurant, res- idential, cinema, and hotel. Do not combine development units into the six classifications and then use one single-use trip generation rate or equation to estimate trip generation for the aggregated land use. If specific land uses are not known at the time of analysis, use a more general category— for example, at zoning, no specific retail categories may be known, so “shopping center” may be the best approximation. The nationally accepted method of estimating site trip gen- eration is to use ITE Trip Generation report (2) trip genera- tion rates, equations, and data and apply them as described in the ITE Trip Generation Handbook (1). However, local agen- cies may have special local rates they prefer to use. Locally determined rates accepted by the reviewing agency can also be used. The choice of trip generation rates/equations should be discussed with the review agency prior to preparing the estimates. Analysts should keep track of the directional split (inbound/ outbound) of the generated trips for each land use. Directional trips are essential to the proper balancing of internal travel demand within the MXD (described in Step 4). If beginning directly with person trips, see the last paragraph of Step 3C. Step 3B: Enter Vehicle Occupancy Enter vehicle occupancy for the trips generated by each land use in Table 103, sub Table 2-A for the A.M. peak hour and corresponding sub Table 2-P of Worksheet 2 for the P.M. peak hour. The vehicle occupancy can be different for enter- ing and exiting vehicles. The vehicle occupancy rate should be based on local data if possible. It is acceptable to use an overall average vehicle occupancy rate based on a survey of a similar mixed-use site or to use land use specific vehicle occupancy rates based on surveys of nearby similar land uses. Metropolitan planning organization (MPO) data could also 97

be used to derive relevant averages for comparable trips in the region. It is important to use vehicle-occupancy rates that reflect travel during the analysis period. Do not use daily vehicle-occupancy rates. Step 3C: Enter Mode Split for MXD This is the mode split for the MXD. It does not apply to the base trip generation rates. Percentages of trips by transit and non-motorized mode (e.g., walk, bike) may be different by direction. Enter the percent of directional trips by each mode for each land use in Table 103, sub Table 2-A for the A.M. peak hour and corresponding sub Table 2-P of Worksheet 2 for the P.M. peak hour. The product of Step 3B is an estimate of the number of person-trips in vehicles entering and exiting each of the mixed-use development land uses. It is important to use mode of access distributions that reflect travel to and from the MXD during the analysis period, not daily. Step 3D: Compute Person Trips Using the vehicle trips entered in Table 103, sub Table 1-A, and vehicle occupancies entered in sub Table 2-A, compute directional trip generation for each land use: Enter the person trips in Table 104, sub Table 7-A for A.M. trips. There are corresponding tables in Worksheet 2 for com- puting P.M. estimates. Step 4: Estimate the Unconstrained Internal Capture Rates for All Land Use Pairs at the Site In this step, unconstrained internal capture rates that are appropriate for the subject development site are determined. This determination begins with the base internal capture rates documented in this research (described in Step 4A); the rates are then modified to account for specific proximity character- istics of the subject site (described in Step 4B). Step 4A: Estimate Base Internal Capture Rates Internal trip capture rates are provided for land use pairs involving the following generic land use classifications: office, retail, restaurant, residential, hotel, and cinema. Trip generation estimates were made in Step 3 for specific types of land uses. For estimating internal capture, land use classifi- cations should be combined into the above general categories before continuing into the estimation process—for example, for a development containing apartments and townhouses, Person trips vehicle trips vehicle occupancy= × . for estimating internal capture, these would be combined as residential. When applying the internal capture estimation methodol- ogy, use the percentages from the third and fourth columns of sub Table 7.1 and sub Table 7.2 within Tables 105 and 106. Each sub table contains both A.M. and P.M. peak-hour data. If a local survey has been conducted using data collection and compilation procedures described in this report, the result- ing internal capture percentages may be used. Users are cau- tioned that data gathered in a method different than the data collection methods described in this report may not be appli- cable and could produce inaccurate internal capture estimates. For land uses other than the six classifications provided herein, users should assume no internal capture (unless com- parable survey data for other land uses are provided). The percentages in Tables 105 and 106 are not applicable to other land uses. They are also only for the weekday A.M. and P.M. peak periods and should not be used for other periods. In some cases, review agencies may set policies to limit the percent internal capture they will permit to