Demand Allocation and Lot-Sizing Solution of Single Product Deterministic Demand
Nguyen Van Hop
DOI: https://doi.org/10.1016/j.eswa.2023.122927
IF: 8.5
2024-01-11
Expert Systems with Applications
Abstract:In this study, a two-phase algorithm has been developed to address the single-product supplier selection and order allocation problem (SP-SSOAP). In the first phase, a novel demand allocation algorithm is developed using a recursive demand equation. Subsequently, an adaptive genetic algorithm is employed in the second phase, with additional constraints for capacity and minimum order quantity, to determine the best lot-sizing plan for each supplier. Experimental results demonstrate that, on average, the proposed heuristic achieves an objective value within 1% of the optimal solution for small-sized instances and outperforms the previous heristic solution by an average of 12% in terms of objective value for larger problems. Additionally, the numerical results emphasize that the proposed method performs better when input parameters such as price, ordering cost, and holding cost exhibit lower fluctuations. This suggests that managers should focus on improving supplier relationships and coordination strategies to reduce variability in price, order cost, and holding cost.
computer science, artificial intelligence,engineering, electrical & electronic,operations research & management science