Computational Modelling of Multi-Level/Multi-Unit Organizations

Johns Hopkins University, Baltimore MD

Investigators

Abstract

Computational Modelling of Multi-Unit/Multi-Level Organizations One of the most important organizational forms in modern economies is the multi-unit organization. In manufacturing, the typical plant is part of a multi-plant manufacturer. In services, the typical store is part of a retail chain. A defining problem of a multi-unit organization is how to balance giving units the freedom to handle the idiosyncratic features of their environments while, at the same time, coordinating the units to take advantage of the commonality among them. How do multi-unit organizations solve this problem and what is the most effective means of doing so? What is the relevant set of instruments and what are the relevant features of the environment determining the best organization and strategy? This project explores these questions and more broadly advances our understanding of multi-unit organizations. Our focus is on retail chains. The central idea underlying our approach is that much of the behavior and performance of a retail chain can be explained by understanding how chains generate, communicate, and use information. How new ideas arise and how they are evaluated and adopted. How the authority to use information is distributed within the chain and how incentives are structured to influence the creation and effective dissemination of information. And how competition and technology interact with these information processes. Implementation of this approach involves the development of a computational model of a retail chain in which corporate headquarters (HQ) and stores are adapting to their environments. HQ and store managers develop new ideas, evaluate their ideas and the ideas of others, and then decide whether to adopt and communicate them to other members of the organization. Competing chains are simultaneously engaging in an analogous exercise while consumers experience and experiment among stores to find the one that best suits their needs. Our analysis explores how various factors influence the rate of improvement of store practices and chain efficiency. These factors include organizational structure (in particular, the authority of store managers to change practices), store manager incentives (as influenced by how they are compensated), the degree of competition within and across geographic markets (and, relatedly, the effect of mergers), the amount of consumer search, the rate of change in consumers' preferences, the degree of informational spillovers across chains, and advances in information technology.