A facility location model for an integrated logistics system in a finite planning horizon with probabilistic customer/supplier participation

• Main Author: Lee, Kim Janeya C.A.
• Format: text
• Language: English
• Published: Animo Repository 2005
• Subjects: Facility management; Integrated logistic support; Production management; Industrial Engineering
• Online Access: https://animorepository.dlsu.edu.ph/etd_masteral/3281; https://animorepository.dlsu.edu.ph/context/etd_masteral/article/10119/viewcontent/CDTG003888_P.pdf
• Institution: De La Salle University

Ideal facility location is one of the most practical methods to minimize costs. Ideal facility location has the transportation costs and facility costs are reduced to a minimum. In most literatures, facility location decisions were applied solely for either forward distribution, or reverse distribution only. Those that combined both forward and reverse distributions do not consider the possibility of sharing both distribution systems in common facilities even though it leads to savings for the company due less capital and operational costs that would be incurred. In this study, a mathematical model that aims to minimize the total operating and capital cost that would be incurred by the system was formulated with both systems sharing facilities as its focal point. The constraints included in the mathematical model are the following: supply and demand constraints, opening, expansion and closing constraints of the facilities, capacity constraints and the probability constraints. The model was translated to the General Algebraic Modeling Systems (GAMS) Language. From the design of experiment, it was found that the operating expenses (variable and fixed expense) of the company are the major factors of the system. In the response surface methodology, the relationships among these factors are analyzed and their effects on the solution are observed. Given the two factors, the optimal solution can be found when the variable expenses are low and the fixed expenses are high. Two methodologies to achieve the optimal solution were compared: sequential heuristics method and integrated heuristic method. The sequential approach considers one side of the distribution first, to which the results would be considered as input on the second run. Using this heuristic, it would only consider the optimal solution of one side of the distribution. The integrated heuristic is better than the sequential heuristic because it considers the forward and reverse distribution system simultaneously. Thus, would be able to assign the best location for both the reverse and forward logistics.