A Knowledge Discovery Way for Reverse Logistics Center Location

In this paper, we present an index system for reverse logistics processing center location. And then we put forward a decision table based on the index system. In the decision table, one potential processing center is regarded as one subject, and the values of its index are regarded as the condition attributes of the table. But because all of the values are continuous, we must discrete them in order to transform the decision table into an information system so as to use rough set way on it. In the information system, the location result of one potential processing center is regarded as the decision attribute of the object. Furthermore, rough set way based on distinguish matrix is introduced to the information system. By doing so, we can get decision rules between the location result and the factors that lead to this result, such as built cost, shipping cost, and so on. All of these rules got from the decision table can construct a knowledge base, and the rules mined from decision table can be used to get decision advice for reverse logistics processing center location when the condition changed.