Sensible Agents in Supply Chain Management: an Example Highlighting Procurement and Production Decisions

Agent-based technologies can be applied to many aspects of supply chain management. The need for responsive, flexible agents is pervasive in this environment due to its complex, dynamic nature. Two critical aspects of agent capabilities are the ability to (1) classify agent behaviors according to autonomy level and (2) adapt problem-solving roles to various problem-solving situations during system operation. Sensible Agents, capable of Dynamic Adaptive Autonomy, have been developed to address these issues. A Sensible Agent’s “autonomy level” constitutes a description of the agent’s problem-solving role with respect to a particular goal. Problem-solving roles are defined along a spectrum of autonomy ranging from command-driven, to consensus, to locally autonomous/master. Dynamic Adaptive Autonomy is a capability that allows Sensible Agents to change autonomy levels during system operation to meet the needs of a particular problem-solving situation. This paper provides an overview of the Sensible Agent Testbed and introduces an example supply chain management domain with a scenario showing how this testbed could be used to simulate agent-based problem solving. INTRODUCTION Supply chain management systems are inherently complex and dynamic. The use of agent-based systems offers significant benefits to a supply chain management system through adaptable automated or semi-automated problemsolving and distribution of control and processing. However, simply applying the agent-based paradigm to supply chain management problems may not be enough to address the realtime demands of these systems. Agent-based systems operating in the supply chain management domain are subject to dynamic situational changes across many dimensions: • certainty of information held by an agent or acquired by an agent about other agents in the system (e.g. speculation about future trends in supply and demand); and • resource accessibility for a particular agent (e.g. machines, money, products); and • goal constraints for multiple goals (e.g. deadlines for goal completion, goal priorities); and • environmental states (e.g. cpu cycles, communication bandwidth). Therefore, supply chain management systems require agent-based problem solving to be flexible and tolerant of faulty information, equipment, and communication links. This research uses Sensible Agent-based systems to extend agent capabilities in dynamic and complex environments (Barber, 1996). Sensible Agents achieve these qualities by representing and manipulating the interaction frameworks in which they plan to achieve their goals. Agent interactions for planning can be defined along a spectrum of agent autonomy as shown in Figure 1. An agent’s SPECTRUM OF AUTONOMY