be applied based on their own justifications. Such limitations would represent agency policy to use what they consider more conservative trip generation estimates (e.g., to avoid the possibility of under- estimating trip generation) and are not to be confused with the findings of this project. Step 4B: Apply Proximity Adjustment Factors (P.M. Peak Period Only) The unconstrained internal capture values presented in Tables 105 and 106 represent rates to be expected between land use pairs that are not constrained by proximity. As some land use pairs become farther apart, the unconstrained inter- nal capture rates will decline. This step accounts for the spe- cific proximity characteristics of the MXD. Proximity adjustment factors are presented below for only the land use pairs for which the available data clearly demonstrates a direct relationship between proximity and internal capture rate. These are comprised of only P.M. peak- period trips; no proximity adjustments are available for the A.M. peak period: • From all land uses to residential and applicable only to the trip origin end; • From office to retail or restaurant, applicable at both ori- gin and destination trip ends; and • From residential to retail or restaurant, applicable at both origin and destination trip ends. These factors are only applicable during the P.M. peak hour. If a land use pair is not included in the above list, use 98

the Table 105 and Table 106 unconstrained internal capture values without adjustment. Locate the appropriate land use pair and direction in Table 107. Compare the MXD proximity to the proximity thresholds in the table: • If the proximity is less than or equal to the value in the third column of Table 107, use the unconstrained internal cap- ture values in Table 105 or Table 106, whichever is appro- priate. Remember that sub Table 7.1 values in Table 105 will be applied to the outbound trips; sub Table 7.2 values in Table 106 will be applied to inbound trips. • If the proximity is within the range shown in the third column of Table 107, use the equation provided in the fourth column to calculate the proximity adjustment factor. Enter the proximity adjustment in the right column of sub Table 7.1 of Table 105 and in sub Table 7.2 of Table 106. • If the value calculated in the fourth column is a proximity adjustment factor of less than 0.10, use the minimum value of 0.10 in subsequent steps. • For each row of sub Table 7.1, multiply the P.M. peak-hour adjustment factors in Column 4 by the P.M. proximity adjustment factor in the right column of that table. Place the resulting product in the right column of sub Table 7.1a of Table 105. Repeat the same for sub Table 7.2 in Table 106. • Enter the adjusted internal capture percentage in the right columns of sub Table 7.1a in Table 105 and in sub Table 7.2a in Table 106. Step 4C: Calculate Proximity-Adjusted Unconstrained Internal Trips at Origin (Outbound) and at Destination (Inbound) In Table 104, sub Tables 7-A(D) and 7-A(O) show the prox- imity adjusted internal vehicle and person trips at the origin and destination, respectively. Be sure that any mode splits from Table 103 sub Table 2-A have been incorporated. In Table 104, each cell in the 8-A(O) and 8-A(D) sub tables is computed as follows: 1. Multiply the direction trips in Table 104, sub Table 7-A(O) (e.g., office exiting trips) by the Table 105 sub Table 7.1 internal capture percentages (e.g., A.M. peak hour 1% to residential). 2. Place the product in Table 104, sub Table 8-A(O). Do the same for the entering trips. 3. Complete the 8-A sub tables of Table 104 in the same manner. P.M. street peak-hour internal trips can be computed the same way using the corresponding P.M. tables. Step 5: Calculate the Balanced Internal Trips between All Land Use Pairs Estimate balanced demand volume by direction by compar- ing the values in the corresponding cells of sub Tables 8-A(O) and 8-A(D) for each land use pair and select the lower value. 99 Table 107. Proximity adjustment factors for P.M. peak hour internal capture rates. From Land Use To Land Use Proximity Distance Equation to Calculate Proximity Adjustment Factor1 Office Residential 770–3,760 ft Factor = 1.23 – 0.0003 x (Distance) Retail Residential 770–3,760 ft Factor = 1.23 – 0.0003 x (Distance) Restaurant Residential 770–3,760 ft Factor = 1.23 – 0.0003 x (Distance) Cinema Residential 770–3,760 ft Factor = 1.23 – 0.0003 x (Distance) Hotel Residential 770–3,760 ft Factor = 1.23 – 0.0003 x (Distance) From Land Use To Land Use Proximity Distance Equation to Calculate Proximity Adjustment Factor2 190–1,524 ft Factor = 1.06 – 0.0003 x (Distance) Office Retail 1,525–2,360 ft Factor = 1.52 – 0.0006 x (Distance) 190–1,524 ft Factor = 1.06 – 0.0003 x (Distance) Residential Retail 1,525–2,360 ft Factor = 1.52 – 0.0006 x (Distance) 190–1,524 ft Factor = 1.06 – 0.0003 x (Distance) Office Restaurant 1,525–2,360 ft Factor = 1.52 – 0.0006 x (Distance) 190–1,524 ft Factor = 1.06 – 0.0003 x (Distance) Residential Restaurant 1,525–2,360 ft Factor = 1.52 – 0.0006 x (Distance) 1 Use 1.00 proximity factor for distances shorter than 770 ft and 0.10 for distances longer than 3,760 ft. 2 Use 1.00 proximity factor for distances shorter than 190 ft and 0.10 for distances longer than 2,360 ft.

