Development of a Deployment Planning Tool for Safety Service Patrols on Freeways in Virginia

Many transportation agencies take part in the freeway incident management (IM) process by deploying safety service patrols (SSPs), typically on highly congested, urban roadways. SSPs are a countermeasure to freeway operations problems, and their implementation can improve safety and mitigate congestion. They are typically deployed in areas that have high traffic volumes (e.g., urban freeways) and are charged with clearing obstructions from roadways such as debris and disabled vehicles as well as assisting state police with traffic control at crash scenes. Although patrols continue to expand throughout the nation, deployment criteria have remained relatively unchanged (e.g., deployment is often driven by congestion patterns in major metropolitan areas). In view of the relatively limited factors used for deployment decisions, a few SSP programs have deemed it necessary to develop empirically driven deployment criteria to help guide decision-makers in making sound, data-driven investment decisions. The purpose of this study was to develop a deployment planning tool that would help Virginia Department of Transportation (VDOT) decision-makers when considering expanding SSP coverage and/or altering existing route coverage. The planning tool developed in this paper is a segment-based ranking scheme that can be applied to rural and urban freeway segments. Within the tool an SSP-assist prediction model (APM) was developed to predict incidents statistically using freeway segment average annual daily traffic (AADT), length, average daily percent of ADT served, and truck percentage. The SSP deployment planning tool can be used when deploying new patrols or altering existing ones. To do this, existing and potential SSP routes should be included in the evaluation. Each route must be divided into its constituent segments (traffic links), and each segment scored using the predicted number of incidents, level of service, planned projects, air quality, maximum access distance, maximum structure length, AADT, and daily truck volume. Route scores are computed by averaging the constituent segment scores. Potential routes are then ranked based on their scores.