You are here

WORK-ZONE SAFETY: Smart at times

Framework determines when to turn to ITS in zones

Safety Article June 24, 2014
Printer-friendly version

Intelligent transportation systems (ITS) are being increasingly deployed in work zones to improve traffic operations and safety.

Also known as smart work-zone systems, these deployments provide real-time information to travelers, monitor traffic conditions and manage incidents. Despite numerous ITS deployments in work zones, a framework for evaluating the effectiveness of these deployments does not exist. To justify the continued development and implementation of ITS in work zones, there is a need to develop a uniform framework to determine ITS effectiveness for specific work-zone projects. In addition, guidance on the circumstances under which ITS deployment is recommended for a work zone is beneficial to agencies. Such a framework is developed in this article.

The framework consists of four steps as shown in Figure 1. In the first step, criteria for selecting work-zone sites for ITS deployment are established. The second step consists of selecting operational and safety performance measures for evaluating a deployment. Data needs for measuring and estimating the chosen measures are then determined. The performance benefits are quantified, total costs of ITS deployment estimated and the benefit-cost ratio is computed in the fourth step.

The site-selection criteria for choosing work zones that offer the greatest potential for traffic and safety improvement through ITS deployment are presented in Figure 2. They are: frequent congestion, high-traffic impact at the work zone, availability of alternative routes and over-capacity demand. Depending on the goals of the deployed ITS, one or more of the following five performance measures is recommended: delay, diversion rate, queue length, crash frequency and speed. These measures were obtained from the synthesis of the existing literature on ITS deployment studies.

Using the same measures across ITS deployments will allow for an easier comparison of the results. Traffic sensors that collect traffic flow, speed and occupancy are key to the chosen performance measures. Additional equipment such as temporary detectors and queue detection trailers also may be needed for accurately measuring queue length. For example, diversion rates can be computed using traffic-flow data collected from temporary traffic sensors deployed on the mainline and ramp. Crash data collected by the law-enforcement agencies are typically archived by the state DOT. Even though benefit-cost methodology has been implemented in other transportation areas, the application to work-zone ITS involves additional wrinkles such as the need to include technology cost factors and the computation of dynamic road-user costs.

Creating a diversion

The proposed work-zone ITS deployment framework is illustrated using two case studies. The case study sites are in the St. Louis region in Missouri. The St. Louis urban region has often been ranked approximately 20th in the U.S. in terms of annual delay according to the Texas A&M Transportation Institute Urban Mobility Report. The first case study, I-70 Blanchette Bridge, is in an urban setting with two major alternative routes. Five lanes were reduced to three lanes in each direction during construction. Only permanent ITS equipment that was already in place was used for the project and no temporary ITS equipment was deployed. The second case study, I-44 Antire Road work zone, is in a rural area with no alternative routes. Three lanes in the eastbound direction were reduced to two lanes. Temporary ITS equipment consisting of four portable dynamic message signs (DMSs), eight queue detection trailers and two Bluetooth travel-time sensors were added to complement the eight existing permanent DMSs.

The expected benefit of ITS in the I-70 site was to improve mobility by encouraging traffic to divert to alternative routes. Thus, the benefits of diverting traffic to alternative routes were estimated. A significant field data-collection effort was employed using portable surveillance blanketing the entire network, including the major alternative routes. This effort was significant, since diversion rates are often not measured in the field but estimated using driver choice models. Diversion rates were measured using traffic data collected before and during the work zone. A traveler survey was conducted to assess the extent to which drivers were influenced by ITS in diverting. The survey revealed that 52% of those that diverted to an alternative route did so due to ITS, specifically DMS. The percentage reliance on DMS was used in a traffic-simulation model to estimate the mobility impacts of ITS. Two scenarios were simulated—a “without DMS” scenario in which the proportion of traffic diverting to each alternative route was adjusted using the 48% value and a “with DMS” scenario in which the observed diversion rates during the work zone were used. The effect ITS has on delay was then computed by subtracting the delays for “without DMS” and “with DMS” scenarios.

