A. Tobail, J. Crowe, A. Arisha

Dublin Institute of Technology (DIT) (IRELAND)
The high level of complexity and uncertainty and various sources of risks create challenges for achieving a satisfactory performance across supply chain networks. Advances in Information Technology can enable supply chain decision makers to predict the magnitude of risks related to their decisions. The proposed framework presents a solution that integrates intelligent-agents, simulation modelling, and optimisation. An interactive, friendly, animated, web-based interface is designed especially to engage the user into a serious game environment. Each user will play a role in the supply chain network and will encounter the consequences of his/her decisions. The optimisation engine developed within the framework is used to advise the users on the optimum decisions and anticipated performances based on these decisions. Genetics Algorithm (GA) and Case-Based Reasoning (CBR) are used to enhance the decision quality. To realise the communication between the client and the server, a high-level communication protocol is designed, developed and implemented. This protocol enables the intelligent-agents to communicate together easily and efficiently. The integration between multi-agent structure and genetic optimisation engine gains benefits from the compatibility in structure and case description by efficiently map the agent structure and case variables to genes. The proposed framework can also be used as training or teaching tool.