Decision Support System for Offline and Online Traffic Management in Urban Areas

A successful traffic management in urban areas undoubtedly determines improvements in life quality. In addition to delays and discomfort for users, traffic congestion has negative impacts on safety and environment. Consequently, the availability of an effective traffic management system is crucial for local administrations and transport operators to make urban mobility and quality of services more efficient. Given this framework, this paper focuses on the development of a decision support system for urban traffic management based on traffic optimization through a local approach for solving the signal setting design of each individual intersection applying equisaturation policy. This problem is addressed as operational planning, thus requiring the simulation of demand-supply interaction performed with dynamic assignment. The procedure occurs automatically through specific Python programming language codes applying both online and offline control strategies for managing non-recurring and recurring traffic situations, respectively. In the latter, the signal setting design results from combined solution of assignment and control issues. The procedure was tested on a real case study concerning 48 signalized intersections in Rome, Italy. Both online and offline applications show performance improvements in terms of delay reduction up to 14% and 20%, respectively. In addition, offline strategies appear sensitive to stability of simulation results and accuracy level of dynamic assignment, so allowing the evaluation of both these variables within the set calculation time. Overall, results confirm the potential of the proposed system, easily transferable to any context and transport network, as a practical tool for automatically and successfully optimizing urban signal plans.