National Academies Press: OpenBook

Using Archived AVL-APC Data to Improve Transit Performance and Management (2006)

Chapter: Chapter 7 - Tools for Analyzing Crowding

« Previous: Chapter 6 - Tools for Analyzing Waiting Time
Page 51
Suggested Citation:"Chapter 7 - Tools for Analyzing Crowding." National Academies of Sciences, Engineering, and Medicine. 2006. Using Archived AVL-APC Data to Improve Transit Performance and Management. Washington, DC: The National Academies Press. doi: 10.17226/13907.
×
Page 51
Page 52
Suggested Citation:"Chapter 7 - Tools for Analyzing Crowding." National Academies of Sciences, Engineering, and Medicine. 2006. Using Archived AVL-APC Data to Improve Transit Performance and Management. Washington, DC: The National Academies Press. doi: 10.17226/13907.
×
Page 52
Page 53
Suggested Citation:"Chapter 7 - Tools for Analyzing Crowding." National Academies of Sciences, Engineering, and Medicine. 2006. Using Archived AVL-APC Data to Improve Transit Performance and Management. Washington, DC: The National Academies Press. doi: 10.17226/13907.
×
Page 53

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.

Crowding is important to passengers for their comfort; to operations, because it can slow the boarding and alighting process; and to planning, as a measure of efficiency. These dif- ferent viewpoints need different measures derived from pas- senger count data. The planning viewpoint is concerned with average load at the peak volume point, that is, the segment whose average load is the greatest. Schedule planning often uses peak point load to determine headway, using a nominal design capacity. Because this measure is a single number that is widely under- stood, it is not covered further. However, impacts on opera- tions and on passengers are strongly affected by the random distribution of passenger crowding, something that can only be analyzed well with the large samples that APCs can afford. This chapter describes methods for analyzing passenger crowding. These methods have been programmed as proto- types in a spreadsheet file which is available on the project description web page for TCRP Project H-28 on the TRB website: www.trb.org. 7.1 Distribution of Crowding by Bus Trip For both the passenger and operations viewpoints, load can be examined on every segment of a trip. However, for most purposes, analysts want to focus on the most crowded segment (the maximum load segment) of each trip. The maximum load segment of any trip may differ from the route’s peak volume point. If averaged over many trips, the aver- age maximum load will often be greater, and cannot be smaller, than average load at the peak volume point. Average maximum load is a measure suggested by the Transit Capacity and Quality of Service Manual (TCQSM) to characterize level of service with respect to crowding (27). The TCQSM defines six levels of service (LOSs) (A through F) and suggests thresholds based on the number of seats and amount of available standing space per standee. The examples in this section use the thresholds shown in Table 8; they were deter- mined using TCQSM default values and assuming a 40-ft bus with 36 transverse seats, 6 longitudinal seats, a stairwell for the rear door, and 6 ft of unused length at the front of the bus. (For greater detail, the TCQSM’s LOS F has been subdivided into levels F1 and F2.) However, while these thresholds account well for passenger comfort levels on an individual trip, they do not mean much when applied to an “average trip.”Neither operations nor pas- sengers care much about average crowding. What really mat- ters is the distribution of crowding and, in particular, extreme values. Because of the large sample sizes afforded by APC data, analysts can derive, from maximum load observations, distri- butions of both trips and passengers by crowding level. To illustrate, the researchers analyzed 30 observations of peak-hour trips on a certain route and found that the mean value of maximum load was 40.3. (The data and analysis described in this chapter can be found in the spreadsheet file on the project description web page for TCRP Project H-28 on the TRB website: www.trb.org.) With the example 42-seat buses, the TCQSM would rate this route-period in LOS C. On average, load is less than the number of seats, which deceptively suggests that everybody should get a seat and that there should be little problem of crowding interfering with boarding and alighting. In fact, the distribution of maximum load over those 30 trips, shown in Figure 17, offers quite a different picture. About 47% of the trips had standees (load > 42); and 20% of the trips were either “crowded” or “overcrowded” (load > 62), which could seriously affect running time. Yet, 27% of the observed trips had at least half their bus seats empty, suggest- ing a possible bunching problem. 7.2 Distribution of Crowding Experience by Passenger 7.2.1 Classification of Crowding Experience Measures of passenger service quality should use passen- gers, not bus trips, as units and should adopt the passenger’s viewpoint. Crowding experience from the passenger’s view- point can be classified as follows: C H A P T E R 7 Tools for Analyzing Crowding 51

