Network-wide multipath progression model for signal coordination using subpath connections

Abstract

Traffic signal coordination is essential for improving urban traffic efficiency. However, existing network-wide models struggle to handle complex multipath turning flows and intense resource competition. To address this limitation, this paper proposes a multipath progression model for network-wide signal coordination. The core methodology decomposes composite critical paths into ordered link-level subpaths, utilizing spatial connection constraints to ensure progression continuity across the network. To address interactions among intersecting paths, the model adopts a hierarchical bandwidth objective that jointly optimizes local, connection, and global bandwidths. This improves coordination under different levels of path competition and helps avoid infeasible solutions. Furthermore, the queue clearance time and maximum queue length constraints are explicitly incorporated to mitigate traffic queue interference. The offsets, phase sequences, and common cycle length are optimized within a unified mixed-integer linear programming framework. Extensive experiments are conducted to evaluate the model’s performance. Results show that the proposed model provides suitable green bands along critical paths. Simulation results demonstrate that the proposed model outperforms Synchro and MaxbandGN across multiple indicators, including average delay and number of stops.