Evaluating Strategies for Work Zone Transportation Management Plans (2020)

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.

158 8.0 Summary of Findings 8.1. Standalone Guidebook Although there is a wealth of information, it is scattered among published research, DOT handbooks, manuals, plans, as well as unpublished documentation. This project developed a TPM Strategy Guidebook, published as NCHRP Research Report 945, that synthesizes useful knowledge from all these diverse sources to create a work zone guidebook. The guidebook provides a compendium of current knowledge on work zone strategies, including suggestions on when to use, benefits, effectiveness, technical issues, design requirements, state of the practice, and cost. 8.2. Field Evaluations Summary of Results This section summarizes field evaluation findings, conclusions, and general considerations for future research. The summaries reference the appropriate NCHRP 03-111 sections: Section 5- Field Evaluations of Truck Lane Restrictions; Section 6-Field Evaluation of Temporary Ramp Metering; and Section 7-Field Evaluation of Reversible Lanes, respectively. 8.2.1 Field Evaluation of Truck Lane Restrictions The team conducted field evaluations of the effectiveness of work zone truck lane restrictions on lane distribution and operations at three work zone sites in Michigan. • Lane Distribution of Trucks. The primary interest in evaluating the effectiveness of truck lane restrictions was the percentage changes of the lane distribution of trucks at the test sites. The team compared the without and with percentages of lane distribution of trucks. All three sites restricted trucks to using the left lane and the data clearly show that the truck lane restrictions effectively created a tangible increase in the number of trucks using the left lane. o With the truck restrictions in place, the percentage change in trucks using the left lane, for all time periods, increased by 84.76% for SB I-75, 502.64% for SB US-23, and 669.04% for NB US-23. For all sites combined, the percentage change in trucks using the left lane, for all time periods, increased by 234.96%. o With truck restrictions in place, the percentage change in trucks using the right lane, for all time periods, decreased by 71.74% for SB I-75, 48.31% for SB US-23, and 51.32% for NB US-23. For all sites combined, the percentage change in trucks using the right lane, for all time periods, decreased by 59.36%. • Effect of Truck Lane Restrictions on Average Vehicle Speeds. The team compared average vehicle speeds to determine if lane-use restrictions caused changes in vehicle

159 speeds. The vehicle speeds were compared separately according to the time period— morning peak period (6:00 to 9:00 a.m.), mid-day period (10:00 a.m. to 1:00 p.m.), and evening peak period (3:00 to 6:00 p.m.) for each site o Average truck speeds were reduced in the right lane at two test sites and increased at the third test site. Average truck speeds increased in the left lane at one test site and decreased at two test sites. o Average passenger car speeds were reduced in the right lane at two test sites and increased at a third test site. Average passenger car speeds increased in the left lane at one test site and decreased at two test sites. Across the three study sites, the overall average truck speeds reduced by approximately 3 mph (5%) with the truck lane restrictions. • Effect on Frequency of Headway. The comparisons of the headways of vehicles on the mainline left lane of the freeway during with and without conditions improved. Lower headways (less than 300 feet) improved between 19%–66%. Headways on the mainline right lane of the freeway saw no improvements but were greater than 300 feet. • Effect on Platoon Headways and Gap Acceptance. The team examined the number of instances where a vehicle leads a platoon of traffic. A platoon is defined as a vehicle traveling with a headway greater than 3 seconds, followed by one or more vehicles with a headway less than 3 seconds. In this analysis, the team evaluated the headway for different vehicle leader–follower pairs (1) car followed by a car or truck (C-C and C-T) and (2) a truck followed by a car or truck (T-C and T-T). The team analyzed headways between vehicles to determine what changes, if any, occurred between the with and without periods. It was hypothesized that car drivers might feel impeded in the left lane because of the increase in the number of trucks and would try to pass slow-moving vehicles by moving into the right lane. The results showed that in the left lane, the headway for a truck following a car or truck increased and both groups decreased in the right lane. These results are in line with the hypothesis and did not substantiate the theory that the restrictions that would cause trucks to bear down on cars more frequently (i.e., restricting trucks to a particular lane decreases safety was not substantiated). • Estimated cost to implement dedicated truck lanes is $15,000–$25,000 (static signs plus PCMS) over 5 mi.

160 Conditions most conducive to favorable application of truck lane restrictions are freeways with two or more lanes in each direction, interchanges spaced more than 2 to 3 mi apart, and with low ramp volumes and truck percentages between 10% to 25% of the total traffic stream. Results also suggest that where crash data are available, transportation agencies conduct the before-and-after evaluations of the safety characteristics. Compliance requires routine enforcement by regular traffic patrols or specialized dedicated truck-enforcement units. Agencies are encouraged to undertake a comprehensive public information campaign about the restriction and inform the public and the trucking community along the corridor to ensure success of the project. 8.2.2 Field Evaluation of Temporary Ramp Metering The team used with–without ramp-metering studies to evaluate the effectiveness of ramp metering at the work zones on MN Route 52 in Rochester, Minnesota, and on I-279 in Ohio Township, Pennsylvania. The team implemented two ramp metering scenarios during the study period and evaluated fixed-cycle and variable-cycle lengths. The team also performed operational and driver behavior evaluations. The following paragraphs summarize the evaluation findings: • The team compared vehicle speeds on the mainline of the freeway to evaluate the effect of implementing ramp metering. The results indicated that the speeds of vehicles on the mainline increased in both ramp-metering scenarios. The t-test results indicated the increases in mean speed in both ramp-metering scenarios were statistically significant and, thus, it can be reasonably concluded that implementing ramp metering seems to have a positive effect on vehicle speeds on the mainline. Overall, under saturated conditions, Fixed-cycle Length Ramp Metering performed slightly better than Variable- cycle Length Ramp Metering, with speed increases of 8.6 mph and 5.18 mph, respectively. When the mainline is less than 80% saturated, Variable-cycle Length Ramp Metering performed better than Fixed-cycle Length Ramp Metering with speed increases of 11–14 mph (right lane/left lane) and 5–8 mph (right lane/left lane), respectively. Left lane in each scenario experienced the larger increase as less vehicles attempt to merge. • The team compared travel time through the work zones to determine the operational effect of implementing ramp metering. The results indicated that the travel time became shorter in both ramp-metering scenarios. The t-test results indicated the decreases in travel time in both ramp-metering scenarios are statistically significant and, thus, it can be reasonably concluded that implementing ramp metering seems to have a positive effect in travel time.