This step is to balance the estimates of directional internally captured trips between the interacting land uses. This must be performed for two reasons: 1. Estimates for each land use are based on the quantity of that land use and its capacity to send or receive internal trips. There is no assurance without balancing that there is enough capacity on the receiving end to accept as many trips as are being sent. 2. The total trips sent internally (i.e., captured trips) from one use to another must equal the number being received at the other end of the trip. Both numbers must be the same. Figure 20 shows this process. To perform this computa- tion, person trips begin from Table 104, sub Tables 8-A(O) and 8-A(D). Compare corresponding cells and select the lowest figure (i.e., the fewest unconstrained internal trips). For exam- ple, compare the “from retail to office” cell. If sub Table 8-A(O) shows 4.4 trips and sub Table 8-A(D) shows 70.4 trips, select the lower value (i.e., 4.4) and enter it into the “retail to office” cell of Table 103, sub Table 4-A in round numbers (4, in this example). Complete sub Table 4-A of Table 103 in this manner. These are the estimated internal trips. As shown above, P.M. street peak-hour estimates can be computed using the corresponding tables. Step 6: Calculate the Overall Internal Capture Rate for the Site In Table 104, sub Tables 9-A (D) and 9-A(O) are used to summarize internally captured trips and compute the exter- nal trips. This is started in person trips. Column 2 of sub Table 9-A(D) is computed by summing the office column of sub Table 4-A of Table 103. Column 4 comes directly from Table 104, sub Table 7-A(D), Column 4. The external trips in Column 3 are the difference between the total and internal person trips in each row. The right three columns in sub Table 9-A(D) are com- puted by multiplying the external vehicle trips in Column 3 of that table by the applicable mode split percentage in the two right columns of sub Table 2-A of Table 103. The transit external trips are computed by multiplying the transit mode split percentage (Column 3, sub Table 2-A) by the number of external person trips (sub Table 9-A(D), Column 3). Non- motorized person trips are calculated similarly. For vehicle trips, use the equation of where the vehicle occupancy comes from Table 103, sub Table 2-A, Column 2. vehicle trips external trips transit trips = − − non-motorized trips vehicle occupancy ⎛⎝⎜ ⎞⎠⎟ For the bottom row in sub Tables 9-A(D) and 9-A(O), there are no internal trips (no data to support such estimates). Val- ues are taken directly from Table 103, sub Table 1-A, the right two columns. Totals from sub Tables 9-A(D) and 9-A(O) are then entered in Table 103, sub Table 5-A. The entering and exit- ing values in Row 1 of that table are the totals from Column 4, sub Tables 9-A(O) and 9-A(D). The total in Column 2 is the sum of the entering and exiting volumes. The second row of sub Table 5-A is the sum of Column 2 of sub Tables 9-A(O) and 9-A(D) divided by the sub Table 5-A, Row 1 entries then multiplied by 100%. The remaining entries in sub Table 5-A are taken from sub Tables 9-A(D) and 9-A(O) in a similar manner. Sub Table 6-A is computed using the row figures in sub Tables 9-A(D) and 9-A(O). Use the corresponding tables to compute P.M. street peak- hour estimates. Reminder The previous computational description follows a spread- sheet workbook designed to have a cover sheet that contains all input and output of interest, with look-up data and inter- mediate computations on the subsequent worksheets. It is intended that the spreadsheet workbook be used to perform computations. If performed manually, analysts may wish to reorder component tables to provide a more logical order. Additional Guidance Site Location The researchers recognize there is internal capture for developments other than single, physically and functionally integrated MXDs (such as CBDs and SACs). The concept of unconstrained internal capture rates constrained by the mix and proximity of land uses also applies to those development patterns. However, the data reported herein include only developments that satisfy the “mixed-use” definition used in this report. Mixed-Use Development Already in ITE Trip Generation Database In a typical shopping center that is included in the ITE Trip Generation report (2), the site restaurants are convenience restaurants that feed off the retail visitors (rather than serve as destination restaurants). In some MXDs, some of the restau- rants may be oriented to the convenience of internal users and not draw heavily from outside the development. These may be snack shops, fast food, or other small restaurants rather than full-scale restaurants that are destination eating places. Ana- lysts may wish to consider them part of a shopping center use if the retail uses generate a large portion of the convenience 100

