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.
100 7.0 Field Evaluation of Reversible Lanes A reversible lane is one in which the direction of traffic flow in one or more lanes is changed to the opposite direction for some period of time. Its utility is derived by taking advantage of the unused capacity of the minor flow direction to increase capacity in the major flow direction, thereby negating the need to construct additional lanes. Reversible lanes are particularly useful in work zones with a directional imbalance in excess of 65%â35% during weekday rush hours, where the existing number of lanes are reduced and the cost to provide additional capacity would be high and, perhaps, not possible because of the cost of ROW or other limitations. A concrete barrier, drums, or other traffic control devices separate the reversible lane from other lanes. Concrete barriers are generally used for longer work zones, which means there is no potential for diversions to on and off ramps and almost no access to or from the work zone area. The primary benefit of reversible lane operation is reducing congestion during periods of high and unbalanced directional travel demand. However, only a few studies have attempted to determine the effects of reversible lanes in work zones of the unused capacity in the minor traffic direction. The goal of this study was to evaluate the operational and safety effectiveness of reversible lanes in work zones. 7.1. Site Selection and Characteristics Through contacts made with state transportation agencies, the team identified the following three locationsâtwo in Michigan and one in Minnesotaâas test sites for evaluation: ⢠I-75 from Dixie Highway to Hess Road, Saginaw County, Michigan. Reversible-lane changeover in place during the weekend and mid-week to accommodate recreational and holiday traffic. ⢠I-94 between East 7th Street in St. Paul and Hwy 120/Century Ave. in Maplewood, Minnesota. Reversible-lane changeover in place after the morning peak period to accommodate the afternoon commuting traffic. ⢠I-75 and I-675 in Zilwaukee, Kochville, and Frankenlust Townships, Saginaw and Bay Counties, Michigan. Reversible-lane changeover in place during the weekend and mid- week to accommodate recreational and holiday traffic. 7.1.1 I-75, Saginaw County, Michigan This project included 3.75 mi of pavement reconstruction with widening for additional lanes, bridge replacements, drainage improvements, and construction of a noise barrier wall on I-75 from Dixie Highway to Hess Road, Saginaw County.
101 The bridge replacements and road reconstruction started in March 2015, and ended in November 2016. The team collected data from June 28 to July 12, 2016, when average daily traffic peaked because of the holiday period. Average daily traffic ranged from 65,000 to 75,000 vehicles; however, during holiday periods the traffic ranged from 100,000 to 110,000 vehicles, an increase of over 50% in any direction with a 70%â30% direction split. This was considered the most severe test of a reversible-lane operation. Three lanes typically serviced each direction; however, during the work zone operation five lanes served both northbound and southbound, with one lane alternating based on daily traffic flow. Posted work zone speed limit was 60 mph. Table 41 shows an example of the reversible-lane changeover. Table 41. I-75 Reversible-lane operational details. Date Day of Week Number of SB lanes after Switch Number of NB lanes after Switch Time Switch Began 06/28/2016 Tuesday 2 3 10:00 a.m. 07/02/2016* Saturday 2 3 8:00 a.m. 07/03/2016 Sunday 3 2 7:30 a.m. 07/06/2016 Wednesday 2 3 10:00 a.m. 07/09/2016 Saturday 3 2 10:00 a.m. *Reversible-lane changeover was to occur on 07/02/16, but much higher holiday traffic volumes in the northbound direction necessitated a change to 07/03/16. 7.1.2 I-94, Maplewood, Minnesota The major part of this project was to resurface the pavement of EB and WB I-94 between Mounds Boulevard to east of Highway 120/Century Avenue. The total length of the project was approximately 5.5 mi. The project also included (1) constructing a new auxiliary lane and extend the existing auxiliary lane along EB I-94 between the exit to East 7th Street and the entrance from Mounds Boulevard, (2) constructing two emergency pull-off sites along EB I-94 between the exit to East 7th Street and the entrance from Mounds Boulevard, (3) resurfacing Highway 61 from north of Burns Avenue to Highway 5, (4) building a new noise wall and replacing part of an existing noise wall between Conway Street and Maple Street, and (5) repairing bridges. The project started in spring 2016 and was completed in October 2018. The reversible lanes were in place during the time period from March to November 2017. The data were collected during the time period from June 24 to July 12, 2016. Average daily traffic ranged from 80,000 to 90,000 vehicles, with a 50%/50% direction split. Typically, three lanes serviced each direction. However, during the work zone operation, five lanes served both WB and EB, with one lane alternating based on peak period (a.m./p.m.) traffic flow. The changeover of the reversible lane
102 occurred dailyâmorning peak period to afternoon peak period between noon and 2:00 p.m., with the other changeover occurring after 8:.00 p.m. for the next day's morning peak period. The posted speed limit was 45 mph. Figure 37 shows the cross-section for p.m. peak hour. Figure 37. I-94 cross-section for p.m. peak hour. 7.1.3 I-75 and I-675, Saginaw and Bay Counties, Michigan The MDOT invested $22.9 million to reconstruct 1 mi of I-75 and I-675 through Saginaw County. The I-75 work zone was located between Exit 154 (Adams Street) to south of Exit 160 (Saginaw Road). The I-675 work zone extended from I-75 in Zilwaukee to Exit 6 (Tittabawassee Road). Work included repairing and reconstructing multiple bridges and ramps and resurfacing with both concrete and asphalt. Project construction started in March 2017 and was completed in November 2017, with the reversible lanes in place from June to October 2017. The team collected data during the time period from June 22 to July 13, 2017. ADT ranged from 65,000 to 75,000 vehicles; however, during holiday periods, the traffic ranged from 100,000 to 105,000 vehicles, an increase of over 50% in any direction with a 70%â30% direction split. Similar to the prior I-75 (Saginaw) evaluation, this project was considered to be one of the most severe tests of a reversible-lane operation. Typically, three lanes serviced each direction; however, during the work zone operation, five lanes served both northbound and southbound, with one lane alternating based on daily traffic flow. Posted speed limit was 60 mph. Table 42 shows an example of the reversible-lane changeover. Table 42. I-75 and I-675 Reversible-lane operational details. Date Day of Week Number of SB lanes after Switch Number of NB lanes after Switch Time Switch Began 6/24/2017 Saturday 3 2 2:15 a.m. 6/28/2017 Wednesday 2 3 11:30 a.m. 7/2/2017 Sunday 3 2 2:40 p.m. 7/6/2017 Thursday 2 3 8:00 a.m.
103 7/8/2017 Saturday 3 2 11:00 a.m. 7/12/2017 Wednesday 2 3 10:30 a.m. 7.2. Study Methodology 7.2.1 Data Collection Duration I-75, Saginaw County, Michigan. Data were collected in March and in June/July 2016. I-94, Maplewood, Minnesota. Data were collected in May and June 2017. I-75 and I-675, Saginaw and Bay Counties, Michigan. Data were collected in June and July 2017. 7.2.2 Data Collection Procedures The team used Wavetronix sensors to collect vehicular data. Wavetronix sensors collect data by emitting a microwave radar beam. The trailer-mounted sensors were stationed perpendicular to the roadway, outside the clear zone; as vehicles pass through the beam, the sensor detects the reflected microwave beam. The sensors can detect volume, vehicle classification, speed, 85th percentile speeds, and vehicle gaps across multiple lanes (up to 200 ft). The team collected all vehicular data by direction and by lane. At each location, the team measured the following data per laneâvolume, speed, vehicle classification, headway, and gap. The data were collected in 1-minute bins. All raw data were screened to exclude missing data values and outliers such as vehicles traveling at very low or very high speeds. 7.2.3 Measures of Effectiveness The team evaluated the following operational MOEs: ⢠Vehicle speed statistics along mainline with and without implementation of the reversible lane. Because the reversible lane would maintain the capacity for the peak direction, this treatment should maintain vehicle operating speeds. ⢠Travel time through the work zone with and without implementing the reversible lane. Travel time through the work zone is a measurement used to determine the effect of implementing the reversible lane. For all sites, the team used the baseline location and work zone location as the reference points to determine the travel time. As stated previously, because the intent of the reversible lane was to maintain the capacity in the peak direction, this treatment should maintain vehicle travel time through the corridor. ⢠Merging headways with and without implementing the reversible lane. Vehicle headway is a measure of the temporal space between two vehicles. As the average of vehicle headways is the reciprocal of flow rate, vehicle headways represent microscopic
104 measures of flow passing a point. To some extent, the minimum acceptable mean headway determines the roadway capacity. 7.2.4 Method for a Statistical Test for Vehicle Speeds The method for the statistical test for vehicle speeds is the same as that described in Section 5.2.5. 7.2.5 Method for a Statistical Test for Travel Time through the Work Zone The method for the statistical test for vehicle speeds is the same as that described in Section 6.2.5. 7.2.6 Method for a Statistical Test for Frequency of Headway The method for the statistical test for vehicle speeds is the same as that described in Section 5.2.6. 7.3. Field Evaluation Results The team compared the different MOEs (i.e., vehicle speed, travel time, and headway) to evaluate the effect of the reversible lane and determine if the reversible lane improved travel characteristics. As indicated in Section 7.1, the team selected three project sites to conduct the evaluation study. The following sections discuss the results of each measure of effectiveness. 7.3.1 Location: I-75, Saginaw County, Michigan 7.3.1.1 Comparison of Results for Vehicle Speeds The team collected data at four locations on I-75âtwo within the work zone (in the reversible- lane configuration) and one each upstream and downstream (outside the work zone). Figure 38 illustrates the data collection locations. Figure 39 illustrates the baseline traffic ADT volumes, as well as the volumes carried by the work zone in the reversible-lane configuration for both northbound and southbound directions. Traffic increased by over 30,000 vehicles during the 4th of July holiday period, showcasing the capability of a reversible lane to change on demand, as occurred on July 2, 2016. The reversible- lane switch was to occur on July 2 to the southbound direction; however, because of the higher northbound traffic volumes, the switch was delayed to July 3, 2016. Figure 40 illustrates the hourly traffic volume changes and the capability of the reversible lane to handle these higher volumes.
