Design of mathematical models for the integration of purchase and production lot-sizing and scheduling problems under demand uncertainty
Milad Mohammadi,Majid Esmaelian,Arezoo Atighehchian
DOI: https://doi.org/10.1016/j.apm.2020.03.021
IF: 5.336
2020-08-01
Applied Mathematical Modelling
Abstract:This study addresses the multi-level lot-sizing and scheduling problem with complex setups and considers supplier selection with quantity discounts and multiple modes of transportation. The present research proposes a mixed-integer linear programming (MILP) model in which the purchase lot-sizing from multiple suppliers, production lot-sizing with multiple machines and scheduling of various products of different families are accomplished at the same time. However, these decisions are not integrated in traditional environments and are taken separately. In this study, two different types of lot-sizing models called aggregated and disaggregated are developed for the problem to evaluate and compare the computational efficiency of them under deterministic and stochastic demands and provide some managerial insights. To deal with the stochastic demands, Chance-Constrained Programming (CCP) approach is applied. Based on the results of this study, the average profit of the separated (purchase from production) lot-sizing model under demand choice flexibility and stochastic demand is 24% and 22% less than the integrated model, respectively. Moreover, the results also confirm the effect of discount structure on the amount of purchases, productions, revenues and costs.
mechanics,engineering, multidisciplinary,mathematics, interdisciplinary applications