A hybrid fuzzy petri net-based approach incorporating extended grey numbers for eco-driving behavior evaluation

Abstract Eco-driving behavior reduces vehicle emissions, and its evaluation is key to enhancing fuel efficiency and mitigating pollution. However, existing assessment models face challenges in integrating multi-dimensional data and addressing uncertainties, which limits their accuracy and practical applicability. To address this gap, this study proposes an eco-driving evaluation framework based on an Extended Gray-Number Weighted Fuzzy Petri Net (EGWPNs). By inputting driving process data such as acceleration values and low-speed duration, the framework yields specific eco-driving scores. The framework raises a weighted extended gray-number set to unify heterogeneous data types, including discrete events like rapid acceleration frequency and continuous variables such as acceleration values. By incorporating the MYCIN confidence method for uncertainty reasoning and the Bonferroni mean operator for multi-attribute aggregation, the EGWPNs model achieves an objective assessment of driving behavior. The framework was validated using 420,000 real-world driving data points collected from 99 vehicles in actual driving emission experiments. The results indicate that frequent rapid acceleration exhibits the strongest negative correlation with eco-driving scores, with a weight coefficient of 0.232, followed by prolonged acceleration events and sustained low-speed acceleration, with weight coefficients of 0.112 and 0.110, respectively. Compared to traditional gray reasoning petri nets, the EGWPNs model improves the evaluation interval shrinkage by 62.18% and demonstrates superior stability. The EGWPNs framework’s adaptability to heterogeneous data enables direct integration into intelligent transportation systems, reducing vehicular emissions through optimized traffic management and enhanced compliance with carbon neutrality policies. This study advances eco-driving methodologies while delivering scalable solutions to mitigate transportation- related environmental impacts.