105 Figure 38. Data collection locations.
106 Figure 39. I-75 daily traffic volumes (with reversible-lane change times).
107 Figure 40. I-75 hourly traffic volumes (with reversible-lane change times).
108 Figures 41 and 42 illustrate the vehicle speeds and traffic volumes on the mainline for northbound direction and southbound direction, respectively. The graphs indicated that I-75 did not have daily peaks; rather, a peak period that spans the entire day from about 10:00 a.m. to 7:00 p.m., regardless of direction. In general, work zones will reduce capacity by at least 5% to 15% when the roadway is at or close to capacity due to narrow lanes or shoulders, barrier close to travel lane and other geometric changes. However, by implementing the reversible lane, it appears that capacity is maintained and vehicle speeds were within ±5% of baselines speeds.
109 Figure 41. Northbound average speed and traffic volumes (10:00 a.m. to 7:00 p.m.). 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 12 :0 0: 00 A M 12 :3 0: 00 A M 1: 00 :0 0 AM 1: 30 :0 0 AM 2: 00 :0 0 AM 2: 30 :0 0 AM 3: 00 :0 0 AM 3: 30 :0 0 AM 4: 00 :0 0 AM 4: 30 :0 0 AM 5: 00 :0 0 AM 5: 30 :0 0 AM 6: 00 :0 0 AM 6: 30 :0 0 AM 7: 00 :0 0 AM 7: 30 :0 0 AM 8: 00 :0 0 AM 8: 30 :0 0 AM 9: 00 :0 0 AM 9: 30 :0 0 AM 10 :0 0: 00 A M 10 :3 0: 00 A M 11 :0 0: 00 A M 11 :3 0: 00 A M 12 :0 0: 00 P M 12 :3 0: 00 P M 1: 00 :0 0 PM 1: 30 :0 0 PM 2: 00 :0 0 PM 2: 30 :0 0 PM 3: 00 :0 0 PM 3: 30 :0 0 PM 4: 00 :0 0 PM 4: 30 :0 0 PM 5: 00 :0 0 PM 5: 30 :0 0 PM 6: 00 :0 0 PM 6: 30 :0 0 PM 7: 00 :0 0 PM 7: 30 :0 0 PM 8: 00 :0 0 PM 8: 30 :0 0 PM 9: 00 :0 0 PM 9: 30 :0 0 PM 10 :0 0: 00 P M 10 :3 0: 00 P M 11 :0 0: 00 P M 11 :3 0: 00 P M Vo lu m e (V eh icl e/ M in ) Sp ee d (M PH ) Time NB Average Speed and Traffic Volumes (Reversible Lane Location vs Baseline Location, 1 Minute Bin) Reversible Lane Location - WorkZone (Speed) - 3 day-Average Baseline Location - Baseline (Speed) - 3 day-Average Reversible Lane Location - WorkZone (Volume) - 3 day-Average Baseline Location - Baseline (Volume) - 3 day-Average Poly. (Reversible Lane Location - WorkZone (Volume) - 3 day-Average ) Poly. (Baseline Location - Baseline (Volume) - 3 day-Average )
110 Figure 42. Southbound average speed and traffic volumes (10:00 a.m. to 7:00 p.m.). 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 12 :0 0: 00 A M 12 :3 0: 00 A M 1: 00 :0 0 AM 1: 30 :0 0 AM 2: 00 :0 0 AM 2: 30 :0 0 AM 3: 00 :0 0 AM 3: 30 :0 0 AM 4: 00 :0 0 AM 4: 30 :0 0 AM 5: 00 :0 0 AM 5: 30 :0 0 AM 6: 00 :0 0 AM 6: 30 :0 0 AM 7: 00 :0 0 AM 7: 30 :0 0 AM 8: 00 :0 0 AM 8: 30 :0 0 AM 9: 00 :0 0 AM 9: 30 :0 0 AM 10 :0 0: 00 A M 10 :3 0: 00 A M 11 :0 0: 00 A M 11 :3 0: 00 A M 12 :0 0: 00 P M 12 :3 0: 00 P M 1: 00 :0 0 PM 1: 30 :0 0 PM 2: 00 :0 0 PM 2: 30 :0 0 PM 3: 00 :0 0 PM 3: 30 :0 0 PM 4: 00 :0 0 PM 4: 30 :0 0 PM 5: 00 :0 0 PM 5: 30 :0 0 PM 6: 00 :0 0 PM 6: 30 :0 0 PM 7: 00 :0 0 PM 7: 30 :0 0 PM 8: 00 :0 0 PM 8: 30 :0 0 PM 9: 00 :0 0 PM 9: 30 :0 0 PM 10 :0 0: 00 P M 10 :3 0: 00 P M 11 :0 0: 00 P M 11 :3 0: 00 P M Vo lu m e (V eh icl e/ M in ) Sp ee d (M PH ) Time SB Average Speed and Traffic Volumes (Reversible Lane Location vs Baseline Location, 1 Minute Bin) Reversible Lane Location - WorkZone (Speed) - 2 day-Average Baseline Location - Baseline (Speed) - 2 day-Average Reversible Lane Location - WorkZone (Volume) - 2 day-Average Baseline Location - Baseline (Volume) - 2 day-Average Poly. (Reversible Lane Location - WorkZone (Volume) - 2 day-Average ) Poly. (Baseline Location - Baseline (Volume) - 2 day-Average )
111 7.3.1.2 Statistical Analysis of Vehicle Speed The team calculated changes in the mean speeds and 85th percentile speeds for vehicles between with and without implementation of the reversible lane. Tables 43 and 44 show the comparison of mean speed and 85th percentile speed of vehicles on the mainline of the freeway and the statistical test results with and without the implementation of the reversible lane. Table 43. Speed comparisonânorthbound direction baseline location vs. reversible-lane location. Speed Comparison for NB I-75 Volume Sample Size Mean Speed (mph) 85th Percentile Speed (mph) SD Mean Speed tstatic 1000- 1100 Baseline 10734 692 66.52 77.00 14.60 5.96 Reversible Lane 11037 507 61.18 70.00 15.81 1100- 1200 Baseline 10950 694 63.85 77.00 16.80 4.00 Reversible Lane 10714 518 59.99 70.00 16.43 1200- 1300 Baseline 10734 692 66.52 77.00 14.60 7.61 Reversible Lane 11093 511 59.40 71.00 17.02 1300- 1400 Baseline 10950 694 63.85 77.00 16.80 5.88 Reversible Lane 10352 499 57.91 70.00 17.50 1400- 1500 Baseline 10734 692 66.52 77.00 14.60 7.85 Reversible Lane 9157 485 58.45 72.00 19.05 1500- 1600 Baseline 10734 692 66.52 77.00 14.60 8.29 Reversible Lane 9813 496 58.30 71.00 18.28 1600- 1700 Baseline 10950 694 63.85 77.00 16.80 3.60 Reversible Lane 10543 488 60.15 72.00 17.80 1700- 1800 Baseline 8258 662 42.74 77.00 28.52 -6.54 Reversible Lane 9560 490 52.74 72.00 23.35 1800- 1900 Baseline 8974 656 57.48 78.00 25.51 10.49 Reversible Lane 6533 458 39.79 72.00 29.15 (Bold indicates significance at the 95% confidence level, α = .05) The t-test results indicated that changes in northbound mean speed during implementation periods of the reversible lane were statistically significant for all time periods. However, the work zone within these congested daily periods was able to maintain traffic volumes and speeds throughout the day.
112 As Table 44 shows, the southbound mean speeds of vehicles on the mainline of the freeway increased for all time periods from 10:00 a.m. to 7:00 p.m. Table 44. Speed comparisonâsouthbound direction baseline location vs. reversible-lane location. Speed Comparison for SB I-75 Volume Sample Size Mean Speed (mph) 85th Percentile Speed (mph) SD Mean Speed tstatic 1000- 1100 Baseline 9537 480 62.37 83.00 27.60 -7.48 Reversible Lane 9347 359 71.97 77.00 4.63 1100- 1200 Baseline 8451 480 44.37 81.00 34.04 -13.50 Reversible Lane 9172 359 67.72 76.00 14.42 1200- 1300 Baseline 8125 480 43.62 80.00 34.48 -4.67 Reversible Lane 7591 332 53.05 73.00 23.07 1300- 1400 Baseline 8287 480 43.51 80.15 34.15 -4.69 Reversible Lane 7453 346 53.25 74.00 25.60 1400- 1500 Baseline 7757 480 44.56 80.00 33.47 -6.06 Reversible Lane 7030 360 56.84 76.00 25.28 1500- 1600 Baseline 7360 480 43.74 80.00 34.18 -5.51 Reversible Lane 6305 360 55.35 77.00 26.87 1600- 1700 Baseline 7051 480 44.27 81.00 34.57 -6.50 Reversible Lane 6337 351 57.83 78.00 25.57 1700- 1800 Baseline 7062 480 44.58 82.00 35.04 -13.61 Reversible Lane 6643 359 69.29 78.00 16.28 1800- 1900 Baseline 7363 480 44.79 82.00 35.27 -18.60 Reversible Lane 6398 359 75.04 79.00 4.44 (Bold indicates significance at the 95% confidence level, α = .05) The t-test results indicated the changes in mean speed during implementation periods of the reversible lane were statistically significant for time periods from 10:00 a.m. to 7:00 p.m. After implementing the reversible lane, the vehicle speeds were maintained on the mainline in both directions of the freewayâdespite volume increases.
