Driving Behaviour, Traffic Friction, and Sustainable Mobility: Evidence from Urban Corridors in Patna, India

Traffic congestion and widespread non-compliance with rules remain major obstacles to sustainable urban mobility in fast-growing Indian cities, where mixed vehicle types, low rule compliance, and informal roadway use complicate traditional traffic-flow assumptions. To understand the joint effect of 19 behavioural and environmental variables on the two traffic-performance indicators of mean spot speed (km/h) and passenger car unit (PCU) flow (PCU/h), a balanced field design was adopted with three land-use categories, three observation days, and three time periods across nine urban corridors in Patna city in India. We used the Shapiro–Wilk test, Pearson, and Spearman correlations, and multiple linear regression (Enter) in SPSS 27. SPEED and PCU showed a moderate negative correlation (Pearson r = −0.394; p < 0.001). The full-sample SPEED model was statistically significant (R2 = 0.957, adjusted R2 = 0.943, F(19,61) = 71.127, p < 0.001) with mean speeds (Q1) and travel-time reliability (Q3) as significant predictors. The PCU model was also statistically significant (R2 = 0.961, adjusted R2 = 0.949, F(19,61) = 78.779, p < 0.001), and the factors mean speeds (Q1), zig-zag lane changes (Q7), rule-following culture (Q11), and slow-moving modes (Q15) were significant predictors. Although the full-sample models had a maximum VIF of 15.5, and the land-use-stratified models had even higher VIFs, interpret individual coefficients with care, especially in the stratified models. Findings indicate that low-cost behaviour change interventions specific to land-use function can complement infrastructure-based approaches to achieve sustainable, efficient, and equitable traffic control in rapidly urbanising Indian cities. The results can support future context-sensitive traffic management, behavioural interventions, and sustainable mobility planning in rapidly urbanising cities with heterogeneous traffic.