Intelligent Systems for Supply Chain Automation

Multi-agent systems have promised to revolutionise decentralised industrial operations by equipping them with learning capability, flexibility and autonomy. A use cases of particular interest is distributed supply chain planning. In this case, agents are used as a mechanism to automate decision making and negotiation between industrial partners, and learn from past actions to optimise outcomes collectively. Past research has shown that this approach can significantly reduce errors and lead times in mundane supply chain operations. However, several of the characteristics that make the approach ideal for these applications, such as autonomy, decentralised control, collective emergent behaviour, hinder it from being applied to real-world problems, because companies can not trace how agents reach decisions, and fear that confidential sales and demand data will be shared by agents, and therefore do not trust in the technology.

This PhD research will explore the use of secure multi-agent systems to automate and optimise decentralised decision making across complex supply chains. Research will be conducted on how a group of agents can reach an agreement on a particular state of affairs and record that agreement without the need for a controlling authority, thereby creating decision traceability and increase trust in distributed supply chain planning.