113 7.3.1.3 Comparison of Results for Travel Time Comparing travel times through the work zone is a good measurement of effectiveness to determine the effect of implementing the reversible lane. The reversible-lane operation maintains capacity through the work zone and is expected to have no effect on vehicle speeds, which will result in similar travel times to the baseline. The team analyzed the travel times of vehicles on the mainline of the freeway through the project limit during the daily the peak hours (10:00 a.m. to 7:00 p.m.) to determine the effects of implementing the reversible lane over the 6.5-mi segment. Figures 43 and 44 illustrate travel time through the project limits for northbound and southbound peak direction periods, respectively. The graphs indicated that the reversible lane was generally able to maintain travel times when compared to the baseline conditions. Travel times were maintained during the later afternoon period when traffic volumes increased by 20% to 30%.
114 Figure 43. Northbound travel time (distance: 6.5 mi). 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 12 :0 0: 00 A M 12 :3 7: 00 A M 1: 14 :0 0 AM 1: 51 :0 0 AM 2: 28 :0 0 AM 3: 05 :0 0 AM 3: 42 :0 0 AM 4: 19 :0 0 AM 4: 56 :0 0 AM 5: 33 :0 0 AM 6: 10 :0 0 AM 6: 47 :0 0 AM 7: 24 :0 0 AM 8: 01 :0 0 AM 8: 38 :0 0 AM 9: 15 :0 0 AM 9: 52 :0 0 AM 10 :2 9: 00 A M 11 :0 6: 00 A M 11 :4 3: 00 A M 12 :2 0: 00 P M 12 :5 7: 00 P M 1: 34 :0 0 PM 2: 11 :0 0 PM 2: 48 :0 0 PM 3: 25 :0 0 PM 4: 02 :0 0 PM 4: 39 :0 0 PM 5: 16 :0 0 PM 5: 53 :0 0 PM 6: 30 :0 0 PM 7: 07 :0 0 PM 7: 44 :0 0 PM 8: 21 :0 0 PM 8: 58 :0 0 PM 9: 35 :0 0 PM 10 :1 2: 00 P M 10 :4 9: 00 P M 11 :2 6: 00 P M Tr av el T im e (M in ut e) Time NB Travel Time (Through Project Limits) Travel Time - Reversible Lane Location (Min) - 3 Day Average Travel Time - Baseline Location (Min) - 3 Day Average
115 Figure 44. Southbound travel time (distance: 6.5 mi). 0.0 2.0 4.0 6.0 8.0 10.0 12.0 12 :0 0: 00 A M 12 :3 7: 00 A M 1: 14 :0 0 AM 1: 51 :0 0 AM 2: 28 :0 0 AM 3: 05 :0 0 AM 3: 42 :0 0 AM 4: 19 :0 0 AM 4: 56 :0 0 AM 5: 33 :0 0 AM 6: 10 :0 0 AM 6: 47 :0 0 AM 7: 24 :0 0 AM 8: 01 :0 0 AM 8: 38 :0 0 AM 9: 15 :0 0 AM 9: 52 :0 0 AM 10 :2 9: 00 A M 11 :0 6: 00 A M 11 :4 3: 00 A M 12 :2 0: 00 P M 12 :5 7: 00 P M 1: 34 :0 0 PM 2: 11 :0 0 PM 2: 48 :0 0 PM 3: 25 :0 0 PM 4: 02 :0 0 PM 4: 39 :0 0 PM 5: 16 :0 0 PM 5: 53 :0 0 PM 6: 30 :0 0 PM 7: 07 :0 0 PM 7: 44 :0 0 PM 8: 21 :0 0 PM 8: 58 :0 0 PM 9: 35 :0 0 PM 10 :1 2: 00 P M 10 :4 9: 00 P M 11 :2 6: 00 P M Tr av el T im e (M in ut e) Time SB Travel Time (Through Project Limits) Travel Time - Reversible Lane Location (Min) - 2 Day Average Travel Time - Baseline Location (Min) - 2 Day Average
116 7.3.1.4 Statistical Analysis of Travel Time Tables 45 and 46 show the comparison of average travel time with and without the reversible lane. As Table 45 shows, the NB average travel time through the project limits was maintained or lower with the reversible-lane operationâexcept during the late afternoon when traffic volumes increased by 20% to 30%. The t-test results indicated the changes in travel time were statistically significant for most time periods. Table 45. Travel time comparisonânorthbound direction baseline location vs. reversible-lane location. Travel Time Comparison for NB I-75 Volume Sample Size Average Travel Time (Min) 85th Percentile Travel Time (Min) SD Mean Travel Time tstatic 1000- 1100 Baseline 10734 180 6.35 8.49 1.79 -2.02 Reversible Lane 11037 179 6.68 8.47 1.26 1100- 1200 Baseline 10950 180 6.82 9.25 2.53 -1.16 Reversible Lane 10714 180 7.15 8.80 2.87 1200- 1300 Baseline 10581 180 7.87 10.40 6.11 1.22 Reversible Lane 11093 180 7.24 8.80 3.43 1300- 1400 Baseline 10621 180 8.16 11.70 5.79 0.83 Reversible Lane 10352 180 7.69 9.47 5.05 1400- 1500 Baseline 10636 180 9.57 10.20 11.00 1.57 Reversible Lane 9157 179 8.06 10.27 6.76 1500- 1600 Baseline 10817 180 8.83 11.74 7.97 1.30 Reversible Lane 9813 179 7.91 9.46 5.08 1600- 1700 Baseline 10337 180 8.58 9.90 7.52 1.83 Reversible Lane 10543 175 7.40 8.95 4.25 1700- 1800 Baseline 10101 180 8.05 10.18 6.71 -3.43 Reversible Lane 9560 180 11.93 17.35 13.57 1800- 1900 Baseline 8258 180 18.61 39.44 16.84 -3.57 Reversible Lane 6533 180 28.46 61.16 32.97 (Bold indicates significance at the 95% confidence level, α = .05) As Table 46 shows, the SB average travel time through the project limits decreased for all time periods from 10:00 a.m. to 7:00 p.m. The t-test results indicated the changes in travel time were statistically significant for all southbound time periods.
117 Table 46. Travel time comparisonâsouthbound direction baseline location vs. reversible-lane location. Travel Time Comparison for SB I-75 Volume Sample Size Average Travel Time (Min) 85th Percentile Travel Time (Min) SD Mean Travel Time tstatic 1000- 1100 Baseline 9537 120 6.25 28.62 16.77 0.54 Reversible Lane 9347 120 5.42 5.66 0.26 1100- 1200 Baseline 8451 120 8.79 54.29 28.11 1.17 Reversible Lane 9172 120 5.76 5.83 4.58 1200- 1300 Baseline 8125 120 8.94 47.27 21.76 0.66 Reversible Lane 7591 120 7.35 17.76 14.99 1300- 1400 Baseline 8287 120 8.96 50.10 21.93 0.62 Reversible Lane 7453 120 7.32 14.09 18.93 1400- 1500 Baseline 7757 120 8.75 48.44 22.44 0.91 Reversible Lane 7030 120 6.86 11.70 3.14 1500- 1600 Baseline 7360 120 8.91 49.09 21.82 0.92 Reversible Lane 6305 120 7.05 12.22 3.79 1600- 1700 Baseline 7051 120 8.81 55.84 24.44 0.92 Reversible Lane 6337 120 6.74 12.35 3.76 1700- 1800 Baseline 7062 120 8.75 53.73 29.40 1.16 Reversible Lane 6643 120 5.63 5.96 2.39 1800- 1900 Baseline 7363 120 8.71 50.62 23.24 1.65 Reversible Lane 6398 120 5.20 5.40 0.21 (Bold indicates significance at the 95% confidence level, α = .05) After implementing the reversible lane, the travel time through the project limits were generally maintained or lower in both directions. 7.3.1.5 Comparison of Results for Frequency of Headway Headway is a good measure of congestion and lack of passing opportunities created by the traffic mix; it is also a good measure of safety as lane changing and frequent passing generally lead to conflicts and the likelihood of crashes. In general, a longer headway accepted by a merging vehicle is safer than a shorter headway. The team examined the headways accepted by following vehicles to see if there were any differences between the with and without implementation of the reversible lane. The team conducted an analysis of vehicles headways on the mainline of the freeway at the baseline location and the reversible-lane location to determine the average values and distribution for both northbound and southbound peak direction periods.
