Nonlinear Impacts of Urban Decentralization on Arterial Traffic Performance: Evidence from Polycentric Mobility Transitions and Microsimulation Analysis

Abstract

Urban mobility systems are undergoing significant transformation due to urban decentralization, increasing telecommuting, multimodal transport adoption, and evolving travel behavior. These changes challenge conventional traffic performance assessment methods that rely on static and volume-based assumptions. This study aims to synthesize how mobility transitions influence traffic performance outcomes, including Level of Service (LOS), speed, delay, density, and congestion redistribution in polycentric and rapidly urbanizing cities. Using a structured literature synthesis approach, the study reviews advances in microsimulation modeling, behavioral mobility representation, and scenario-based traffic analysis. It examines calibration methods such as trajectory-based calibration, machine learning-assisted parameter estimation, and hybrid optimization techniques, as well as mobility transition scenarios including decentralization, telecommuting, multimodal integration, and connected and autonomous vehicle adoption. Key indicators analyzed include LOS, travel time reliability, queue length, density distribution, and delay variability. The findings show that mobility transitions mainly redistribute congestion spatially and temporally rather than reduce it absolutely. Decentralization shifts congestion across corridors, telecommuting smooths peak demand with context-dependent effects, and multimodal integration improves accessibility while potentially increasing congestion at transfer points. Traffic responses are highly non-linear, making traditional static LOS approaches insufficient, particularly in mixed-traffic environments. Enhanced microsimulation models with data-driven calibration provide more accurate representations of real-world traffic behavior. The study concludes that future urban transport planning requires integrated, behavior-sensitive, and data-driven approaches using multidimensional performance indicators to support adaptive and resilient mobility management.