Developing Adaptive Strategies in Supply Chain Management

Supply Chain Management (SCM) is a complex process which includes a number of activities such as: negotiating with suppliers for raw materials, competing for customer orders, managing inventory, scheduling production, and delivering goods to customers. The activities are interconnected and interdependent; it is not obvious which strategies to follow in each part of the supply chain in order to achieve better overall performance. This paper introduces a multi-agent system as a tool for investigating SCM strategies. The system was tested in the Trading Agent Competition SCM game, which offers a realistic simulated environment for studying SCM approaches. The experimental results showed that the ability to predict and schedule in advance is vital in each stage of the chain. Among others, the paper concentrates on the strategies for predicting customer order prices. In particular, it demonstrates that an advantage over competitors can be gained by modeling their behavior and predicting their prices. We use neural networks learning technique for making price predictions.