The IoV-TraSim: IoT-Enabled V2X Adaptive Traffic Simulation for Real-World Urban Congestion Mitigation

Abstract

Population growth, increasingly large numbers of vehicles on the roads, and the lack of sufficient infrastructure to respond to the dynamic nature of roadway intersection traffic-control requirements are putting growing pressure on urban transportation systems globally. Existing simulation frameworks do not easily integrate large-scale real-world topology extraction with V2X-aware agent-level modelling and heuristic calibration for developing world cities. In this paper, an IoV-TraSim framework for adaptive traffic simulation based on vehicle-to-X communications is proposed, which incorporates the road topology derived from OpenStreetMap, the V2X sensing infrastructure modelled by RSU/OBU, microscopic agent-based vehicle movement, congestion-aware speed modulation, and a heuristic calibration engine to optimise adaptive signal phase timing. The framework is tested on a real road network for Baghdad consisting of 166,002 nodes and 368,548 edges with four cases: baseline fixed-phase operation, expansion of the high-speed corridor with V2X infrastructure, a controlled expansion at a moderate speed limit under fixed-phase control, and moderate-speed adaptive signal coordination. The proposed adaptive V2X coordination in Scenario 3 raises the calibration score from 0.783 to 0.897, an improvement of 14.6% over the fixed-phase operation of Scenario 1 and of 39.5% over the uncontrolled expansion of Scenario 2, lowers the per-zone tuning error from 31.6% to 10.2%, a reduction of 21.4 percentage points, and raises the vehicle arrival rate from 88.4% to 91.3%. A paired statistical test applied to over 30 independent replications per scenario verifies an improvement in Scenario 3 over Scenario 1 for the smaller size, with $p < 0.001$. The mean network speed of Scenario 3 is 10.9 km/h, which is lower than that of the other configurations because the expanded corridors are operated under a 50 km/h posted limit rather than 80 km/h; calibration fidelity and distribution balance are therefore reported separately from raw speed. Within the scope of the simulated Baghdad network, the results indicate that adaptive IoT-V2X signal coordination mitigates congestion more effectively than uncontrolled physical expansion of the road network.