A Benefit Cost Model for Traffic Incident Management

Every day, non-recurring incidents cause delay on major roadways and cost the public valuable time and, hence, money. Some public agencies have systems that help them to identify and mitigate the problems associated with incidents such as delay. These systems are referred to as Traffic Incident Management systems. Many different agencies have searched for a way to quantify the benefits that these programs provide. This thesis details how a model was developed to help these agencies easily quantify the benefits that can be derived from saving delay through efficient incident management. This model uses real traffic data collected by roadway traffic sensors to find the actual delay experienced by roadway users. It consists of 30 second aggregate data that measured the speed, occupancy, and flow at over 200 stations along the interstate system in Knoxville. It was collected by the Tennessee Department of Transportation in Knoxville, Tennessee and was used to help quantify the delay associated with incidents on the interstate. Another part of the dataset for the model included an incident log that recorded every incident that was reported or identified in Knoxville for the year 2009. Once the delays were quantified for each of the stations for every thirty seconds for the whole year, the benefits of a traffic incident management system could be calculated. The benefits were quantified by using the delay savings due to the incident management program, and thus, how much money was saved for each incident. The benefit in delay saved for the city of Knoxville was found to be $12.1 million dollars. The costs for the Knoxville system were $1.43 million dollars, thus giving a benefit cost ratio of approximately 8.5:1. The model developed to determine this benefit cost ratio can be applied to other places and reused with only minor adjustments that do not require extensive data collection. This is advantageous because it requires less time and effort to calculate the benefits of any particular incident management system.