Allocation methods for a multicommodity distribution network design problem
- Pathum Thani : Asian Institute of Technology, 2012
- 72 p. : col. ill., charts
- Thesis ; no. ISE-12-13 .
- Asian Institute of Technology. Thesis ; no. ISE-12-13 .
A thesis submitted in partial fulfillment of the requirements for thedegree of Master of Engineering inIndustrial and Manufacturing Engineering, School of Engineering and Technology
Thesis (M. Eng.) - Asian Institute of Technology, 2012
This research presents new allocation schemes which are 2highest-total-demand customer to nearest DC3 and 2nearest all types of facilities3 to solve multicommodity distribution network design problem (MDNP). This is a strategic planning model where the important decisions to select of the locations of suppliers, plants and distribution centers (DCs), and to make allocation decisions about the number of raw materials and products from suppliers to plants, plant to DCs and DCs to customers so as to minimize the total costs. A mathematical model is formulated for the 3-stage MDNP (suppliers, plants, DCs) and also includes the bill of material so that a product can be made from several raw materials. The comparisons between two metaheuristics which are Differential Evaluation (DE) algorithm and Particle Swarm Optimization with multiple social learning terms (GLNPSO) are made to find which method is better in solving MDNP. A new way of finding parameter setting of GLNPSO is proposed based on self adapted parameters.The proposed allocation methods are then compared using the GLNPSO version of the MDNP algorithm.The 2highest-total-demand customer to nearest DC3 approach is then compared with2nearest all types of facilities3 algorithm on the 3-stage cases except the largest case.