118 As mentioned above, the team used the K-S test to judge how faithfully a distribution fits the sample data. The team adopted the K-S test to determine the goodness-of-fit in the work zone traffic condition. Table 47 summarizes the K-S test results of the northbound peak direction period vs. reversible- lane location. Table 47. K-S test results for the northbound peak direction period. Northbound Peak Direction Period Baseline Location Reversible-Lane Location Sample Size 3,940 3,782 Mean Headway (seconds) 3.31 3.14 Median Headway (seconds) 3.00 2.86 Maximum difference (D) 0.08 Significance Yes Figure 45 presents a visual performance comparison of headway distributions through a cumulative distribution function in the peak hours (10:00 a.m. to 7:00 a.m.) at the baseline location and the reversible-lane location. The team observed a shift in the headway distribution toward the shorter headway at the reversible-lane location compared to baseline location. The mean value of headway was 3.31 seconds at baseline location as opposed to 3.14 seconds at reversible-lane location. With the significance level α of 0.05 and the sample size more than 40, the critical statistic of the K-S test for the maximum difference between the cumulative distributions, D, was 0.03. The results of the K-S test (baseline location vs. reversible-lane location) show a value of D of 0.08 (greater than the critical value of 0.03), which suggests the differences in the two cumulative distributions are statistically significant.
119 Figure 45. Cumulative headway distribution plotânorthbound peak direction periodâbaseline location vs. reversible-lane location (10:00 a.m. to 7:00 p.m.). 0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% 80.00% 90.00% 100.00% 0 1 2 3 4 5 6 7 8 9 Cu m ul at iv e pe rc en ta ge Headway (Sec) Baseline Location Reversible Lane Location
120 Table 48 summarizes the K-S test results of the southbound peak direction vs. reversible-lane location. Table 48. K-S test results for the southbound peak direction period. Southbound Peak Direction Period Baseline Location Reversible-Lane Location Sample Size 3,392 2,945 Mean Headway (seconds) 3.56 3.10 Median Headway (seconds) 3.53 2.86 Maximum difference (D) 0.16 Significance Yes Figure 46 presents a visual performance comparison of headway distributions through a cumulative distribution function in the peak hours (10:00 a.m. to 7:00 p.m.) at the baseline location and the reversible-lane location. The team observed a shift in the headway distribution toward shorter headway at the reversible-lane location compared to the baseline location. The mean value of headway was 3.56 seconds at baseline location as opposed to 3.10 seconds at the reversible-lane location. With the significance level α of 0.05 and the sample size more than 40, the critical statistic of the K-S test for the maximum difference between the cumulative distributions, D, was 0.03. The results of the K-S test for (baseline location vs. reversible-lane location) shows a value of D of 0.16 (greater than the critical value of 0.03), which suggests the differences in the two cumulative distributions are statistically significant.
121 Figure 46. Cumulative headway distribution plotâsouthbound peak direction periodâ baseline location vs. reversible-lane location (10:00 a.m. to 7:00 p.m.). The headways of vehicles on the mainline of the freeway at reversible-lane location exhibited a decrease in both northbound peak direction and southbound peak direction periods. The result of the K-S test indicated that the differences in the two cumulative distributions were statistically significant. The late afternoon traffic volume increase may have contributed to shorter headways. It can be reasonably concluded that although the mean headway decreased, it remained above 3 seconds. 0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% 80.00% 90.00% 100.00% 0 1 2 3 4 5 6 7 8 9 Cu m ul at iv e pe rc en ta ge Headway (Sec) Baseline Location Reversible Lane Location
122 7.3.2 Location: I-94, Maplewood, Minnesota 7.3.2.1 Comparison of Results for Vehicle Speeds The team collected data at four locations on I-94âtwo within the work zone (in the reversible- lane configuration) and one each upstream and downstream (outside the work zone). Figure 47 illustrates the data collection locations. Figure 48 illustrates the baseline traffic ADT volumes as well as the volumes carried by the work zone in the reversible-lane configuration for both northbound and southbound directions. I-94 carries commuting traffic and this evaluation showcases the capability of a reversible lane to manage demand. Based on daily peak period demand, as Figure 48 shows, the reversible operation was changed over during the mid-afternoon or early evening.
123 Figure 47. Data collection locations.
124 Figure 48. I-94 hourly traffic volumes.
125 Figures 49 and 50 illustrate the vehicle speeds and traffic volumes on the mainline for the a.m. peak westbound direction and the p.m. peak eastbound direction, respectively. In general, for the a.m. peak westbound direction, it appears that the reversible lane resulted in improved mainline performance. However, for the p.m. peak eastbound direction, implementing the reversible lane reduced the mainline speeds. In general, work zones will reduce capacity by at least 5% to 15% when the roadway is at or close to capacity due to narrow lanes or shoulders, barrier close to travel lane and other geometric changes. However, by implementing the reversible lane, it appears that capacity was generally maintained and vehicle speeds were within ±5% of baselines speeds.
126 Figure 49. A.M. eastbound average speed and traffic volumes. 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 6: 00 :0 0 AM 6: 05 :0 0 AM 6: 10 :0 0 AM 6: 15 :0 0 AM 6: 20 :0 0 AM 6: 25 :0 0 AM 6: 30 :0 0 AM 6: 35 :0 0 AM 6: 40 :0 0 AM 6: 45 :0 0 AM 6: 50 :0 0 AM 6: 55 :0 0 AM 7: 00 :0 0 AM 7: 05 :0 0 AM 7: 10 :0 0 AM 7: 15 :0 0 AM 7: 20 :0 0 AM 7: 25 :0 0 AM 7: 30 :0 0 AM 7: 35 :0 0 AM 7: 40 :0 0 AM 7: 45 :0 0 AM 7: 50 :0 0 AM 7: 55 :0 0 AM 8: 00 :0 0 AM 8: 05 :0 0 AM 8: 10 :0 0 AM 8: 15 :0 0 AM 8: 20 :0 0 AM 8: 25 :0 0 AM 8: 30 :0 0 AM 8: 35 :0 0 AM 8: 40 :0 0 AM 8: 45 :0 0 AM 8: 50 :0 0 AM 8: 55 :0 0 AM 9: 00 :0 0 AM 9: 05 :0 0 AM 9: 10 :0 0 AM 9: 15 :0 0 AM 9: 20 :0 0 AM 9: 25 :0 0 AM 9: 30 :0 0 AM 9: 35 :0 0 AM 9: 40 :0 0 AM 9: 45 :0 0 AM 9: 50 :0 0 AM 9: 55 :0 0 AM Vo lu m e (V eh icl e/ M in ) Sp ee d (M PH ) Time AM WB Average Speed and Traffic Volumes (at White Bear Avenue, 1 Minute Bin) With Reversible Lane Implementation (Speed) Without Reversible Lane Implementation (Speed) With Reversible Lane Implementation (Volume) Without Reversible Lane Implementation (Volume) Poly. (With Reversible Lane Implementation (Volume)) Poly. (Without Reversible Lane Implementation (Volume))
127 Figure 50. P.M. eastbound average speed and traffic volumes. 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 3: 00 :0 0 PM 3: 05 :0 0 PM 3: 10 :0 0 PM 3: 15 :0 0 PM 3: 20 :0 0 PM 3: 25 :0 0 PM 3: 30 :0 0 PM 3: 35 :0 0 PM 3: 40 :0 0 PM 3: 45 :0 0 PM 3: 50 :0 0 PM 3: 55 :0 0 PM 4: 00 :0 0 PM 4: 05 :0 0 PM 4: 10 :0 0 PM 4: 15 :0 0 PM 4: 20 :0 0 PM 4: 25 :0 0 PM 4: 30 :0 0 PM 4: 35 :0 0 PM 4: 40 :0 0 PM 4: 45 :0 0 PM 4: 50 :0 0 PM 4: 55 :0 0 PM 5: 00 :0 0 PM 5: 05 :0 0 PM 5: 10 :0 0 PM 5: 15 :0 0 PM 5: 20 :0 0 PM 5: 25 :0 0 PM 5: 30 :0 0 PM 5: 35 :0 0 PM 5: 40 :0 0 PM 5: 45 :0 0 PM 5: 50 :0 0 PM 5: 55 :0 0 PM 6: 00 :0 0 PM 6: 05 :0 0 PM 6: 10 :0 0 PM 6: 15 :0 0 PM 6: 20 :0 0 PM 6: 25 :0 0 PM 6: 30 :0 0 PM 6: 35 :0 0 PM 6: 40 :0 0 PM 6: 45 :0 0 PM 6: 50 :0 0 PM 6: 55 :0 0 PM Vo lu m e (V eh icl e/ M in ) Sp ee d (M PH ) Time PM EB Average Speed and Traffic Volumes (at White Bear Avenue, 1 Minute Bin) With Reversible Lane Implementation (Speed) Without Reversible Lane Implementation (Speed) With Reversible Lane Implementation (Volume) Without Reversible Lane Implementation (Volume) Poly. (With Reversible Lane Implementation (Volume)) Poly. (Without Reversible Lane Implementation (Volume))
128 7.3.2.2 Statistical Analysis of Vehicle Speed The team calculated changes in the mean speeds and 85th percentile speeds for vehicles between with and without implementation of the reversible lane. Tables 49 and 50 show the comparison of mean speed and 85th percentile speed of vehicles on the mainline of the freeway and the statistical test results with and without implementation of the reversible lane. As Table 49 shows, the a.m. peak westbound mean speeds of vehicles on the mainline of the freeway generally improved in spite of additional traffic volumes. The t-test results indicated statistically significant changes in mean speed during the periods of implementation of the reversible lane between 7:00 a.m. to 10:00 a.m. Table 49. Speed comparisonâ(a.m. peak westbound) without reversible lane vs. with reversible lane. Speed Comparison for WB I-94 Volume Sample Size Mean Speed (mph) 85th Percentile Speed (mph) SD Mean Speed tstatic 0600- 0700 Baseline 13200 540 60.46 72.47 15.08 -0.62 With Reversible Lane 9748 531 60.87 64.00 3.37 0700- 0800 Baseline 12073 539 39.36 68.00 24.08 -15.68 With Reversible Lane 12870 525 56.21 60.00 6.47 0800- 0900 Baseline 10954 540 53.67 71.79 21.29 -3.95 With Reversible Lane 11348 525 57.43 62.00 5.84 0900- 1000 Baseline 8679 540 63.41 72.23 8.61 12.09 With Reversible Lane 8342 528 54.83 62.00 13.91 (Bold indicates significance at the 95% confidence level, α = .05) As Table 50 shows, the p.m. peak eastbound mean speeds of vehicles on the mainline of the freeway decreased for all time periods from 3:00 p.m. to 7:00 p.m. The t-test results indicated statistically significant increases in all mean speed during the periods with the reversible lane.
