Predicting on-time deliveries in trucking: A model based on the working conditions of drivers

Abstract

Over a period of two years, 26.3 thousand road freight shipments were recorded. The records include information about truckload companies, drivers, and the causes of noncompliance and delays in deliveries. Logistic regression based in working conditions as independent variables was used to predict non-compliance deliveries attributed to cargo drivers. Results show that vehicle type, medical coverage and social security, level of stress, work dissatisfaction, and transit time were strongly associated with out-of-timedelays in deliveries. The proposed model is a promising tool to improve the performance of truckload companies and it may motivate to benefit working conditions of truckers.