In the face of the seriously economical crisis in today?s world, businesses in each industry should save costs and make an effort to find profitable opportunities in the situation where the global economics is in a state of deep depression. The main purpose of this study is to construct an integrated VMI system. The VMI system (e.g., a three-level supply chain) includes a core TFT-LCD company (i.e., AU Optronics Corporation (AUO)) and its up-stream vendors. This study first defines the notations of operational variables and makes the basic assumptions. Then, an optimization package (e.g., LINGO) is used for the small-scale problems since it has been proven to be NP-hard. For the reason of difficultly obtaining an optimal solution, an effective heuristic method (i.e., genetic algorithm) is applied for saving the large-scale problems. In addition, an experiment and a computational efficiency analysis will be also conducted. It is hoped that our model can be used as an effective approach for VMI supply chain practitioners to cooperatively make their procurement, production, and delivery decisions, to maintain long-term stable partnerships, and to improve global competition. Most importantly, the results of this empirical study are shown to be correct since they are extremely similar to the view points of practitioners and real-world practices. We expect to provide practitioners who plan to install VMI systems in the future with the experience of the three case companies. Finally, the proposed mathematical programming model and the heuristic method are verified by this research for the aptness of practical use.
