Decision Sciences Volume 42 Number 3 August 2011 C  2011 The Authors Decision Sciences Journal C  2011 Decision Sciences Institute A Profit-Maximizing Approach to Disposition Decisions for Product Returns * Mark E. Ferguson Moore School of Business, University of South Carolina, Columbia, SC 29208, e-mail: mark.ferguson@moore.sc.edu Moritz Fleischmann University of Mannheim, Business School, PO Box 10 34 62, 68131 Mannheim, Germany, e-mail: Moritz.Fleischmann@bwl.uni-mannheim.de Gilvan C. Souza † Kelley School of Business, Indiana University, Bloomington, IN 47405, e-mail: gsouza@indiana.edu ABSTRACT Motivated by the asset recovery process at IBM, we analyze the optimal disposition decision for product returns in electronic products industries. Returns may be either remanufactured for reselling or dismantled for spare parts. Reselling a remanufactured unit typically yields higher unit margins. However, demand is uncertain. A common policy in many firms is to rank disposition alternatives by unit margins. We propose a profit-maximization approach that considers demand uncertainty. We develop single period and multiperiod stochastic optimization models for the disposition problem. An- alyzing these models, we show that the optimal allocation balances expected marginal profits across the disposition alternatives. A detailed numerical study reveals that our approach to the disposition problem outperforms the current practice of focusing exclu- sively on high-margin options, and we identify conditions under which this improvement is the highest. In addition, we show that a simple myopic heuristic in the multiperiod problem performs well. [Submitted: August 2, 2010. Revisions received: January 18, 2011; February 3, 2011. Accepted: February 8, 2011.] Subject Areas: Closed-Loop Supply Chains, Inventory Management/ Coordination, Markovian Decision Processes, and Yield Management. INTRODUCTION In this article, we address the disposition decision that a manufacturer has to make for product returns. We motivate our analysis by describing the asset recovery process at IBM. Information technology products such as personal computers, servers, storage systems, and mainframes coming off of lease are returned to ∗ We thank the IBM executives for exposing us to this problem. † Corresponding author. 773