The permanent DMS equipment in the study corridor serves multiple purposes, and therefore the costs should not be attributed solely to the work-zone project. However, the equivalent costs during the work-zone period were included in this study in order to estimate the benefit-cost ratio of permanent ITS deployments.

The total costs of ITS deployment were estimated to be $198,530 if equipment costs are included and $59,130 if only operating costs are included. The delay-reduction benefits were estimated to be $407,694.05, resulting in a benefit-cost ratio of 2.1 to 1 with equipment costs and 6.9 to 1 without equipment costs.

The use of ITS in the I-44 Antire Road work zone focused on safety. By warning drivers of downstream traffic speeds and queuing, the potential for rear-end crashes can be reduced. Accordingly, the safety benefits were computed based on the reduction in different types of crashes due to ITS deployment. Fatal, injury and property-damage-only crash severities were considered in the analysis. It was estimated that 5.6 property-damage-only crashes and 0.96 injury crashes were eliminated due to ITS deployment in the work zone. There were no fatal crashes that occurred during the work zone. The safety benefits were then quantified using the AASHTO Red Book unit cost values, which resulted in a total benefit of $345,900. The benefit-cost ratio was then computed using the actual ITS deployment cost of $106,700, which produced a benefit-cost ratio of 3.2 to 1.

Need a starting point

This study recommends that state DOTs interested in evaluating the effectiveness of work-zone ITS deployments must collect data without ITS to have a baseline to compare the results. Three approaches were discussed to collect such data. First, a control work-zone site similar in characteristics to the treatment site may be selected. The selection of a control site may not always be possible due to various reasons such as the uniqueness of the treatment work zone in terms of the type of work, geometrics and traffic characteristics. When choosing a control site is not feasible, traffic simulation is an alternative approach to generate control measurements. The data needs for calibrating a simulation model can be high, depending on the complexity of the work zone.

A third approach is to not use the ITS system for a few days during the work-zone period for measuring control data. Although the third approach provides the most accurate data for assessing ITS benefits, agencies might be hesitant to turn off functioning equipment due to liability issues or concerns over the public’s expectation of benefiting from such equipment. One possible alternative is to delay the deployment of ITS in the work zone for a few days to collect the “without ITS” data before turning on ITS. Thus, the “without ITS” data is collected before the public has experienced the ITS system.

But ITS may have provided valuable traveler information during the crucial initial period when drivers are least familiar with the work-zone conditions. Therefore, not having ITS in the initial period also could be undesirable.

The proposed framework also can be applied to estimate the benefits of planned work-zone ITS deployments. Planning and simulation models are needed to estimate the operational and safety measures due to ITS use in work zones. A few assumptions will be needed to assess the ITS benefits. For example, if the ITS is intended for diverting traffic to an alternative route, the possible diversion rate must be assumed based on historical data or the observed rates at other similar sites.
 
Acknowledgements
This research was conducted under the Midwest Smart Work Zone Deployment Initiative (SWZDI) and Federal Highway Administration (FHWA) Pooled Fund Study TPF-5(081). The authors would like to thank the FHWA, the Iowa DOT and the other pooled-fund state partners—Missouri, Wisconsin, Nebraska and Kansas—for their financial support and technical assistance.

The authors are thankful for the assistance provided by Missouri DOT staff Dan Smith, Jon Nelson, Linda Wilson, Tom Blair and Ryan Pierce, and Noah Jenkin and Pete Krikelis of ASTI for assisting with the I-44 data. The authors wish to acknowledge the contributions of University of Missouri research assistants Tyler Horn, Igor Caus, Dallas Crain, Tim Cope and Sawyer Breslow. Thanks also are due to Tracy Scriba with the FHWA for serving on the technical advisory committee of the project and reviewing the draft report. ST

About the author: 
Edara is an associate professor, Department of Civil Engineering, at the University of Missouri, Columbia.
Email Subscriptions