Replenishment management with stockout effects
The purpose of this project is to investigate the effect of stockouts on a firm's operational policy and total discounted profit. We find that the financial impact of ignoring stockout effects can be significant, especially when a customer's memory is short and profit margin of the product is small. These results provide insight for operational decisions and customer services to effectively manage inventory systems with stockout-dependent demand.
The application of this project requires MATLAB software. We use MATLAB software to do simulation experiment. Specifically, we use MATLAB software to randomly generate simulated data for demands and parameters randomly produced by uniform distribution. Then we use MATLAB software to test the profit loss caused by ignorance and investigate the performance of heuristic policies when considering stockout effects.
(1) File - “profit loss caused by ignorance”
Under this folder, we created two files, the “main” file is the main program, and the “pllu” file is the subprogram called by the main program. In this file, the simulated demand data is generated from uniform distribution or truncated Normal distribution. We consider a series of scenarios with different customer memory factors and marginal profit ratios. For each scenario, we generate 1000 instances with parameters randomly drawn from uniform distributions. We calculate the heuristic policy when ignoring the stockout effects and the optimal policy when considering the stockout effects. The profit loss caused by ignorance means that the loss of profit resulting from the adoption of the heuristic policy compared with the maximum profit obtained when the optimal policy is adopted.
(2) File - “Performance of expected service level”
Under this folder, we created two files, the “performance” file is the main program, and the “perishable” file is the subprogram called by the main program. In this file, the simulated demand data is also generated from uniform distribution or truncated Normal distribution. We also consider a series of scenarios with different customer memory factors and marginal profit ratios. For each scenario, we generate 1000 instances with parameters randomly drawn from uniform distributions. We calculate the heuristic policy through defining the service level by its mean and the optimal policy when defining the service level by the actual shortages. The performance of the heuristic policy means that the loss of profit resulting from the adoption of the heuristic policy compared with the maximum profit obtained when the optimal policy is adopted. We find that the heuristic policy performs well, with close-to-optimal performance. Its average performance ranges from 0.012% to 0.031% and its worst performance ranges from 0.094% to 0.837%. In comparison, the heuristic policy of ignoring the stockout effect does not perform as well, especially when the profit margin ratio is low and the customer memory is short. In this case, the profit loss can easily exceed 10%.
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》