161 • The team also analyzed headways between vehicles on the mainline of the freeway at the merging area of ramp and mainline to determine the changes, if any, that occurred between with and without implementing ramp metering. The results were somewhat different for the two study areas, primarily because of roadway capacity condition. For MN Route 52 in Rochester, Minnesota, the results showed that headways of vehicles on the mainline increased in both ramp-metering scenarios (Fixed-cycle Length Ramp Metering = 2.42 seconds; Variable-cycle Length Ramp Metering = 2.44 seconds) from Meter-off scenario (2.3 seconds); however, the K-S test results indicated that the differences in the two cumulative distributions were not statistically significant. In general, the longer headway at the merging area is safer than a shorter headway. For I-279 in Ohio Township, Pennsylvania, the results showed that headways of vehicles on the mainline slightly decreased in both ramp-metering scenarios (from Meter-off scenario for both right lane and left lane). The K-S test results indicated that the differences in the two cumulative distributions were statistically significant for three scenario comparisons (Right lane–Meter-off vs. Variable-cycle Length; Right lane–Meter- off vs. Fixed-cycle Length; Left lane–Meter-off vs. Variable-cycle Length), and not statistically significant for one scenario-comparison (Right lane–Meter-off vs. Fixed- cycle). The reason for the decrease of headway on the I-279 project may resulted from the increase of the mainline traffic by approximately 10% to 20% that was directly related to implementing the ramp-metering strategy. Regardless, average headways were greater than 2.4 seconds across all time periods. • Based on the observation of video for morning peak period (7:30 to 8:30 a.m.), the team determined that driver compliance rate for Variable-cycle Length Ramp Metering was higher than Fixed-cycle Length Ramp Metering and the compliance rate range was between 60% to 90%. This is considered good-to-excellent, as it was not accompanied by any type of enforcement. • Once ramp metering is installed, it is estimated that the pay-off period will be approximately 3 to 5 months, depending on the level of congestion. (i.e., longer-term work zone greater than 5 months will yield significantly positive cost savings. Including crash costs and other savings would yield a shorter pay-off period. Overall, this first evaluation of ramp metering in work zones during peak conditions proved very successful. The main lesson is to determine time of saturation and set the ramp metering at least 15–30 minutes prior. The traditional ramp-metering design volume criteria will not work in work zone conditions. Through this evaluation, the team recommends that the total lane/ramp vehicle volumes should not be greater than 1,600 vph (ramp volumes should not

162 exceed 400–600 vph). An ideal traffic volume range would be closer to 1,400 combined vph per lane with the ramp volumes not to exceed 600 vehicles per hour for maximum effectiveness. 8.2.3 Field Evaluation of Reversible Lanes The team used with–without reversible lane studies to evaluate the effectiveness of the reversible lane at work zones on three project sites. The team performed operational evaluations. The following is a summary of the evaluation findings: • The team compared vehicle speeds on the mainline of the freeway to evaluate the effect of implementing the reversible lane. The results indicated that the speeds of vehicles on the mainline were generally maintained across all test sites. • The team also compared travel time through work zone to determine the effect of implementing the reversible lane. The results indicated that the travel time became shorter in most of the time periods analyzed at three project sites. The t-test results indicated the decreases in travel time for most of the time periods analyzed were statistically significant. On average travel time improvements across all sites ranged from 5.6% to 15%. Thus, it can be reasonably concluded that implementing the reversible-lane operation seems to have a positive effect on travel time. • The team analyzed headways between vehicles on the mainline of the freeway at the reversible-lane location and baseline location to determine the changes, if any, that occurred between with and without implementation of the reversible lane. The results showed that the headways in the reversible-lane configuration on a few occasions decreased because of the significant increase in traffic volumes; however, the minimum average headway was 2.7 seconds. • Key to a successful reversible-lane operation is understanding the traffic flow pattern, daily and weekday, and knowing when to change over the lanes. Operation must be flexible enough to adjust to changes in demand. • The reversible lane did not carry less traffic than other lanes—as previously thought— with a maximum traffic flow per lane from 1,600 to 2,250 vph. • The capacity reduction factor for reversible-lane operation appeared to be 0.90 to 1.20; the latter occurred in cases where the reversible-lane operation was within barriers and not affected by ramps and other merging traffic. • The number of crashes were higher when compared to a non-work zone condition, but less than expected for a work zone condition. Advanced signs and pavement markings on the approach to the taper can help enhance the operation of the reversible lane.

163 8.3. Suggestions for Future Research The field evaluations conducted as part of this research effort are the first of their kind in a work zone setting and act as a good starting point for future research. Future research could add to the existing study by including work zones in different settings (congested, uncongested, rural, arterial freeways, etc.). Additional studies could also be used to validate the CMFs.