ANALYSIS OF COORDINATION MECHANISMS IN DECENTRALIZED, HIERARCHICAL DECISION MAKING PROCESSES (OPERATIONS RESEARCH, TRANSFER PRICING, DECISION PROCESS MODELS).

摘要

This research concentrates on the analysis of coordination procedures in decentralized decision processes. The decision process under consideration concerns the allocation of resources for the production of one or more products in a hypothetical organization characterized by multiple organizational units and levels, shared corporate resources, and goods and/or services transferred between divisions of the organization.;Composition-based mathematical programming models, which are shown to be more descriptive of decision processes than decomposition based models are used as the vehicle for analysis. An investigation of dual variable-based transfer pricing approaches confirms the severe limitations of using such approaches for coordination. Thus, a new method of valuing goods and coordinating the production and flow of divisional products is developed. The method is based on flows of information other than dual variables.;The new method, called a price schedule algorithm, enacts coordination through the levying of subsidies and/or taxes on the production of divisional products. The value of the subsidy or tax is determined in a negotiation process between the central unit and the divisions, and is based on sensitivity analysis information. The decision process models evidence learning at both levels of the organization and are shown to satisfy a set of criteria constructed to judge how well a particular decision process model represents a particular decision process. The model is also used to drive a simulation operating in an experimental design setting to illustrate the flexibility of the composition-based mathematical decision process model for investigating research hypotheses concerning decision processes.;The goal of the research is to merge the quantitative and accounting literatures dealing with decentralized decision making. The result is a framework that joins mathematical programming approaches to resource allocation with transfer pricing approaches to control and coordination.