restaurant business. If this is to be done, it is suggested that not more than the first 5% of overall retail/restaurant square footage be considered as convenience restaurant. This should exclude all destination and free-standing restaurants. In such a case, if greater than 5%, assume the amount above 5% to be restaurant—generate trips accordingly and keep this portion separate for determining the internal capture rates. Add the convenience restaurant square footage to the retail space. The ITE Trip Generation report (2) and the Trip Generation Handbook (1) already include several types of MXDs. Current ITE land use classifications that already account for internal trip-making include the following: • Shopping center: shopping-center trip-generation rates are based on retail developments that already normally include restaurant, cinema, and limited other entertainment uses; however, “if a shopping center is planned to have out-parcel development of a significantly different land use classifica- tion or a very large percentage of overall gross leasable area, the site could be considered a mixed-use development for the purpose of estimating site trip generation” (1). • Office park with retail: “A subdivision or planned unit development containing general office buildings and sup- port services such as banks, restaurants and service stations arranged in a park- or campus-like atmosphere should be considered as an office park (ITE Land Use Code 750 form ITE Trip Generation report),” not as an MXD (1). • Office building with ground floor retail or onsite cafeteria: “An office building with support retail or restaurant facili- ties contained inside the building should be treated as a gen- eral office building (Land Use Code 710) because the trip generation rates and equations already reflect the presence of such support uses” (1). • Hotel with limited retail and restaurant space. “A hotel with an onsite restaurant and small retail falls within Land Use Code 310 and should not be treated as a MXD” (1). Land Use Split between Retail and Restaurant The internal capture rates presented earlier in Tables 103 through 106 treat retail and restaurant as separate land uses. To use these rates, it is necessary to differentiate between retail and restaurant uses at the mixed-use site. It is possible that the analyst will only know total retail (i.e., retail plus restaurant) square footage. In that situation, two different approaches are suggested for estimating internal capture: 1. Assume the same retail/restaurant split found at the six sites for which data were available; and 2. Assume and test different retail/restaurant splits (within a reasonable range) to determine whether the retail/restaurant split changes site trip generation and internal capture significantly. The six-step estimation procedure is merely a mathemat- ical technique for estimating internal capture; the researchers are not trying to suggest how to adhere to specific local TIA requirements. Proximity of Land Uses If the analyst knows (or can confidently assume) the land uses and their sizes but does not know their proximities, the analyst must prepare at least a schematic site plan. Do not simply assume that each pair of land uses consists of build- ings adjacent to each other (e.g., within 200 ft). At the mini- mum, test different proximities and observe their effects on overall internal capture at the mixed-use site. If development information is not yet detailed enough to permit a direct estimate of proximity distances, use the site size and Figure 23 to estimate the average probable separa- tion, then use that distance for the proximities between each land use pair. This will produce rough estimates of internal capture, at least related to proximity. As the site plan evolves, use more specific information for proximity. The recommended approach is to separate each land use into blocks, with a block being the building faces along both sides of a street (see Figure 24). Locate the centroid of the entrances of a specific land use for each block and measure dis- tances between each block of that land use and the other inter- acting land use. If there are multiple blocks, then make mea- surements between all pairs of blocks and use the weighted average distance, using trip generation involved in each inter- change as the weighting factor. Figure 24 shows an example of a multi-block scenario. If the dots represent the centroid of the entrances for Land Use A in each block, and if d1 and d2 are the respective distances to Land Use B in Block 3, then the weighted average distance between Land Uses A and B is d d1 2×( )+ × sq ft in Block 1 of Land Use A sq ft in Block 2 of Land Use A sq ft in Block ( ) s 1 and 2 of Land Use A( ). 101 1 2 3 A A A A A B d2 d1 A Figure 24. Sample blocks including Land Uses A and B.

