Assessment of road accident risk in India under a type-2 picture fuzzy framework with novel distance measures.

This study presents a comprehensive analysis of road accidents in India by using official government data sets from 53 major cities across 28 states. This research is conducted under a Type-2 picture fuzzy environment with advanced aggregation operators such as the arithmetic mean, geometric mean, hamy mean, Einstein weighted operator, and Aczel-Alsina operator. To enhance decision-making under uncertainty, a novel distance measure is proposed and theoretically validated. Five prominent distance-based multi-criteria decision-making (MCDM) methods such as TOPSIS, VIKOR, WASPAS, CODAS, and COPRAS were applied to rank the cities based on 35 critical road accident-related criteria. The results of the model revealed that Chennai consistently secured the top ranking in TOPSIS, COPRAS, and FIR. Amritsar also performed exceptionally, ranking 1st in VIKOR and achieving 3rd in TOPSIS. In contrast, cities such as Kanpur, Lucknow, and Ahmedabad consistently ranked in the bottom quartile, with Kanpur receiving the lowest position (53rd) in TOPSIS and FIR, indicating they are in critical risk zones. Mid-ranked cities such as Hyderabad, Pune, and Rajkot showed variations across methods but maintained stable FIR positions. To validate the robustness of the rankings, a comprehensive sensitivity analysis was conducted by various decision parameter q, confirming the stability of rankings under fluctuating conditions. The proposed framework improved the precision of accident severity and demonstrated adaptability to broader decision-making scenarios involving uncertainty. This study contributes a novel, integrated, and mathematically rigorous methodology for assessing urban safety for policymakers to reduce the road accident risks across Indian cities.