Institute for Transportation

About the research

Transportation agencies strive to mitigate the safety and mobility impacts of work zones using a variety of strategies as part of work zone transportation management plans. Closing one side of a divided multilane highway and crossing over traffic to the other side as a two-way operation is being used more frequently by transportation agencies. Work zones need to be designed within a given paved width, especially with counter-directional flow when traffic is crossed over to opposing lanes. The challenge faced by designers is in understanding the safety and mobility implications of the allocation of lanes and shy distances to barriers for a given paved width. For example, if a paved width of 26 ft were available, would it be better to have two 12-ft lanes with 1-ft shy distances or two 11-ft lanes with 2-ft shy distances?

The recently completed SWZDI project: Mobility and Safety Impacts of Work Zone Lane and Shoulder Widths attempted to address this question. While the project demonstrated that mobility and safety impacts of lane/shoulder widths in work zones can be modeled, it also had some limitations. The primary limitation was that safety analysis only considered right-side departures of vehicles in the right lane. Furthermore, the impact on vehicles straddling lanes was not quantified and all the locations with 1-ft shy distances were very short sections. Hence, there is a need to build upon the previous research to address these limitations and provide a more complete understanding of the safety and mobility impacts of lane and shoulder widths.

The current research will achieve the following objectives:

• Collect data from work zones with different lane widths and shy distances.
• Quantify mobility and safety impacts of the different combinations. Mobility performance measures (PMs) will include speed, speed-flow curves, and impact of enforcement or speed feedback signs. Safety PMs will include probabilities of edge line encroachment/barrier contact, and lane straddling. Impact of presence of adjacent vehicles on lane position and speed will also be evaluated.

About the research

The objective of this research project is to develop enhanced guidelines to help transportation practitioners develop temporary traffic control plans for highway construction projects with alternative intersections. Attainment of the project objective will help to fill gaps in existing knowledge and provide transportation practitioners with tools to help improve mobility and safety in work zones on highway construction projects with alternative intersections. The research approach will include a literature review and the gathering of information from various states regarding their best practices for implementing works zones on projects with alternative intersections.

About the research

Nighttime work zones pose increased risks for both drivers and workers due to the challenge of ensuring adequate visibility while minimizing glare. Current lighting practices vary widely across public agencies, often resulting in excessive illumination and poor glare control. This study evaluated the performance of nighttime lighting systems through two phases: (1) an industry survey and (2) a controlled field experiment along with a qualitative task assessment and analysis. The survey, with 116 responses, identified persistent issues related to glare and uneven light coverage, primarily attributed to improper lighting configurations rather than insufficient light output. The field study tested 126 lighting setups using LED and halogen lighting sources set at varying mounting heights, aiming angles, and rotation angles. Measurements included horizontal and vertical illuminance, pavement luminance, and veiling luminance ratio (VLR). Results showed that the lighting setups—rotation angle and mounting height—had the greatest impact on visibility and glare. Optimal configurations (rotation: 40°–50°, height: 12–13 ft, aiming angle: 20°–40°) could provide better visibility and minimize harmful glare in the vicinity of nighttime work zones. The report provides recommendations for optimal lighting configurations, offering agencies practical guidance to improve safety and visibility in nighttime work zones.

About the research

The late merge, or “zipper” merge, has become a common strategy to increase work zone capacity by encouraging drivers to stay in their lanes until they reach a defined merge area, where they alternately merge. While the zipper merge has been shown to provide improved operational performance, there is considerable variability in driver familiarity and behavior when encountering the zipper merge. In addition to determining where and when merging should occur, agencies also determine whether lane merge control is static or dynamic. This study provides insights into the use and efficacy of various types of lane merge control strategies. The research included a literature review, a state department of transportation (DOT) survey, a road user survey, and field evaluations conducted at several freeway work zones in Michigan and Missouri. The state DOT survey found that 93% of agencies use static lane merge compared to only 40% that use dynamic lane merge, the latter of which is more widely used in urban freeways than rural freeways. Various factors are considered when deciding whether to use a dynamic lane merge, including annual average daily traffic, peak hour volumes, and duration of work. The road user surveys showed that drivers would typically merge closer to the taper under zipper merge lane control compared to early merge; however, compliance with this strategy increased significantly when a portable changeable messages sign (PCMS) was used as a supplementary device. Drivers also indicated that traffic signs with textual messages, either with or without supplementary graphical messages, were preferred over graphical-only signs. Driver merging behavior was also found to vary depending upon both the merging strategy (i.e., early versus late/zipper) and the vehicle’s location with respect to the start of the taper. Interestingly, there was significant variability in respondents’ perceptions of whether the zipper merge impacts traffic safety and operations. Familiarity and comfort with the zipper merge were strong determinants of driver behavior and perceptions. The results suggest that outreach campaigns may help to raise awareness of the zipper merge. The field evaluations showed that the zipper merge tended to result in better use of available capacity. However, at low volumes, drivers tend to merge earlier without any adverse impact on operations. For example, the M-53 study location (average volume of 940 vehicles/hour) had utilization rates of less than 15% on average but did not experience any substantive negative impacts on operations. The field evaluations also assessed the impact of varying the location of the PCMS within the work zone, including near the taper and one mile upstream of the taper. It is recommended from the field evaluations that if only a single PCMS is to be used, it should be positioned nearly one mile upstream of the taper displaying USE BOTH LANES/DURING BACKUPS. If an additional PCMS is available, it is recommended to be positioned within 1,000 ft upstream of the lane closure displaying MERGE HERE/TAKE TURNS.

