Work

A Genetic Algorithm for Medicine Inventory Management Under Uncertain Demand

Year: 2026
Type: article
Abstract: Objective: This study aimed to propose an optimal drug inventory management approach for government hospitals under uncertain demand, particularly during emerging disease scenarios like Coronavirus Disease 2019 (COVID-19). The goal was to minimize inventory management costs, including holding, ordering, and drug costs, by determining the optimal order quantity and reorder point.Material and Methods: Drugs were categorized using the K-Means Clustering method to identify similar demand patterns. A mathematical model and a genetic algorithm (GA) were developed to determine the optimal ordering policy. These methods were evaluated based on total inventory management costs and processing times. Sensitivity analysis was performed to evaluate the effects of cost variations, with increments ranging from 10% to 50%.Results: The mathematical model achieved a total inventory management cost of 80,652,330.9 Thai Baht (THB), a 4.7% reduction (3,800,278.1 THB) compared to the genetic algorithm’s 84,452,609.0 THB. However, the genetic algorithm significantly reduced processing time to 60.2 minutes, compared to 368 minutes for the mathematical model, representing an 83.6% time reduction. Compared to the current policy’s cost of 93,442,791.9 THB, the mathematical model lowered costs by 13.7% (12,790,461.0 THB), while the genetic algorithm achieved a 9.6% reduction (8,990,182.9 THB).Conclusion: The proposed methods effectively reduced inventory management costs and processing times compared to the existing policy. This study introduces an integrated approach that combines K-Means clustering, a mathematical model, and a genetic algorithm to efficiently manage hospital drug inventories under uncertainty, reducing both costs and processing time.
Language: English
Open Access status: diamond
OpenAlex ID: W7127048930
FWCI:
Cites: 20
Cited by:
Related to: 10