This paper discusses the results of study on network surveillance performed by Mitretek Systems. The purpose of this study is to determine the value of traffic information provided by various levels of network surveillance by measuring travel time performance of a route guided vehicle population. The approach of this study is to employ the INTEGRATION traffic simulation model and an inter-urban network with both freeways and arterials. The network used in this study is based on the Cherry Hill, New Jersey area. For near-term deployment, it is unlikely that devices such as video cameras or inductive loops will be installed in all areas of a transportation network. For inter-urban freeway networks, state and regional highway agencies may have little or no real-time information about traffic conditions. Therefore it may be significantly more cost-effective to obtain network conditions from a relatively limited unguided probe population than by installing a system of fixed surveillance devices. This report examines the problems and effects of using unguided vehicles as probes in the network. A fixed population of guided vehicles do not function as probes. Results indicate that a partial surveillance system using 20% to 25% unguided probe vehicles can provide enough information to the Traffic Management Center for guided vehicles to experience almost all of the benefit associated with a network under full surveillance. More than half of this potential benefit may be realized with as low as a 1% probe vehicle population. (A)
Abstract