A two-stage stochastic inventory management model for an intermodal trucking company

Abstract

Intermodal transportation faces several challenges due to uncertainty in rail schedules and customer demand. However, this uncertainty is rarely considered for determining asset management at the Intermodal rail yards. Typically, each Intermodal rail yard requires certain inventory of chassis to serve the demand for either empty containers or loaded containers. It is crucial for any transportation firm to optimally allocate and move chassis between rail ramps to overcome random demand. This thesis develops a two stage stochastic optimization model to determine the optimal allocation and repositioning decisions for chassis and empty boxes across the rail yards to minimize costs and meet service levels. The first stage formulation contains the initial chassis allocation decisions which are independent from random parameters in the following time periods. The second stage formulation determines the empty boxes and chassis repositining decisions for subsequent time periods when the random demand is realized. This thesis applies the L-Shaped Method to efficiently solve this problem. Using numerical experiments, this thesis analyzes the impact of system parameters on the run time performance. The thesis also analyzes the impact of initial chassis inventory and demand patterns on the optimal decisions. We observe that the higher initial inventory or demand at one location than the other results in an increase in the required repositioning moves and expected cost. Conversely, the model is fairly robust to how inventory and demand values are distributed between resource types.