• Seated next to an empty seat (corresponds to LOS A) • Seated but not next to an empty seat (corresponds to LOS B-C) • Standing but not crowded (3.85 sq. ft. or more per standee; for this example, load is no more than 53) (corresponds to LOS D) • Standing and bus is full (2.2 to 3.85 sq. ft. per standee; for this example, load is between 54 and 62) (corresponds to LOS E) • Standing and crowded (1.6 to 2.2 sq. ft. per standee; for this example, load is between 62 and 69) (corresponds to LOS F1) • Standing and overcrowded (less than 1.6 sq. ft. per standee) (corresponds to LOS F2) In this classification, if a bus with 42 seats has 70 passen- gers, then 42 passengers will be classified as “seated but not next to an empty seat” (they experience LOS B-C); the other 28 will be classified as “standing and overcrowded” (they expe- rience LOS F2). When all the passengers on the 30 trips used in the previ- ous example are classified, their distribution is as shown in Figure 18. A comparison of Figures 17 and 18 shows how dif- ferent passenger experience is compared to the “experience” of the buses. While 6.7% of the buses were “overcrowded,” only 4.6% of the passengers were overcrowded standees—because more than half the passengers on the crowded buses were seated. Very few passengers fall into the category “standing but not crowded”because even though 17% of the buses had loads in this category, only a few of the passengers on those buses were standing. And while 27% of the buses were in the least crowded category, only 14% of the passengers are in that cate- gory, because not many passengers were on those buses. Distribution by passenger (rather than by trip) better indi- cates the level of service that passengers experience. For exam- ple, Figure 18 shows that 14% of the passengers had excellent service (during peak hour, they were able to sit with an empty seat next to them); and the vast majority had service equiva- lent to a carpool (a seat next to an occupied seat). However, about 5% experienced overcrowding as standees; they are the passengers who are likely to complain. A distribution of passengers by experienced crowding can be used to support a service quality standard such as • “During peak periods, at least 80% of our customers should be seated at the maximum load point, and 52 Load (pax) Basis Passenger Comfort TCQSM LOS 21 0.5 * (no. of seats) Can sit next to unoccupied seat A 32 0.75 * (no. of seats) Can choose seat B 42 no. of seats Seated C 53 3.85 sq. ft. per standee Standing but not crowded D 62 2.2 sq. ft. per standee Full E 69 1.6 sq. ft. per standee Borderline of crowded and overcrowded F1† †LOS F from theTCQSM has been subdivided into levels F1 and F2. Table 8. Example crowding thresholds. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Load more than 69 Load up to 69 Load up to 62 Load up to 53 Load up to 42 Load up to 32 Load up to 21 mean = 40.3 Figure 17. Distribution of trips by maximum load. Figure 18. Distribution of passengers by crowding experience. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Standing, load > 69 Standing, load up to 69 Standing, load up to 62 Standing, load up to 53 Seated next to occupied seat Seated next to empty seat

• “Fewer than 2% of our customers should have to stand on an overcrowded bus (a bus with less than 1.6 sq. ft. stand- ing space per person).” This standard is much more customer oriented than, for example, specifying that average load factor (load divided by number of seats) not exceed 1.2. The latter is a common format for load standards. However, passenger experience of crowding is not based on average load, but is sensitive to extremes; fur- thermore, even on crowded trips, passenger experience is very different depending on whether one has a seat. 7.2.2 Assumptions About Seated Passengers Passenger counts do not provide direct observation of how many passengers are seated or whether they are next to an empty seat. However, those figures can be estimated to a rea- sonable level of accuracy using three assumptions: passengers sit if they can, seats are in pairs (as is most often the case on buses), and passengers sit next to an empty seat if possible. Thus, for a bus with S seats and passenger load L, then • If L ≤ S/2, everybody sits next to an empty seat; • If L ≥ S, nobody sits next to an empty seat; and • If L lies between S/2 and S, S − L passengers sit next to an empty seat, because there are S − L empty seats. For exam- ple, if there are 42 seats and load is 32, 10 people sit next to an empty seat and the other 22 people sit in pairs next to an occupied seat. This formula is incorporated in the spreadsheet file that ana- lyzes crowding. (The spreadsheet is available on the project description web page for Project H-28 on the TRB website: www.trb.org.) Of course, the assumptions about where passengers sit or stand are sometimes violated; some passengers choose to stand when seats are available, and friends often choose to sit together, even if a bus is nearly empty. In such cases, however, the service received by those passengers is still appropriately classified based on the assumptions, because they had the option of sitting, or of sitting next to an empty seat, but chose something that they apparently preferred. 7.2.3 Alternative Classifications Differentiating seated passengers depending on whether a seat next to them is empty may be considered excessive detail. Of course, the two classes of seated passengers can be col- lapsed. However, to the extent that market research shows that passengers like having an empty seat next to them, both classes are worth measuring, particularly as these classes can be estimated using APC data. For some purposes, all the passengers, seated and standing in a crowded bus, may be said to experience overcrowding. This classification may be especially appropriate where the vehicle’s ventilation system allows overcrowding to result in unpleasant heat and odor. 53

Next: Chapter 8 - Passenger Count Processing and Accuracy »
Using Archived AVL-APC Data to Improve Transit Performance and Management Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB's Transit Cooperative Research Program (TCRP) Report 113: Using Archived AVL-APC Data to Improve Transit Performance and Management explores the effective collection and use of archived automatic vehicle location (AVL) and automatic passenger counter (APC) data to improve the performance and management of transit systems. Spreadsheet files are available on the web that provide prototype analyses of long and short passenger waiting time using AVL data and passenger crowding using APC data. Case studies on the use of AVL and APC data have previously been published as appendixes to TCRP Web-Only Document 23: Uses of Archived AVL-APC Data to Improve Transit Performance and Management: Review and Potential.

  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!