Integrating Microscopic Simulation and Optimization: Application to Freeway Work Zone Traffic Control

This paper presents a methodology for optimizing performance of a traffic system on the basis of simulated observations of its microscopic behavior. The method integrates simulation and optimization submodels for describing traffic flow on urban freeway lane closures. The stochastic nature of traffic is accounted for in determining the true system response to traffic control variables. The simulation submodel has been validated at a series of work sites in the Chicago area expressway system. The optimization submodel optimizes a single objective function subject to a set of linear constraints. Preliminary model applications included the determination of an optimum merging strategy to be adopted by traffic entering the work zone in lanes to be closed for traffic. The model recommendation yielded the lowest average travel time in the work zone and, interestingly, did not incorporate many early merges; the latter is often viewed as a desired merging strategy. In addition, the optimum merging strategy varied with the traffic flow level entering the work zone and with the character of the objective function to be optimized.