A Multi-Stage Stochastic Mixed-Integer Linear Programming to Design an Integrated Production-Distribution Network under Stochastic Demands

Supply chain management has gained much interest from researchers and practitioners in recent years. Proposingpractical models that efficiently address different aspects of the supply chain is a difficult challenge. This researchinvestigates an integrated production-distribution supply chain problem...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Industrial Engineering & Management Systems Jg. 17; H. 3; S. 417 - 433
Hauptverfasser: Derakhshi, Mohammad, Niaki, Seyed Taghi Akhavan, Niaki, Seyed Armin Akhavan
Format: Journal Article
Sprache:Englisch
Veröffentlicht: 대한산업공학회 01.09.2018
Schlagworte:
산업공학
Supply chain
Pareto front
Bi-objective optimization
Scenario reduction
Production-distribution
Stochastic programming
ISSN:1598-7248, 2234-6473
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Supply chain management has gained much interest from researchers and practitioners in recent years. Proposingpractical models that efficiently address different aspects of the supply chain is a difficult challenge. This researchinvestigates an integrated production-distribution supply chain problem. The developed model incorporates partieswith a specified number of processes to obtain raw materials from the suppliers in order to convert them to semi andfinal products. These products are then distributed through warehouses to end-distributors having uncertain demands. This uncertainty is captured as a dynamic stochastic data process during the planning horizon and is modeled into amulti-stage stochastic mixed integer linear program using a scenario tree approach. For large-size instances, a hybridexact-approximate algorithm is proposed, where its effectiveness is assessed via several numerical cases. Furthermore,the model is generalized to its bi-objective version by considering the accessibility of the products based on the safetystock policy of the companies involved. In the end, an existing algorithm is combined with the ε-constraint method toobtain an approximate Pareto front. KCI Citation Count: 0
ISSN:1598-7248
2234-6473
DOI:10.7232/iems.2018.17.3.417