An improved way based on rough set for multi-stage reverse logistics processing center location

• Published in: 2009 International Conference on Machine Learning and Cybernetics Vol. 1; pp. 138 - 144
• Main Author: Jiang-Hong Man
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.07.2009
• Subjects: Computer network management; Conference management; Costs; Cybernetics; Discretization; Fuzzy sets; Information entropy; Information systems; Location problem; Machine learning; Processing center; Reverse logistics; Rough set theory; Set theory
• ISBN: 9781424437023; 1424437024
• ISSN: 2160-133X
• DOI: 10.1109/ICMLC.2009.5212561

In this paper, firstly a multi-stage reverse logistics network model is described and an index system for reverse logistics processing center location is presented. Secondly, a programming model is designed for the system where the values are continuous. Thirdly, the computational results is transformed into an information system, where one potential processing centers is regarded as a subject, and its values of lower index are regarded as condition attributes, the location result of it is regarded as decision attribute. And then, rough set way is introduced to the information system to mine the relation between the location result and factors such as built cost, shipping cost, maintenance cost to build a knowledge base. By doing so, when the conditions are changed, new decision advice can be given by checking the knowledge base instead of recalculating the programming model. Finally, the rules mined from decision table are used to instruct the decision support system for reverse logistics processing center location.