Abstract
Multi-agent systems (MASs) have emerged as the most natural intrinsic paradigm for modelling complex systems, such as supply chains (SCs). However, their development process remains quite involved and extremely time consuming, hindering wide-spread adoption in industrial-strength applications. A software agent-component based framework is proposed to simplify and speedup multi-agent (MA) modelling and simulation for SC applications. With the help of pre-developed libraries of reusable components—organizational agents (OAs), SC agents (SCAs), behaviour and policy objects; the framework allows model developers to quickly configure a MAS to simulate SC dynamics and study control and coordination issues. Being generic, flexible, and scalable; it supports development of either pseudo-centralized models by a single model developer, or distributed models by either a single or group of enterprises constituting a SC. The framework is unique; based on requirements it allows for representing different segments of SC network at either aggregated or detailed levels resulting in models of hybrid resolution. It facilitates studies involving intra- and inter-organizational dynamics (either independently or collectively), considering information asymmetry explicitly. The framework is validated through MA-simulation of Tamagotchi SC. Its results are presented, and the research extensions outlined.
Additional information
Notes on contributors
R. Govindu
Ramakrishna Govindu works as a Management Consultant in Chicago. He received his B.S. degree in Mechanical with specialization in Production Engineering from Osmania University (India), M.S. degree in Quality, Reliability and Operations Research from Indian Statistical Institute (ISI), and Ph.D. degree in Industrial Engineering from Wayne State University. As a management consultant he executed numerous projects relating to Business Reengineering, Operations Research and IT applications, and Productivity/Process/Quality Improvement; in a wide-range of industries in the United States and in India. During his doctoral studies, as research and teaching assistant in the Ford and Visteon Engineering Management Masters Program (EMMP), he rendered analytical and modelling help for leadership projects. He also teaches several undergraduate and graduate courses regularly. He has research interests in modelling and decision support for supply chain and enterprise system applications using traditional O.R. based approaches, multi-agent systems, and information technology driven solutions. He is a member of INFORMS, IEEE, IIE, SIAM, and ASQ.
R.B. Chinnam
Ratna Babu Chinnam is an associate professor with the Department of Industrial and Manufacturing Engineering at Wayne State University. He received his B.S. degree in Mechanical Engineering from Mangalore University (India) and his M.S. and Ph.D. degrees in Industrial Engineering from Texas Tech. He authored over 50 technical publications in the areas of smart engineering systems, intelligent manufacturing, autonomous diagnostics and prognostics, advanced quality and reliability engineering, operations management, supply chain management, and computational intelligence. His publications appeared in IEEE Transactions on Reliability Engineering,Semiconductor Manufacturing, Neural Networks; and International Journal of Production Research. He regularly serves on conference planning committees for IJCNN, ANNIE, and IASTED’s NIC. He conducts extensive collaborative research with Ford and DaimlerChrysler. He received research funding from the NSF; and consulted for such companies as Energy Conversion Devices and Tecton. He is a member of Alpha Pi Mu, INFORMS, International Neural Network Society, and the North American Manufacturing Research Institute.