129 Table 50. Speed comparisonâ(p.m. peak eastbound) without reversible lane vs. with reversible lane. Speed Comparison for EB I-94 Volume Sample Size Mean Speed (mph) 85th Percentile Speed (mph) SD Mean Speed tstatic 1500- 1600 Without Reversible Lane 13783 540 64.66 70.29 9.28 17.43 With Reversible Lane 8486 357 54.81 64.00 7.55 1600- 1700 Without Reversible Lane 16305 540 64.35 69.45 9.57 18.31 With Reversible Lane 8556 349 45.30 63.00 17.86 1700- 1800 Without Reversible Lane 15253 540 65.45 69.65 6.18 22.89 With Reversible Lane 7995 351 40.70 63.00 19.63 1800- 1900 Without Reversible Lane 10979 540 67.39 72.60 6.73 19.54 With Reversible Lane 5710 329 43.44 66.00 21.60 (Bold indicates significance at the 95% confidence level, X = .05) After implementing the reversible lane, the speeds of vehicles on the mainline of the freeway generally increased in the a.m. peak westbound direction. The speeds of vehicles on the mainline of the freeway decreased in the p.m. peak eastbound direction. However, the reversible-lane operation was able to maintain average vehicle speeds at or close to the work zone posted speed limit (45 mph). Therefore, it seems that implementing the reversible lane was able to achieve the mobility goal of the work zone. 7.3.2.3 Comparison of Results for Travel Time The comparison of travel times through the work zone is a good MOE to determine the effect of reversible lanes. The team conducted an analysis of travel time of vehicles on the mainline of the freeway through the project limits of the project area in the peak hours (6:00 a.m. to 10:00 a.m. for WB and 3:00 p.m. to 7:00 p.m. for EB) to determine the effects of implementing the reversible lane. Figure 51 illustrates that travel times through the project limits for westbound direction significantly improved during the period of highest congestion. Figure 52 illustrates that although speeds were lower at the mid-point of the reversible lane, travel time through the project limits (4.6 mi) for the eastbound direction improved.
130 Figure 51. A.M. peak westbound travel time (distance: 4.6 mi). 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 6: 00 :0 0 AM 6: 05 :0 0 AM 6: 10 :0 0 AM 6: 15 :0 0 AM 6: 20 :0 0 AM 6: 25 :0 0 AM 6: 30 :0 0 AM 6: 35 :0 0 AM 6: 40 :0 0 AM 6: 45 :0 0 AM 6: 50 :0 0 AM 6: 55 :0 0 AM 7: 00 :0 0 AM 7: 05 :0 0 AM 7: 10 :0 0 AM 7: 15 :0 0 AM 7: 20 :0 0 AM 7: 25 :0 0 AM 7: 30 :0 0 AM 7: 35 :0 0 AM 7: 40 :0 0 AM 7: 45 :0 0 AM 7: 50 :0 0 AM 7: 55 :0 0 AM 8: 00 :0 0 AM 8: 05 :0 0 AM 8: 10 :0 0 AM 8: 15 :0 0 AM 8: 20 :0 0 AM 8: 25 :0 0 AM 8: 30 :0 0 AM 8: 35 :0 0 AM 8: 40 :0 0 AM 8: 45 :0 0 AM 8: 50 :0 0 AM 8: 55 :0 0 AM 9: 00 :0 0 AM 9: 05 :0 0 AM 9: 10 :0 0 AM 9: 15 :0 0 AM 9: 20 :0 0 AM 9: 25 :0 0 AM 9: 30 :0 0 AM 9: 35 :0 0 AM 9: 40 :0 0 AM 9: 45 :0 0 AM 9: 50 :0 0 AM 9: 55 :0 0 AM Tr av el T im e (M in ut e) Time AM WB Travel Time at Work Zone Travel Time - With Reversible Lane Implementation (Min) Travel Time - Without Reversible Lane Implementation (Min)
131 Figure 52. P.M. peak eastbound travel time (distance: 4.6 mi). 0.0 1.0 2.0 3.0 4.0 5.0 6.0 3: 00 :0 0 PM 3: 05 :0 0 PM 3: 10 :0 0 PM 3: 15 :0 0 PM 3: 20 :0 0 PM 3: 25 :0 0 PM 3: 30 :0 0 PM 3: 35 :0 0 PM 3: 40 :0 0 PM 3: 45 :0 0 PM 3: 50 :0 0 PM 3: 55 :0 0 PM 4: 00 :0 0 PM 4: 05 :0 0 PM 4: 10 :0 0 PM 4: 15 :0 0 PM 4: 20 :0 0 PM 4: 25 :0 0 PM 4: 30 :0 0 PM 4: 35 :0 0 PM 4: 40 :0 0 PM 4: 45 :0 0 PM 4: 50 :0 0 PM 4: 55 :0 0 PM 5: 00 :0 0 PM 5: 05 :0 0 PM 5: 10 :0 0 PM 5: 15 :0 0 PM 5: 20 :0 0 PM 5: 25 :0 0 PM 5: 30 :0 0 PM 5: 35 :0 0 PM 5: 40 :0 0 PM 5: 45 :0 0 PM 5: 50 :0 0 PM 5: 55 :0 0 PM 6: 00 :0 0 PM 6: 05 :0 0 PM 6: 10 :0 0 PM 6: 15 :0 0 PM 6: 20 :0 0 PM 6: 25 :0 0 PM 6: 30 :0 0 PM 6: 35 :0 0 PM 6: 40 :0 0 PM 6: 45 :0 0 PM 6: 50 :0 0 PM 6: 55 :0 0 PM Tr av el T im e (M in ut e) Time PM EB Travel Time at Work Zone Travel Time - With Reversible Lane Implementation (Min) Travel Time - Without Reversible Lane Implementation (Min)
132 7.3.2.4 Statistical Analysis of Travel Time The team used a comparison of travel times to evaluate the effect of the reversible lane and determine if the reversible lane caused changes in travel characteristics. The team used the t- statistic to evaluate the effect of implementing the reversible-lane condition. The team calculated changes in average travel time and the 85th percentile travel time for vehicles traveling through the project limits between the with and without the reversible lane. Tables 51 and 52 show the comparison of average travel time and the 85th percentile speed of travel along with the statistical test results with and without the reversible lane. Table 51 shows that WB average travel time through the project limits generally decreased. The t-test results indicated the changes of travel time during the periods of implementing the reversible lane were statistically significant. Table 51. Travel time comparisonâ(a.m. peak westbound) without reversible lane vs. with reversible lane. Travel Time Comparison for WB I-94 Volume Sample Size Average Travel Time (Min) 85th Percentile Travel Time (Min) SD Mean Travel Time tstatic 0600- 0700 Without Reversible Lane 13200 180 5.23 4.84 3.86 2.31 With Reversible Lane 9748 177 4.57 4.77 0.21 0700- 0800 Without Reversible Lane 12073 180 12.63 19.74 17.26 5.96 With Reversible Lane 12870 175 4.96 5.26 0.38 0800- 0900 Without Reversible Lane 10954 180 7.60 12.28 8.44 4.39 With Reversible Lane 11348 175 4.84 5.03 0.39 0900- 1000 Without Reversible Lane 8679 180 4.31 4.62 0.26 -12.20 With Reversible Lane 8342 176 5.02 5.84 0.73 (Bold indicates significance at the 95% confidence level, X = .05) As Table 52 shows, the EB average travel time through the project limits increased for time periods from 3:00 p.m. to 7:00 p.m. The t-test results indicated the changes in travel time during the periods of implementing the reversible lane were statistically significant.
