Optimization of Urban Shortest Path Considering Traffic Congestion and Air Quality Using ICAW and Genetic Algorithm (Case Study: Tehran)

This study proposes a multi-objective framework for optimizing urban shortest paths by simultaneously considering traffic congestion and air quality. The objective is to minimize a weighted combination of travel time and pollutant exposure, with binary decision variables representing route selection. Real-world air quality data (SO₂, NO₂, CO) were collected from IoT sensors on ۵۰۰ vehicles and ۵۰ fixed stations across Tehran, integrated with traffic data via spatio-temporal synchronization using inverse distance weighting and ۵۰۰m grid mapping. The transportation network was extracted from OpenStreetMap (۱۴,۲۸۷ nodes, ۳۲,۴۵۶ edges). The Incremental C-means Adaptive Weights (ICAW) algorithm was applied for real-time clustering of congestion and pollution hotspots, with optimal clusters (c=۵) determined via the elbow method. Route optimization was performed using a hybrid Genetic Algorithm (population=۲۰۰, tournament selection) with adaptive edge weights updated dynamically. The framework was evaluated across three pollution scenarios with ۱۰۰ routing instances per scenario. Results demonstrate statistically significant improvements (p<۰.۰۱): pollutant exposure reduced by ۱۸.۷% (S۱: good downtown air), ۲۳.۴% (S۲: moderate pollution), and ۳۱.۶% (S۳: severe pollution), with travel time reductions of ۱۲.۴%, ۱۰.۸%, and ۸.۹%, while distance increased by only ۳.۲-۶.۹%. Comparative analysis shows ICAW-GA achieves ۲۸% lower computational time than NSGA-II and ۴۲% faster convergence than PSO. The framework successfully balances transportation efficiency and environmental sustainability, providing real-time adaptive routing suitable for smart city deployment.

کلیدواژه ها:

Intelligent Transportation Systems ، Air Quality Index ، Traffic Congestion ، Genetic Algorithm ، Internet of Things