A MODEL AND FAST OPTIMIZATION METHOD FOR SIGNAL COORDINATION IN A NETWORK

The coordination of signal controls in a traffic network has a significant impact on throughput and journey times. Traditionally, signal coordination is optimized as a separate model step, after route flows have been obtained from an assignment. A better integration of both steps requires a model of signal coordination which admits a fast solution method. We describe such a model and a solution based on mathematical optimization techniques for mixed-integer problems. Compared to existing approaches, our method supports non-uniform cycle lengths and yields optimal solutions (w.r.t. the objective function of our model) in less computation time. We use microsimulation to compare the resulting signal phasing to one obtained from an industry standard model and report on effectiveness and computational effort.