133 Table 52. Travel time comparisonâ(p.m. peak eastbound) without reversible lane vs. with reversible lane. Travel Time Comparison for EB I-94 Volume Sample Size Average Travel Time (Min) 85th Percentile Travel Time (Min) SD Mean Travel Time tstatic 1500- 1600 Without Reversible Lane 13783 180 4.28 4.63 0.39 -12.09 With Reversible Lane 8486 120 4.96 5.23 0.52 1600- 1700 Without Reversible Lane 16305 180 4.30 4.55 0.32 -13.39 With Reversible Lane 8556 120 5.61 6.14 1.05 1700- 1800 Without Reversible Lane 15253 180 4.23 4.45 0.23 -18.62 With Reversible Lane 7995 120 6.12 6.76 1.10 1800- 1900 Without Reversible Lane 10979 180 4.10 4.32 0.23 -15.15 With Reversible Lane 5710 120 6.17 7.16 1.49 (Bold indicates significance at the 95% confidence level, α = .05) After implementing the reversible lane, the travel time through project limits improved in the morning peak period but showed a slight increase in the afternoon peak period. This increase was most likely attributable to the decreased capacity of the overall work zone due to other temporary lane changes. Although the changes varied, implementing the reversible-lane operation appeared to meet the work zone goal (i.e., maintaining vehicle speeds at or close to the posted speed limit 45 mph). 7.3.2.5 Comparison of Results for Frequency of Headway The team examined the headways accepted by following vehicles to see if there were any differences between the with and without implementation of the reversible lane. The team conducted an analysis of headways of vehicles on the mainline of the freeway at one reversible- lane location to determine the average values and distribution for the a.m. peak westbound and p.m. peak eastbound peak directions. As mentioned above, the team used the K-S test to judge how faithfully a distribution fits the sample data. The K-S test was adopted to determine the goodness-of-fit in the work zone traffic condition. Table 53 summarizes the K-S test results for the a.m. peak westbound direction.
134 Table 53. K-S test results for the a.m. peak westbound direction. A.M. Peak Westbound Direction Baseline Location Reversible-Lane Location Sample Size 2,063 1,906 Mean Headway (seconds) 3.09 3.16 Median Headway (seconds) 2.86 3.00 Maximum difference (D) 0.06 Significance Yes Figure 53 presents a visual performance comparison of headway distributions through a cumulative distribution function in the a.m. peak westbound direction (6:00 a.m. to 10:00 a.m.) at the reversible-lane location. The team observed a slight shift in the headway distribution toward longer headways with the without reversible-lane condition compared to with the reversible-lane condition. The longer headways were observed for the with reversible-lane condition compared to the without reversible-lane condition starting from cumulative percentage at 35% and continuing to about a cumulative 95% of the distribution, with a maximum headway difference of 0.3 sec. The mean value of headway was 3.09 seconds at the without reversible-lane condition as opposed to 3.16 seconds at the with reversible-lane condition. With the significance level α of 0.05 and the sample size more than 40, the critical statistic of K-S test for the maximum difference between the cumulative distributions, D, is 0.04. The results of K-S test for (without reversible lane vs. with reversible lane) shows a value of D of 0.06 (greater than the critical value of 0.04), which suggests the differences in the two cumulative distributions are statistically significant. Table 54 summarizes the K-S test results for the p.m. peak eastbound direction.
135 Figure 53. Cumulative headway distribution plotâ(a.m. peak westbound) without reversible lane vs. with reversible lane (6:00 a.m. to 10:00 a.m.). Table 54. K-S test results for the p.m. peak eastbound direction. P.M. Peak Eastbound Direction Baseline Location Reversible-Lane Location Sample Size 2,145 1,338 Mean Headway (seconds) 2.52 2.89 Median Headway (seconds) 2.40 2.73 Maximum difference (D) 0.17 Significance Yes Figure 54 presents a visual performance comparison of headway distributions through a cumulative distribution function for the p.m. peak eastbound direction (3:00 p.m. to 7:00 p.m.) at the reversible-lane location. The team observed a slight shift in the headway distribution toward longer headways with the reversible-lane condition compared to without the reversible- lane condition. The mean headway value was 2.52 seconds for the without reversible-lane condition, as opposed to 2.89 seconds for the with reversible-lane condition. With the significance level α of 0.05 and the sample size more than 40, the critical statistic of K-S test for the maximum difference between the cumulative distributions, D, is 0.05. The results of K-S test for (without reversible lane vs. with reversible lane) shows a value of D of 0.17 (greater than the 0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% 80.00% 90.00% 100.00% 0 1 2 3 4 5 6 7 8 9 Cu m ul at iv e pe rc en ta ge Headway (Sec) Baseline Condition Reversible Lane Condition
136 critical value of 0.05), which suggests the differences in the two cumulative distributions are statistically significant. Figure 54. Cumulative headway distribution plotâ(p.m. peak eastbound) without reversible lane vs. with reversible lane (3:00 p.m. to 7:00 p.m.). The vehicles headways on the mainline of the freeway at the reversible-lane location exhibited an increase in both a.m. peak westbound direction and p.m. peak eastbound direction. The result of the K-S test indicated that the differences in the two cumulative distributions were statistically significant. It can be reasonably concluded that the headway increase, a positive effect, was a result of implementing the reversible-lane operation. 0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% 80.00% 90.00% 100.00% 0 1 2 3 4 5 6 7 8 9 Cu m ul at iv e pe rc en ta ge Headway (Sec) Baseline Condition Reversible Lane Condition
137 7.3.3 Location: I-75 and I-675, Saginaw and Bay Counties, Michigan 7.3.3.1 Comparison of Results for Vehicle Speeds The team collected data at four locations on I-75âtwo within the work zone (in the reversible- lane configuration) and one each upstream and downstream (outside the work zone). Figure 55 illustrates the data collection locations. Figure 56 illustrates the baseline traffic ADT volumes as well as the volumes carried by the work zone in the reversible-lane configuration for both northbound and southbound directions. The effect of a holiday weekend or recreational traffic was to significantly increase vehicles (>30,000) in the work zone, showcasing the capability of a reversible lane to change on demand. Figure 57 illustrates the hourly traffic volume changes and the capability of the reversible lane to handle these higher volumes.
138 Figure 55. Data collection locations.
139 Figure 56. I-75 daily traffic volumes.
140 Figure 57. I-75 hourly traffic volumes.
141 Figures 58 and 59 illustrate the vehicle speeds and traffic volumes on the mainline for the northbound and southbound directions, respectively. The graphs indicated that I-75 did not have daily peaks; rather, a peak period spans the entire day from about 10:00 a.m. to 7:00 p.m., regardless of direction. In general, work zones will reduce capacity by at least 5% to 15% when the roadway is at or close to capacity due to narrow lanes or shoulders, barrier close to travel lane and other geometric changes. However, by implementing the reversible lane, it appears that capacity was maintained and vehicle speeds were within ±5% of baselines speeds.
142 Figure 58. Northbound average speed and traffic volumes. 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 12 :0 0: 00 A M 12 :3 0: 00 A M 1: 00 :0 0 AM 1: 30 :0 0 AM 2: 00 :0 0 AM 2: 30 :0 0 AM 3: 00 :0 0 AM 3: 30 :0 0 AM 4: 00 :0 0 AM 4: 30 :0 0 AM 5: 00 :0 0 AM 5: 30 :0 0 AM 6: 00 :0 0 AM 6: 30 :0 0 AM 7: 00 :0 0 AM 7: 30 :0 0 AM 8: 00 :0 0 AM 8: 30 :0 0 AM 9: 00 :0 0 AM 9: 30 :0 0 AM 10 :0 0: 00 A M 10 :3 0: 00 A M 11 :0 0: 00 A M 11 :3 0: 00 A M 12 :0 0: 00 P M 12 :3 0: 00 P M 1: 00 :0 0 PM 1: 30 :0 0 PM 2: 00 :0 0 PM 2: 30 :0 0 PM 3: 00 :0 0 PM 3: 30 :0 0 PM 4: 00 :0 0 PM 4: 30 :0 0 PM 5: 00 :0 0 PM 5: 30 :0 0 PM 6: 00 :0 0 PM 6: 30 :0 0 PM 7: 00 :0 0 PM 7: 30 :0 0 PM 8: 00 :0 0 PM 8: 30 :0 0 PM 9: 00 :0 0 PM 9: 30 :0 0 PM 10 :0 0: 00 P M 10 :3 0: 00 P M 11 :0 0: 00 P M 11 :3 0: 00 P M Vo lu m e (V eh icl e/ M in ) Sp ee d (M PH ) Time NB Average Speed and Traffic Volumes Reversible Lane Location vs Baseline Location, 1 Minute Bin) Reversible Lane Location - WorkZone (Speed) - 3 day-Average Baseline Location - Baseline (Speed) - 3 day-Average Reversible Lane Location - WorkZone (Volume) - 3 day-Average Baseline Location - Baseline (Volume) - 3 day-Average Poly. (Reversible Lane Location - WorkZone (Volume) - 3 day-Average ) Poly. (Baseline Location - Baseline (Volume) - 3 day-Average )
143 Figure 59. Southbound average speed and traffic volumes. 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 12 :0 0: 00 A M 12 :3 0: 00 A M 1: 00 :0 0 AM 1: 30 :0 0 AM 2: 00 :0 0 AM 2: 30 :0 0 AM 3: 00 :0 0 AM 3: 30 :0 0 AM 4: 00 :0 0 AM 4: 30 :0 0 AM 5: 00 :0 0 AM 5: 30 :0 0 AM 6: 00 :0 0 AM 6: 30 :0 0 AM 7: 00 :0 0 AM 7: 30 :0 0 AM 8: 00 :0 0 AM 8: 30 :0 0 AM 9: 00 :0 0 AM 9: 30 :0 0 AM 10 :0 0: 00 A M 10 :3 0: 00 A M 11 :0 0: 00 A M 11 :3 0: 00 A M 12 :0 0: 00 P M 12 :3 0: 00 P M 1: 00 :0 0 PM 1: 30 :0 0 PM 2: 00 :0 0 PM 2: 30 :0 0 PM 3: 00 :0 0 PM 3: 30 :0 0 PM 4: 00 :0 0 PM 4: 30 :0 0 PM 5: 00 :0 0 PM 5: 30 :0 0 PM 6: 00 :0 0 PM 6: 30 :0 0 PM 7: 00 :0 0 PM 7: 30 :0 0 PM 8: 00 :0 0 PM 8: 30 :0 0 PM 9: 00 :0 0 PM 9: 30 :0 0 PM 10 :0 0: 00 P M 10 :3 0: 00 P M 11 :0 0: 00 P M 11 :3 0: 00 P M Vo lu m e (V eh icl e/ M in ) Sp ee d (M PH ) Time SB Average Speed and Traffic Volumes (Reversible Lane Location vs Baseline Location, 1 Minute Bin) Reversible Lane Location - WorkZone (Speed) - 3 day-Average Baseline Location - Baseline (Speed) - 3 day-Average Reversible Lane Location - WorkZone (Volume) - 3 day-Average Baseline Location - Baseline (Volume) - 3 day-Average Poly. (Reversible Lane Location - WorkZone (Volume) - 3 day-Average ) Poly. (Baseline Location - Baseline (Volume) - 3 day-Average )
144 7.3.3.2 Statistical Analysis of Vehicle Speed The team calculated changes in the mean speeds and 85th percentile speeds for vehicles between with and without implementing the reversible lane. Tables 55 and 56 show the comparison of mean speed and 85th percentile speed of vehicles on the freeway mainline and the statistical test results with and without the implementation of the reversible lane. As Table 55 shows, the mean speeds of vehicles on the northbound mainline of the freeway increased for all time periods. Table 55. Speed comparisonânorthbound direction baseline location vs. reversible-lane location. Speed Comparison for NB I-75 Volume Sample Size Mean Speed (mph) 85th Percentile Speed (mph) SD Mean Speed tstatic 1200- 1300 Baseline 11134 523 64.71 73.00 9.55 -10.17 Reversible Lane 11916 550 69.42 75.00 4.72 1300- 1400 Baseline 12071 523 61.15 73.00 11.89 -12.55 Reversible Lane 12677 544 68.24 74.00 5.11 1400- 1500 Baseline 12251 524 59.22 73.00 12.13 -11.86 Reversible Lane 13063 543 66.14 72.00 5.69 1500- 1600 Baseline 11169 528 62.69 76.00 14.90 -8.84 Reversible Lane 12411 547 68.89 76.00 6.28 (Bold indicates significance at the 95% confidence level, α = .05) The t-test results indicated the increases in mean speed during the periods of implementing the reversible lane were statistically significant for all time periods. As Table 56 shows, the mean speeds of vehicles on the southbound mainline of the freeway increased for all time periods.
