Proceedings of the 2010 Industrial Engineering Research Conference A. Johnson and J. Miller, eds. Supervisory Control of a Synchronized Supply Chain Using Petri Nets Abstract ID: 504 Julie Drzymalski Department of Industrial Engineering Western New England College, Springfield MA 01119 USA Nicholas G. Odrey Department of Industrial and Systems Engineering Lehigh University, Bethlehem PA USA Abstract This research presents a framework for a multi-echelon, multi-stage synchronized supply chain (SSC) that is reactive to supplier changes and has the objective of maintaining the due date. Modeling of the SC environment is accom- plished using a Generalized Stochastic Petri net (GSPN) which is hierarchical and modular. The hierarchy includes a Supply Chain Manager (SCM) controlling the procurement activities of the lower SC members through the use of supervisory control nets. However, since the supplier subset is dynamic, traditional supervisory control net methodol- ogy for Petri nets was extended to a dynamic situation. This dynamic control net technique was further decentralized to the facility level by a technique utilizing Petri net transition guards which enables a Petri net state space analysis used to calculate remaining delivery times for a large-scale supply chain. Two delivery leadtimes distributions were considered: normal and beta and are conditioned upon one of 3 discrete supplier states. A chance constrained program determines the optimal supplier subsets and order quantities under leadtime uncertainty. The results of this research resulted in a dynamically controlled, reactive SSC which in certain cases resulted in lower overall costs, improved on-time shipment performance and lower total inventory than an uncontrolled SC. Keywords supply chain, Petri nets, supervisory control 1 Introduction Today’s business environment is highly competitive and global. More and more companies are relocating their pro- ductions overseas or using foreign sources. This can make a supply chain (SC) difficult to coordinate and induces risk and the potential for noise and unreliability. Supply chain disruptions can be categorized broadly as uncertainty in the demand (downstream) side or supply (upstream) side and can stem from a wide range of sources: natural disasters, labor disputes or strikes, logistic problems and quality issues to name a few [3] [14] [4]. The root cause can be either internal or external to the company. The consequences of a disruption can have an extensive financial impact on com- panies. In 2000, Becker [1] reported that high growth firms (firms which less risk averse) saw a 10.81% drop in stock on a day with a supply chain glitch while firms which are more risk averse (small-growth) saw about a 5.89% loss. In a follow-up study in 2004, supply-chain failures averaged about 7% lower sales growth, 11% higher costs and a 14% increase in inventories, averaging over all types of firms [2]. Due to demanding and unstable business environments, companies must be able to quickly react to disturbances from outside sources. Supply chain modeling has become necessary as companies face more complex and global interac- tions. Theoretical models exist to enable firms to plan for unforeseen errors [7][4][12][13]. Gaonkar and Viswanadham [7] developed two models which determine optimal partnering selection; the first minimizes operational cost and its variability and the second minimizes risk of backorders in the event of a supply disruption. Sheffi [12] lists three