Proximity Adjustment Factor The proximity adjustment factors presented previously are based on the observed effects of changing walk distances on internal capture rates between land use pairs. These walking distances were measured along available routes that did not require the use of stairways or other obstacles or delaying fac- tors. The paths were also along clear, adequately wide walk- ways in very good repair. Mode Share of Internal Trips The great majority of internal trips to a site will be either walked or driven onsite (many sites also accommodate bicy- cle travel)—that is, few internal trips within a site will use major public streets either on the periphery of the develop- ment or internal to the development. An exception will be trips driven within a multi-block area of complementary and interacting land uses. For those MXD sites or areas where all internal trips will be walked, bicycled, or driven onsite (on private internal streets or through parking areas), the mode of access to the site should be used to factor vehicle external trip generation for the analy- sis period. The ITE trip generation rates and equations gen- erally incorporate suburban mode splits. Limited observa- tions within the ITE dataset point toward about 2% of the trips arriving by walking, bicycling, or transit. This would be a reasonable assumption to apply to ITE data. The other 98% arrives as either vehicle drivers or passengers. After arriving on the site, internal trips that could or might be driven will be limited by (1) the convenience (or lack thereof) of driving versus walking, and (2) the availability of a motor vehicle for the trip. Some MXDs may also have inter- nal shuttles. Hence, it is a reasonable assumption that the beginning point for internal trips by personal vehicle will be no higher than the mode of access to the site and possibly far lower if the development is walkable. The mode split is used at the end of the process to deter- mine the number of external person trips being taken by personal vehicle, transit, and non-motorized modes such as bicycle or walking. The mode split can be derived from sur- veys of similar land uses near the study site or from other estimates relevant to the study location and land uses (e.g., regional or localized travel data available from MPOs or other credible sources). Pass-By Trips “The application of pass-by trip reductions should be applicable to (mixed-use) sites. However, none of the inter- nal trips can be of a pass-by nature because they do not travel on the adjacent (external) street system” (1, p. 100). Pass-by trip percentages are applicable only to external trips—those trips that enter or exit the adjacent street system. They should be applied after the external trips are estimated, not to the base vehicle-trip generation. Unconstrained Internal Capture Rates The unconstrained internal capture rates presented in Tables 99 through 102 reflect data collected at as many as six MXDs. For several of the land uses, the potential sums of internal capture rates appear to be illogical—for example, the sum of 124% for internal capture for trips to office in the P.M. peak hour (see Table 102) is impossible. However, these “illogical” sums will not occur for three reasons: • First, they would require unlikely balances of interacting land uses. For example, to maximize the inbound office internal capture rate during the P.M. peak hour, the retail space would need to be 20 times the office space and the restaurant space would be half of the office space; a mix with so little office is essentially a shopping center. While the internal capture for trips to the office from retail would be high, the opposite would not be the case. The overall internal capture rate would be modest. • Second, the proximity adjustment factors will reduce the effective unconstrained internal capture rates because of the possibility that all office and residential uses will not be located within 200 ft of all retail and restaurant at the mixed-use site. • Third, it appears to be mathematically impossible for all unconstrained maximum internal capture percent- ages to occur at the same time within a development because each maximum requires a different ratio of development units for the pair of land uses involved. For example, using Tables 99 through 101, for the A.M. peak to achieve 65% internal capture for trips from office to restaurant, restaurant would need to have 63⁄23 times the square footage of the office for a balance to be achieved (the ratio of sq ft of each land use to achieve a complete balance between sending and receiving land uses neces- sary to obtain the unconstrained internal capture per- centage). To achieve the 28% internal capture of trips from office to retail, the square footage of office would have to be 28⁄32 times the square footage of retail. The office to residential is 0% due to the value in Table 101. So, for 100,000 sq ft of retail, office would have to have 87,500 sq ft of office and about 239,700 sq ft of restaurant to reach 91% internal capture. At average size of about 5,000 sq ft per restaurant, that would amount to 48 restaurants, a very unlikely balance—and that is only the balance results 102

for trips from office. To achieve the maximum for inter- nal trips to office, office square footage needs to be 4⁄2 times the retail square footage, which would not maxi- mize trips in the opposite direction. To maximize trips from restaurant to office, the office would have to have 41⁄3 times the square footage—again, different from what would be required for the opposite direction to maxi- mize. Following the same process, one can quickly see that it is mathematically impossible to achieve all uncon- strained internal capture percentages concurrently for a given MXD. In the unlikely occurrence that the sum of internal trips should total over 100% of the total trip generation for a land use, it is recommended that the total internal trips be reduced to 100%, and the inter- changes from the affected land use to other interacting land uses be proportionally reduced. This would not yield a total internal capture of 100 percent; rather, it would be one interchange and one direction that would be estimated to be 100%. Validation of Estimation Procedure This estimation procedure was tested against development and cordon count data for several developments and found to replicate actual results for MXDs fairly well. The validation procedure and results are described in Appendix F. 103

Next: Chapter 4 - Conclusions, Recommendations, and Suggested Research »
Enhancing Internal Trip Capture Estimation for Mixed-Use Developments Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s National Cooperative Highway Research Program (NCHRP) Report 684: Enhancing Internal Trip Capture Estimation for Mixed-Use Developments explores an improved methodology to estimate how many internal trips will be generated in mixed-use developments—trips for which both the origin and destination are within the development.

The methodology estimates morning and afternoon peak–period trips to and from six specific land use categories: office, retail, restaurant, residential, cinema, and hotel. The research team analyzed existing data from prior surveys and collected new data at three mixed-use development sites. The resulting methodology is incorporated into a spreadsheet model, which is available online for download.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!