This paper proposes a new framework that integrates simulated and physical agents to provide an efficient way for companies to form supply chains dynamically. In this integrated framework, physical agents coordinate with inter-organizational physical agents to conduct business processes whereas simulated agents model and analyze business processes to support physical agents in making rational decisions under uncertain situation and with incomplete information. This paper surveys different techniques used for dynamic process coordination and explains how the proposed integrated framework can be used by companies to reach a commonly accepted goal in dynamic supply chains. This paper also elaborates the efficient supply chain formation using a business process example of the mold industry, and finally discusses the development issues of this framework and future research directions.