Risk assessment model based on multi-agent systems for complex product design

Design of a complex product takes a long and intricate process, and mostly, it has the multidisciplinary nature. The design can become even challenging if the product to be designed is totally new since numerous technical challenges and uncertainties are involved. In this paper, the risk assessment for complex product design is discussed, and the quantification of the risk is focused. A new risk model has been proposed; which is based on Multi-Agent Systems (MAS). With the consideration of the similarity between the design of a product and that of an adaptive system, a hierarchical network is used to represent the design process based on an agent-based approach. In the proposed model, the concept of material flow is conceived to describe the interactions among agents, and the risks associated with the design activities are analyzed using the probabilistic assessment tools to simulate the risk propagation. The risk assessment is parameterized and implemented in the platform of the Recursive Porous Agent Simulation Toolkit (Repast), and the risk level was quantified by using the Monte Carlo Simulation method. To demonstrate the effectiveness of the proposed model, experiments are conducted with two typical scenarios, and simulation results are analyzed in details. Finally, the future work based on the developed model is discussed.