Data-driven robust model for container slot allocation with uncertain demand

Abstract

The global container transportation market is complex and changeable, resulting in high uncertainty of container cargo demand. Effective container slot allocation (CSA) decisions are difficult for shipping companies to make. A data-driven robust model was developed for optimizing the CSA under highly uncertain cargo demands to maximize the revenues of shipping companies. The features of the empty container transportation were integrated into the model due to the imbalance of container import and export. The Copula method was employed to construct the uncertain set to deal with the limited historical demand information. Moreover, the model can be transformed into a simple and manageable robust optimization problem by introducing the protection level. Finally, the effectiveness of the optimal robust CSA policy was verified by numerical examples. Results demonstrated that the robust optimization model effectively balances the relationship between the revenue and the risk preference of shipping companies and maximizes the revenue by using limited cargo demand information. The optimal robust slot allocation policy is more stable under the heavy tail cargo demand. (Received in August 2021, accepted in November 2021. This paper was with the authors 1 month for 1 revision.).