Scalable Enterprise Systems: Integration of Optimization, Inventory Control and Dynamic Models for the Management of Large-Scale Supply Chains

Carnegie Mellon University, Pittsburgh PA

Investigators

Abstract

This project addresses the problem of running enterprise systems in a flexible, responsive, and reliable manner to capture the economical potential that the market might offer, while minimizing the impact of fluctuations in the market. To achieve these goals, systematic control and optimization tools will be developed that can be used to assess the ability of supply chain systems to react to fluctuations, as well as to control the system to improve its responsiveness. Large scale problems will be addressed, the plan is to investigate the integration of supply optimization, inventory control and dynamics to achieve cost optimal performance, while reducing inventory levels and coping effectively with disturbances in supplies, orders and delivery times. The specific objective objectives are to develop a novel framework for multi-level planning that integrates optimization, inventory control and dynamics in the context of a supply chain with decentralized control. Moreover, to investigate a novel approach to model the dynamics and control of the entire supply chain. For the implementation, the plan is to develop a prototype of an internet-based system that will allow the effective collaboration among the various parties in the supply chain. A major expected benefit of the research is the creation of a science base that provides a framework and guidance for developing decision support tools for enterprise systems that leverage the new information technologies. Regarding the proposed integration of optimization, inventory control and dynamics, the major benefit will be the coordinated decision making in the supply chain that operates as a decentralized system, which reflects more closely the real-world behavior of these systems. Finally, the research should provide a scalable enterprise solution, which is urgently needed as software companies are scrambling to meet the demand, and companies are investing in automated systems without a coherent plan or understanding.