Estimating regional shared trailer demand in drop-and-pull transport

Abstract

With the advancement of modern information technology, the demand for shared trailers in drop-and-pull transport has become a research focus. In this study, a method for estimating regional shared trailer demand is proposed. This method comprehensively considers the regional drop-and-pull level and the total tonnage of operational freight vehicles to calculate the total tonnage of shared trailers in the region, thus estimating the shared trailer demand. This provides a reference for the allocation of shared trailers in regional drop-and-pull transport. The regional drop-and-pull level is measured by the ratio of drop-and-pull freight volume to total freight volume. Three different machine learning models, namely linear regression, random forest, and XGBoost, were used to predict the freight volume. The shared trailer demand was then calculated based on the freight volume. The comparison of prediction results showed that the XGBoost model had the best predictive performance. Finally, the effectiveness of the shared trailer demand estimation method was verified through a case study.

Keywords

• Drop-and-pull transport
• Shared trailer
• Demand estimation
• Freight volume prediction
• Machine learning