An options-based approach to coordinating distributed decision systems
| Autor: | Nikunj Kapadia, Daniel R. Ball, Abhijit Deshmukh |
|---|---|
| Rok vydání: | 2015 |
| Předmět: |
Information Systems and Management
Knowledge management General Computer Science business.industry Multiple forms Computer science Multi-agent system Information technology Management Science and Operations Research Industrial and Manufacturing Engineering Risk analysis (engineering) Distributed decision Modeling and Simulation Uncertainty modeling State (computer science) business Risk management |
| Zdroj: | European Journal of Operational Research. 240:706-717 |
| ISSN: | 0377-2217 |
| DOI: | 10.1016/j.ejor.2014.05.037 |
| Popis: | Engineering and operations management decisions have become increasingly complex as a result of recent advances in information technology. The increased ability to access and communicate information has resulted in expanded system domains consisting of multiple agents, each exhibiting autonomous decision-making capabilities, with potentially complex logistics. Challenges regarding the management of these systems include heterogenous utility drivers and risk preferences among the agents, and various sources of system uncertainty. This paper presents a distributed options-based model that manages the impact of multiple forms of uncertainty from a multi-agent perspective, while adapting as both the stream of information and the capabilities of the agents are better known. Because the actions of decision makers may have an impact on the evolution of underlying sources of uncertainty, this endogenous relationship is modeled and a solution approach developed that converges to an equilibrium system state and improves the performance of agents and the system. The final result is a distributed options-based decision-making policy that both responds to and controls the evolution of uncertainty in large-scale engineering and operations management domains. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect