Fixed costs and shared resources allocation in two-stage network DEA

Abstract

Data envelopment analysis (DEA) is an approach for performance evaluation and benchmarking. In many real managerial decisions, fixed costs are used in building common facilities used by an entire organization, and need to be shared by sub-components in decision-making units (DMUs). The current paper discusses the problem of fixed cost allocation when DMUs are composed of two linked stages. We develop a set of DEA models to measure the performance by treating fixed cost as an additional input factor shared in two stages among a set of DMUs. We propose three procedures based on different objectives in reality to obtain a fair cost allocation plan. The models are developed under the assumption when the DEA best-practice is of the constant returns to scale, but can be readily applied under the assumption of the variable return to scale (VRS). The proposed models are illustrated using a randomly generated data set.