Enhancing Urban Intersection Efficiency through Signal Retiming: A SIDRA-based Case Study from Kathmandu

Abstract

Enhancing intersection operations is crucial for reducing congestion and improving traffic safety, especially in rapidly urbanizing areas. This study analysed the congested Jorpati Intersection in Kathmandu, a key unsignalized junction handling 47,610 vehicles daily, with traffic increasing by 277% from 2012 to 2014, despite no major upgrades. Using detailed field surveys, video-based traffic data collection, and SIDRA Intersection software, the study assessed current conditions and evaluated four signal retiming scenarios during morning and evening peak hours under mixed-traffic conditions with vehicle-specific inputs. The SIDRA model was calibrated and validated using local conditions to replicate the real-world scenario by comparing queue lengths in each leg. The results revealed serious operational inefficiencies under current manual control systems, with vehicle delays of up to 97 seconds, queue lengths exceeding 280 metres (in the worst-performing lanes), and an average performance index (PI) of 1340 during peak hours. Among the simulated scenarios, the ‘left controlled phase’ with optimal cycle timing demonstrated the greatest improvements, reducing maximum queues by 63%, delays by 30%, and PI by 36%, while upgrading the level of service from C to B. These findings underscore the effectiveness of signal retiming in enhancing intersection performance in mixed-traffic environments. The study’s methodology provides a replicable framework for similar urban contexts and offers practical insights for managing traffic in rapidly urbanising areas where standard engineering solutions may need local adaptation.