Identification of factors influencing potential collisions and risk prediction of professional taxi drivers in China

Traffic safety among taxi drivers is a critical concern, particularly in high-density urban environments. This study proposes a quantitative model to predict crash risk by examining the symmetry and asymmetry of various influencing factors, including personal characteristics, workload, risky driving behaviors, and crash history. An anonymous survey was conducted among taxi drivers in four Chinese cities, collecting 1010 valid responses. Using an ordered logistic regression model, we identified 14 key risk indicators, such as severe sleep problems, high income dissatisfaction, and 12 types of risky driving behaviors. The results indicate asymmetric effects, where some risk factors disproportionately contribute to crash risk, revealing an imbalance in workload distribution and behavioral tendencies. Additionally, descriptive analysis shows that Chinese taxi drivers face significant workload pressures, with an average of 12-hour workdays, which further disrupts driving stability and safety. To address these asymmetries, the study recommends government regulations to limit working hours and encourages taxi companies to provide targeted safety education. The findings provide valuable insights for policymakers and industry stakeholders to design interventions that restore balance in driver workload and risk exposure, ultimately contributing to a safer urban transportation system.