About the research

This project designed, deployed, and validated a statewide platform for real-time incident detection, daily operations analysis, and data-driven support for smart work zone planning known as ReactorIQ. Unlike traditional systems that depend on fixed roadside sensors, manual contractor logs, or 511 updates, ReactorIQ is built around high-resolution connected vehicle (CV) telemetry as its primary data source. By processing these data at scale, ReactorIQ offers a granular and behaviorally grounded view of how drivers respond to lane closures, queues, and work zone configurations in both urban and rural environments. Insights are made available through an integrated web-based dashboard where users can access CV performance metrics, lane closure detection outputs, queue warning indicators, camera-based analytics, and machine learning results within a single decision-support environment.

This report details the development and key features of ReactorIQ, the challenges overcome and major accomplishments throughout the project, and directions for future work.

About the research

The goal of this project was to quantify the mobility and safety impacts of different combinations of lane width and shy distance to a barrier for a given paved width. The research team developed a device to measure lateral distance and derive speed, vehicle length/type, and headway information under day and night conditions. Data collected at 17 locations in Illinois, Michigan, and Wisconsin were used for the analyses. Lateral distance data of over a quarter of a million vehicles were used for the safety analysis. Extreme value theory (EVT) modeling was conducted to estimate the probabilities of right edge line encroachment and right barrier contact. Wider lanes were found to have decreased edge line encroachment and barrier contact, while wider shy distances were associated with increased edge line encroachment and decreased barrier contact. The speeds of over 125,000 free flow vehicles were used to quantify the mobility impact. Linear regression modeling was conducted to develop models for estimating free flow speeds in work zones. Work zone free flow speed increases with an increase in speed limit, lane width, and left/right shy distances to a barrier. A case study of a 55 mph posted work zone with two open lanes and barriers on both sides with an available paved width of 26 ft is presented. The results indicate that 11 ft lanes with 2 ft shy distances have a slightly lower probability of right barrier contact (for vehicles in the right lane) than 12 ft lanes with 1 ft shy distances while having a greater free flow speed. This research demonstrates how lateral distance can be collected and modeled along with speed data to assess safety and mobility impacts in work zones. Limitations of the study are acknowledged, and recommendations for future research are presented.

