A Multi-item Joint Inventory Replenishment Policy with Full Truckload Case Study of the Office Supplies E-Commerce Store in Thailand

Abstract

The objective of this research is to propose joint replenishment policies with full truckload for an office supply company, which sells through an online channel. The current policy is an (s,S) policy for each individual item, and it is not suitable for coordinated replenishment. Our two proposed methods are            1) a periodic review with an order-up-to level, and 2) a can-order system (s,c,S). We will show that our policies outperform the current policy in 2018. We first compute the demand forecasts using the historical records in 2015–2017 and select the best forecasting model that minimize the error measurement (e.g., MAD, MSE, MAPE). Then, we determine the mean and the standard deviation of the effective demand, and use them to determine parameters for two inventory policies, given the desired customer service level of 95%. Finally, we apply our inventory policies in 2018 with the constraints of full pallet and full truckload. We conclude that when the truck utilization is 100%, the costs associated with the first and second policies are 77.47% and 67.86% lower than the current policy, respectively. The company can improve its inventory performance by choosing either if the two proposed policies.