The average hourly income of taxi drivers could be improved by understanding the realized income of taxi drivers and investigating the variables that determine their income. Based on 4.85 million taxi-trajectory GPS records in Shenzhen, China, this study built a multi-layer road index system in order to reveal the behavioral patterns of drivers with varying income levels. On this basis, late-shift drivers were further selected and classified into two categories, namely high-earning and low-earning groups. Each driver within these groups was further classified into three income levels and four categories of factors were defined (i.e., occupied trips and duration, operational region, search speed, and taxi service strategies). The sample-based multinomial logit model was used to reveal the significance of these income-influencing factors. The results indicate significant differences in the drivers’ behavioral habits and experience. For instance, high-earning drivers focused more on improving efficiency using mobility intelligence, while low-earning drivers were more likely to invest in working hours to boost their revenue.

The effective recognition of risky driving behaviors holds technical potential for supporting accident prevention and sustainable transportation. However, existing intelligent algorithms for optimizing deep learning models in this field often suffer from slow convergence and high errors. This study proposes a novel hybrid model (ICPO-XGBoost) for risky driving behavior classification. The improved crested porcupine optimizer (ICPO) was developed using logistic-tent composite mapping for population initialization, a hybrid mechanism combining refraction opposition-based learning and Cauchy mutation to avoid local optima, and an adaptive variable spiral search with inertia weight to balance global and local search. The ICPO was then employed to optimize the hyperparameters of the XGBoost classifier. The ICPO demonstrated superior optimization accuracy and convergence speed compared to benchmark algorithms. The ICPO-XGBoost model achieved accuracy, precision, recall, and F1 scores of 96.2%, 95.4%, 95.8%, and 95.6%, respectively, for classifying and identifying risky driving behaviors. Compared to various benchmark models, these results represent increases of 12.7–24.8%, 14.8–31.8%, 14.9–31.0%, and 15.0–32.4%, respectively. For specific driving behavior categories (normal driving, slow driving, short-distance tailgating, sudden acceleration/deceleration, frequent lane changing, and forced lane changing), the precision, recall, and F1 scores of the ICPO-XGBoost model fell within the ranges of 84.8–99.2%, 87.5–100.0%, and 86.2–99.2%, respectively. Compared to benchmark models, these metrics show increases of 1.5–75.8%, 5.8–68.1%, and 3.3–72.6%, respectively. Notably, the model significantly improved accuracy in identifying sudden acceleration/deceleration behaviors. The results of this model facilitate the classification and early warning of risky driving behaviors, thereby reducing the frequency of such behaviors, lowering the risk of traffic accidents, and enhancing road traffic safety.