Multi-echelon supply chain design considering unreliable facilities with facility hardening possibility

Mohammad. Rohaninejad | Rashed. Sahraeian | Reza Tavakkoli-Moghaddam

---

URL :   http://research.shahed.ac.ir/WSR/WebPages/Report/PaperView.aspx?PaperID=85215
Date :  2018/06/15
Publish in :    Applied Mathematical Modelling
DOI :  https://doi.org/10.1016/j.apm.2018.06.001
Link :  https://www.sciencedirect.com/science/article/pii/S0307904X18302464
Keywords :Facility location Reliability Multi-echelon supply chain design Approximation algorithm Sample average approximation algorithm Greedy adding and substitution algorithm

Abstract :

---

This paper considers the design of a multi-echelon supply chain network with unreliable facilities when the facilities can be “hardened” (their capacity increased and their probability of full or partial failures decreased) by more investment. The objective function of the problem is the minimization of the expected costs, including the fixed and the service costs, in all the echelons of the supply chain such that maintaining the flow of the demand at different echelons of the network interconnects them. In order to see how the decisions of each echelon affect other echelons, we consider a scenario in which facilities at the lower echelons of the chain are customers to the higher echelons and vice versa. A new scenario-based formulation is proposed for the problem to effectively cover the outcomes of this uncertainty. Moreover, we discuss how to use this model to choose the location and the hardening level of each facility to be both inexpensive and reliable. These decisions lead to some managerial insights. Given the computational complexity of the model, especially for the large-scale problems, several efficient approximation algorithms are presented. Finally, the computational results are provided to show the validity of the proposed model and the efficiency of the algorithms in providing high-quality solutions within acceptable times. The results show that the adoption of the integrated approach at different echelons of the network through the cooperation between the owners of echelons is much more effective than the independent optimization of each one.