145 Table 56. Speed comparisonâsouthbound direction baseline location vs. reversible-lane location. Speed Comparison for SB I-75 Volume Sample Size Mean Speed (mph) 85th Percentile Speed (mph) SD Mean Speed tstatic 1200- 1300 Baseline 10670 534 69.25 74.00 4.18 -0.65 Reversible Lane 11361 550 69.42 75.00 4.72 1300- 1400 Baseline 10655 528 64.50 74.00 13.68 -5.89 Reversible Lane 11282 544 68.24 74.00 5.11 1400- 1500 Baseline 10339 531 56.75 74.00 20.12 -10.36 Reversible Lane 10909 543 66.14 72.00 5.69 1500- 1600 Baseline 9025 531 67.26 75.00 11.56 -2.86 Reversible Lane 9500 547 68.89 76.00 6.28 (Bold indicates significance at the 95%confidence level, α = .05) The t-test results indicated the increases in mean speed during the periods of implementing the reversible lane were statistically significant for time periods from 1:00 p.m. to 4:00 p.m. After implementing the reversible lane, the speeds of vehicles on the mainline of the freeway increased in both northbound and southbound peak direction periods. Therefore, it seems that implementing the reversible lane had a positive effect. 7.3.3.3 Comparison of Results for Travel Time Comparing the travel times through a work zone is a good MOE to determine the effect of the implementing reversible-lane operations. The reversible lane provides capacity to the mainline of the freeway and it is expected to maintain vehicle speeds, which will result in travel times similar to the baseline condition. Figures 60 and 61 illustrate average 3-day travel times through the project limits for northbound and southbound peak direction periods, respectively. In general, travel times were maintained across the daily peak periods.
146 Figure 60. Northbound average travel time (distance: 7.3 mi). 0.0 2.0 4.0 6.0 8.0 10.0 12.0 12 :0 0: 00 A M 12 :2 8: 00 A M 12 :5 6: 00 A M 1: 24 :0 0 AM 1: 52 :0 0 AM 2: 20 :0 0 AM 2: 48 :0 0 AM 3: 16 :0 0 AM 3: 44 :0 0 AM 4: 12 :0 0 AM 4: 40 :0 0 AM 5: 08 :0 0 AM 5: 36 :0 0 AM 6: 04 :0 0 AM 6: 32 :0 0 AM 7: 00 :0 0 AM 7: 28 :0 0 AM 7: 56 :0 0 AM 8: 24 :0 0 AM 8: 52 :0 0 AM 9: 20 :0 0 AM 9: 48 :0 0 AM 10 :1 6: 00 A M 10 :4 4: 00 A M 11 :1 2: 00 A M 11 :4 0: 00 A M 12 :0 8: 00 P M 12 :3 6: 00 P M 1: 04 :0 0 PM 1: 32 :0 0 PM 2: 00 :0 0 PM 2: 28 :0 0 PM 2: 56 :0 0 PM 3: 24 :0 0 PM 3: 52 :0 0 PM 4: 20 :0 0 PM 4: 48 :0 0 PM 5: 16 :0 0 PM 5: 44 :0 0 PM 6: 12 :0 0 PM 6: 40 :0 0 PM 7: 08 :0 0 PM 7: 36 :0 0 PM 8: 04 :0 0 PM 8: 32 :0 0 PM 9: 00 :0 0 PM 9: 28 :0 0 PM 9: 56 :0 0 PM 10 :2 4: 00 P M 10 :5 2: 00 P M 11 :2 0: 00 P M 11 :4 8: 00 P M Tr av el T im e (M in ut e) Time NB Travel Time (Through Project Limits) Travel Time - Reversible Lane Location (Min) - 3 Day Average Travel Time - Baseline Location (Min) - 3 Day Average
147 Figure 61. Southbound average travel time (distance: 7.3 mi). 0.0 2.0 4.0 6.0 8.0 10.0 12.0 12 :0 0: 00 A M 12 :2 8: 00 A M 12 :5 6: 00 A M 1: 24 :0 0 AM 1: 52 :0 0 AM 2: 20 :0 0 AM 2: 48 :0 0 AM 3: 16 :0 0 AM 3: 44 :0 0 AM 4: 12 :0 0 AM 4: 40 :0 0 AM 5: 08 :0 0 AM 5: 36 :0 0 AM 6: 04 :0 0 AM 6: 32 :0 0 AM 7: 00 :0 0 AM 7: 28 :0 0 AM 7: 56 :0 0 AM 8: 24 :0 0 AM 8: 52 :0 0 AM 9: 20 :0 0 AM 9: 48 :0 0 AM 10 :1 6: 00 A M 10 :4 4: 00 A M 11 :1 2: 00 A M 11 :4 0: 00 A M 12 :0 8: 00 P M 12 :3 6: 00 P M 1: 04 :0 0 PM 1: 32 :0 0 PM 2: 00 :0 0 PM 2: 28 :0 0 PM 2: 56 :0 0 PM 3: 24 :0 0 PM 3: 52 :0 0 PM 4: 20 :0 0 PM 4: 48 :0 0 PM 5: 16 :0 0 PM 5: 44 :0 0 PM 6: 12 :0 0 PM 6: 40 :0 0 PM 7: 08 :0 0 PM 7: 36 :0 0 PM 8: 04 :0 0 PM 8: 32 :0 0 PM 9: 00 :0 0 PM 9: 28 :0 0 PM 9: 56 :0 0 PM 10 :2 4: 00 P M 10 :5 2: 00 P M 11 :2 0: 00 P M 11 :4 8: 00 P M Tr av el T im e (M in ut e) Time SB Travel Time (Through Project Limits) Travel Time - Reversible Lane Location (Min) - 3 Day Average Travel Time - Baseline Location (Min) - 3 Day Average
148 7.3.3.4 Statistical Analysis of Travel Time The team used a comparison of travel times to evaluate the effect of the reversible lane and determine if the reversible lane caused changes in travel characteristics. The team employed the t-statistic to evaluate the effect of implementing the reversible lane. The team calculated changes in average travel time and the 85th percentile travel time for vehicles traveling through the project limits between the with and without reversible lane implementation. Tables 57 and 58 show the comparison of average travel times and the 85th percentile travel time and the statistical test results with and without the reversible lane. As Table 57 shows, the NB average travel time through the project limits decreased for all time periods. The t-test results indicated that decreases in travel time during the periods of implementing the reversible lane were statistically significant. Table 57. Travel time comparisonânorthbound direction baseline location vs. reversible-lane location. Travel Time Comparison for NB I-75 Volume Sample Size Average Travel Time (Min) 85th Percentile Travel Time (Min) SD Mean Travel Time tstatic 1200- 1300 Baseline 11134 180 6.85 8.24 1.28 5.86 Reversible Lane 11916 180 6.28 6.50 0.23 1300- 1400 Baseline 12071 180 7.49 8.53 2.33 6.15 Reversible Lane 12677 180 6.42 6.76 0.33 1400- 1500 Baseline 12251 180 7.83 9.20 2.26 6.00 Reversible Lane 13063 180 6.73 7.07 0.95 1500- 1600 Baseline 11169 180 7.54 9.61 2.67 5.68 Reversible Lane 12411 180 6.39 6.92 0.53 (Bold indicates significance at the 95% confidence level, α = .05) Table 58 shows that the SB average travel times in general were reduced. The t-test results indicated the decreases in travel time during the periods of implementing the reversible lane were statistically significant for all time periods from 12:00 p.m. to 4:00 p.m.