About the research

About the research

This study investigated methods for improving the effectiveness of speed feedback trailers (SFTs) when used as a speed management strategy in highway work zones. The research included a literature review, a state department of transportation (DOT) survey, and field evaluations conducted at several freeway work zones. The findings were synthesized to provide recommendations on methods for optimizing the deployment of SFT in freeway work zones. The state DOT survey revealed that SFTs are widely implemented in work zones across the United States, most commonly for lane closures and traffic shifts. Their use varies across states, ranging from optional to mandatory under specific conditions. SFTs are most commonly positioned near the work area or in advance of the lane closure taper and are often relocated as the work progresses. From there, a series of field studies were conducted within freeway work zones in Michigan and Missouri to evaluate the effectiveness of various SFT deployment strategies towards reducing work zone speeds and improving speed compliance. These evaluations, conducted in multiple phases and at five freeway work zone locations, sought to yield insights and recommendations for optimizing SFT deployment and introducing measures to improve their overall effectiveness. The evaluations specifically assessed the impact of strategically placing SFTs at various locations within the work zones, including near the start of a lane closure, approaching a work area, approaching a lane shift, and within a freeway crossover. Additionally, the effectiveness of SFTs were also assessed when combined with other strategies, like digital speed limits (DSLs) signs and police vehicle presence within the work zone. Although SFTs were generally effective at reducing work zone speeds regardless of the deployment characteristics, they tended to be more effective when positioned closer to the work area, including ingress/egress locations, where speeds were up to 3.6 mph lower when the SFT was present and active. SFTs were also effective at lowering work zone speeds when positioned within 1,000 beyond the end of the lane closure taper, within 1,000 ft in advance of the start of the taper, and within freeway crossovers. The speed reduction effects were generally sustained for at least one-half mile beyond the SFT. SFTs were also found to improve speed reductions measured near a police vehicle positioned within the lane closure by an additional 1.4 mph. Additionally, when paired with DSL signs on the same trailer assembly, the speed feedback display reduced speeds near the work area by an additional 1.8 mph. It is recommended that if only a single SFT is to be used, it should be positioned near the work area, approximately 200 ft in advance of the active work. If additional SFTs are available, then it is recommended that one be positioned within 1,000 ft upstream of the lane closure, shift, or crossover. Additionally, an SFT should be placed shortly beyond the end (e.g., within 1,000 ft) of any lane closure taper, preferably adjacent to the initial speed limit sign.

About the research

The ability for agencies to notify vehicles that they are approaching a work zone has the potential to reduce crashes by increasing motorists’ awareness of the conditions. A critical element of this ability involves providing accurate work zone information and reducing the potential for false alerts that motorists ignore, which can be accomplished through connected temporary traffic control devices (cTTCDs) such as connected arrow boards. Guidance is needed for agencies on integrating work zone data into traffic management operations to enhance safety and efficiency. This study focused on cTTCDs such as connected arrow boards, connected traffic cones, and other smart devices that improve the accuracy of work zone information. The research documents the current state of the practice, evaluates various cTTCDs, and explores methods for their integration into an agency’s work zone management system through technology such as an advanced traffic management system (ATMS).

For a broad view of how arrow boards can be integrated, arrow board data were summarized across 18 states. In total, 498,358 arrow board activations were captured between January 2023 and August 2024. As the search radius around each arrow board decreased, the percentage of locations that were unambiguously associated with a single roadway increased, with 86.6% of locations being unambiguous at a search radius of 25 ft. In total, 62% of arrow board activations had the closest roadway within 50 ft of the arrow board, with a majority of arrow board activations within close proximity to a roadway and two-thirds of the closest activations being less than 25 ft.

In terms of the potential benefits, the arrow board activations near a work zone represented only 11.3% of the total activations in Wisconsin, 1.2% of the total activations in Colorado, and 31.1% of the total activations in Iowa. In Iowa, arrow board activations occurred on average 290 minutes before the reported start time and 0.99 miles before the reported start location for verified work zones and on average 139 minutes before the reported start time and 0.51 miles before the reported start location for estimated work zones.

The use of cTTCDs such as connected arrow boards is expected to continue to increase, which will result in the need for additional research to continue to explore how this information can be utilized for real-time and historical analysis. The potential for automating the process of associating arrow boards and work zones still faces implementation challenges, but these can be overcome with continued evaluation and deployment of cTTCDs.

About the research

This study investigated the viability of using crowdsourced data sets, specifically segment speed data (SSD) and connected vehicle data (CVD), for providing real-time traffic information to the public. After a literature review and interviews with state department of transportation personnel were conducted, the study focused on work zone queue warning systems (QWS). Data from six work zones in Iowa were analyzed and compared in terms of data completeness, accuracy, and latency between SSD, CVD, and sensor data. The SSD showed high data completeness but poor performance in terms of missed and false calls, latency, and queue warning display. CVD, despite having challenges with overnight data coverage, achieved low missed calls and better latency than SSD. A virtual QWS approach was developed to evaluate the effectiveness of combining SSD and CVD. This involved using the archived data as a data feed to determine whether a queue warning would have been displayed. The SSD and CVD were compared against when the sensor data would have supplied a warning. In addition, an option combining both the SSD and CVD was tested. For this test, CVD performed better than SSD. The option combining both CVD and SSD was incrementally better than CVD alone. The study suggests that CVD has some potential for QWS applications, although low data coverage during overnight hours may be challenging. While SSD has good data coverage, it is less effective at identifying congestion. Further refinement in data processing and integration methods may be able to reduce false calls and improve overall performance.