An Empirical Analysis of Last-Mile Delivery Efficiency: Evaluating Operational and Demographic Determinants in E-commerce Logistics

The rapid expansion of the e-commerce and food delivery sectors has placed unprecedented pressure on last-mile logistics. This study investigates the factors influencing delivery rider efficiency through a comprehensive statistical lens. Utilizing a dataset of monthly performance metrics, the research employs descriptive analysis, paired t-tests, independent t-tests, ANOVA, and multiple linear regression to identify productivity drivers. Key findings reveal a significant "delivery gap" between orders attempted and orders delivered (t = 26.99, p < 0.001), primarily attributed to operational friction such as customer cancellations and Return to Origin (RTO) incidents. The regression model (R2 = 0.89) identifies Orders Attempted and Working Hours as the strongest positive predictors of success, while Cancelled Orders and RTOs significantly impede performance. Notably, the study finds that demographic factors, including gender (p = 0.77) and education level, have no significant impact on delivery productivity. The study concludes that last-mile efficiency is predominantly determined by operational variables rather than rider demographics. Recommendations focus on technological interventions to reduce order cancellations and optimize route density to bridge the attempt-delivery gap.