149 Table 58. Travel time comparisonâsouthbound direction baseline location vs. reversible-lane location. Travel Time Comparison for SB I-75 Volume Sample Size Average Travel Time (Min) 85th Percentile Travel Time (Min) SD Mean Travel Time tstatic 1200- 1300 Baseline 10670 180 6.26 6.46 0.22 -5.55 Reversible Lane 11361 180 6.80 7.51 1.30 1300- 1400 Baseline 10655 180 7.45 6.91 3.65 1.83 Reversible Lane 11282 180 6.92 8.15 1.27 1400- 1500 Baseline 10339 180 9.45 15.49 5.60 5.83 Reversible Lane 10909 180 6.94 8.93 1.44 1500- 1600 Baseline 9025 180 6.87 6.45 2.41 4.12 Reversible Lane 9500 180 6.13 6.33 0.19 (Bold indicates significance at the 95% confidence level, α = .05) After implementing the reversible lane, the travel times through project limits decreased in most of the time periods for both northbound and southbound peak direction periods. The travel time savings were similar and statistically significant. It can be reasonably concluded that improving travel time is the result of implementing the reversible lane. Therefore, implementing the reversible lane seems to have a positive effect on travel time. 7.3.3.5 Comparison of Results for Frequency of Headway Headway is a good measure of congestion and lack of passing opportunities created by the traffic mix; it is also a good measure of safety as lane changing and frequent passing generally lead to conflicts and the likelihood of crashes. In general, a longer headway accepted by a merging vehicle is safer than a shorter headway. The team examined the headways accepted by following vehicles to see if there were any differences between the with and without implementation of the reversible lane. The team conducted an analysis of headways of vehicles on the mainline of the freeway at the baseline location and the reversible-lane location to determine the average values and distribution for both northbound and southbound peak direction periods. As mentioned, above, the team used the K-S test to judge how faithfully a distribution fits the sample data. The K-S test was adopted to determine the goodness-of-fit in the work zone traffic condition. Table 59 summarizes the K-S test results of the northbound peak direction vs. reversible-lane location.
150 Table 59. K-S test results for the northbound peak direction period. Northbound Peak Direction Period Baseline Location Reversible-Lane Location Sample Size 1,970 2,140 Mean Headway (seconds) 2.85 2.76 Median Headway (seconds) 2.61 2.50 Maximum difference (D) 0.03 Significance No Figure 62 presents a visual performance comparison of headway distributions through a cumulative distribution function in the peak hours (12:00 p.m. to 4:00 p.m.) at baseline and reversible-lane location. The team observed a slight shift in the headway distribution toward longer headway with the with reversible-lane condition compared to without the reversible-lane condition. The mean value of headway was 2.85 seconds at the without reversible-lane condition as opposed to 2.76 seconds at the with reversible-lane condition. With the significance level α of 0.05 and the sample size more than 40, the critical statistic of K-S test for the maximum difference between the cumulative distributions, D, was 0.04. The results of K-S test for (without reversible lane vs. with reversible lane) shows a value of D of 0.03 (less than the critical value of 0.04), which suggests the differences in the two cumulative distributions are not statistically significant. Figure 62. Cumulative headway distribution plotânorthbound peak direction periodâ baseline location vs. reversible-lane location (12:00 p.m. to 4:00 p.m.). 0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% 80.00% 90.00% 100.00% 0 1 2 3 4 5 6 7 8 9 Cu m ul at iv e pe rc en ta ge Headway (Sec) Baseline Location Reversible Lane Location
151 Table 60 summarizes the K-S test results of the southbound peak direction vs. reversible- lane location. Table 60. K-S test results for the southbound peak direction period. Southbound Peak Direction Period Baseline Location Reversible-Lane Location Sample Size 1,905 2,056 Mean Headway (seconds) 3.31 3.25 Median Headway (seconds) 3.00 3.00 Maximum difference (D) 0.035 Significance No Figure 63 presents a visual performance comparison of headway distributions through a cumulative distribution function in the peak hours (12:00 p.m. to 4:00 p.m.) at baseline and reversible-lane location. The team observed a slight shift in the headway distribution toward shorter headway at reversible-lane location compared to baseline location. The mean value of headway was 3.31 seconds at baseline location as opposed to 3.25 seconds at reversible-lane location. With the significance level α of 0.05 and the sample size more than 40, the critical statistic of K-S test for the maximum difference between the cumulative distributions, D, is 0.04. The results of K-S test for (baseline location vs. reversible-lane location) shows a value of D of 0.035 (less than the critical value of 0.04), which suggests the differences in the two cumulative distributions are not statistically significant.
152 Figure 63. Cumulative headway distribution plotâsouthbound peak direction periodâ baseline location vs. reversible-lane location (12:00 p.m. to 4:00 p.m.). The headways of vehicles on the mainline of the freeway at the reversible-lane location exhibited an increase in both northbound peak direction and southbound peak direction periods. The result of the K-S test indicated that the differences in the two cumulative distributions were not statistically significant. 7.4. Effects of Reversible Lane Operation on Traffic flow in Work Zones All work zones at the three sites were considered long term and in place for more than four months. A concrete barrier separates the reversible lane from other lanes, which means that there was no potential for diversion to on and off ramps and almost no access to or from the work zone activities. The following discussion is an attempt to understand the relationship between a reversible-lane operation as part of the work zone strategy and traffic flow characteristics. Speed-flow graphs are shown followed by test site and for all lanes by direction. Note that the reversible lane (#3) is shown twice (i.e., by direction when changed over). The estimated volumes per lane in the work zones were significantly higher than that found using Highway Capacity Manual 2010 (HCM 2010), which ranged from 1500+ to 1700+. Capacity (Qmax) in the non-work zone condition ranged from 1,400 to 1,800 compared to the work zone where the range was between 1,600 to 2,250 vehicles per lane. In typical congested conditions, the work zone capacity is lower than non-work zone conditions; however, the data collected across all sites clearly demonstrated that using a reversible-lane configuration can maintain or 0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% 80.00% 90.00% 100.00% 0 1 2 3 4 5 6 7 8 9 Cu m ul at iv e pe rc en ta ge Headway (Sec) Baseline Location Reversible Lane Location
153 increase the directional capacity, regardless of whether it supports daily (commuting) or weekday (holiday/recreational) traffic. The reversible lane in all lanes evaluated were separated from opposing traffic with concrete barriers. Figures 64â66 illustrate speed-flow characteristics by lane by direction for each test site (3âreversible lane, 2âcenter lane, 1âright lane). Figure 64. I-75 in Saginaw County, Michigan, speed-flow plots. Reversible-lane changeover took place during weekends and mid-week to accommodate recreational and holiday traffic.
154 Figure 65. I-75 and I-675 in Saginaw and Bay Counties, Michigan, speed-flow plots. Reversible-lane changeover took place during weekends and mid-week to accommodate recreational and holiday traffic.
155 Figure 66. I-94 in Maplewood, Minnesota, speed-flow plots. Reversible-lane changeover took place after the morning peak period to accommodate afternoon commuting traffic.
156 7.5. Work Zone Crash Modification Factor for Reversible Lane Table 61 shows the expected and actual crash results for deploying a reversible lane. The Total Hours column indicates the number of hours of data analyzed. Table 61. Expected and actual crash results for reversible lane Treatment Total Hours Expected Crashes Actual Crashes Percent Change Reversible-Lane Deployment 6600 132.5 97 -27 For the reversible-lane condition, there was a 27% decrease from expected to actual crashes. In order to determine the proportional effects of the treatments on the numbers of crashes, an odds ratio analysis was undertaken according to the following equations: Where: CMFD = crash modification factor = proportional effect of a deployment on crashes: TAD = total actual crashes during a deployment (equal to 97 in this case); TED = total expected crashes during a deployment (equal to 132.5 in this case); TAND = total actual crashes when nothing was deployed (equal to 81 in this case); TEND = total expected crashes when nothing was deployed (equal to 117.1 in this case); and SD (CMFD) = standard error. Table 62 shows the results from the CMF calculation. The calculated CMF for reversible lanes approximates 1, indicating that this treatment had no effect on reducing the number of crashes, without taking standard error into account.
157 Table 62. CMF results for reversible lane. Treatment CMFD SE(CMFD) ADT Reversible Lane (Interstate) 1.029 0.200 Up to 100,000 Vehicles The CMF is limited because of the few test sites. Agencies should only use this as a guide, monitor all work zones, and take appropriate action to mitigate any increase in crashes (i.e., severity and number). A general review of the data at these sites indicate that the upstream taper and the initial entry to the work zone appears to pose the most risk for drivers when in a reversible-lane configuration. Additional signing and pavement markings on the approach to the taper can help mitigate this issue.
