Valuing Information In Complex Systems: An Integrated Analytical Approach To Achieve Optimal Performance In The Beer Distribution Game

Keywords

Beer distribution game; Control theory; Decision analysis; System dynamics; Value of information

Abstract

Even seemingly simple systems can produce complex dynamics, which leads management professionals to develop tools for training, monitoring, and improving performance. Management simulators provide useful insights about human behavior and interactions, while computational and informational decision support tools offer opportunities to reduce inconsistencies, errors, and non-optimal human choices, particularly for complex systems that involve multiple decision makers, uncertainty, variability, and time. We use the context of a popular management simulator that teaches students about the bullwhip effect (i.e., the beer distribution game) to explore an integrated decision analytic, control theory, and system dynamics approach to the game that recognizes the value of available (imperfect) information and considers the value of perfect information to provide the optimal strategy. Using a discrete event simulation, we characterize optimal decisions and overall team scores for the situation of actual available information and perfect information. We describe our implementation of the strategy in the field to win the 2007 beer game world championship played at the 25th conference of the International System Dynamics Society. This paper seeks to demonstrate that better understanding of the system and use of available information leads to significantly lower expected costs than identified in prior studies. Understanding complex systems and using information optimally may increase system stability and significantly improve performance, in some cases even without better information than already available.

Publication Date

12-17-2015

Publication Title

IEEE Access

Volume

3

Number of Pages

2677-2686

Document Type

Article

Personal Identifier

scopus

DOI Link

https://doi.org/10.1109/ACCESS.2015.2505730

Socpus ID

84961076875 (Scopus)

Source API URL

https://api.elsevier.com/content/abstract/scopus_id/84961076875

This document is currently not available here.

Share

COinS