Simulation-Based Optimization Algorithm for the Vendor-Managed Inventory Problem for Blood Platelets

Abstract

Objective: Estimate an optimal policy for the blood platelets supply chain distribution problem using a vendor-managed inventory problems approach. Methods and materials: This paper uses an integrated simulation-based optimization model to develop a Vendor-Managed Inventory approach for blood platelets. Simulation is used to estimate the performance of a defined inventory policy. On the other hand, a genetic algorithm finds optimal or near-optimal inventory policies. This approach is evaluated using a case study inspired by a real blood center in Colombia. Results and discussion: Using the proposed approach, key indicators in the blood supply chain such as total cost and outdated units are significantly improved while maintaining the service level.  In terms of costs, the VMI model shows a 19.19% advantage over the non-VMI solution. Moreover, the proposed VMI solution can reduce by 42.25% the number of expired platelets. Conclusions: Using a VMI-based distribution system and a simulation-based optimization approach with genetic algorithms offers promising results in the proposed use case. This mixed methodology allows for flexible system configurations without the need for complex changes in the algorithm, and it does so without the need for excessive computational resources.