Evaluation and optimization methods based on e-business for vendor selection

In order to solve the multi-objective optimization problem of vendor selection, this paper analyzes the influence of e-business technologies, and figures out the necessity of building an index system and improving the evaluation methods after studying many related literatures. Some new indexes including many quantitative and qualitative ones are added to the traditional measurement index system, in consideration of rebuilding it under e-business environment. Further, three evaluation methods are researched: AHP (Analytical Hierarchical Process), DEA (Data Envelopment Analysis) and AHP/DEA. DEA and AHP methods are usually used in vendor selection, but both have limitations. AHP/DEA model has two stages: first, the DEA is run for each pair of vendors separately, second, the pair-wise evaluation matrix generated in the first stage is utilized to rank the vendors via the AHP. This competitive model AHP/DEA takes the best use of both methods, by avoiding the pitfalls of each. At the end of the paper, an instance of vendor selection is studied to compare the three methods, and the validity of the AHP/DEA method is proved.