Warranty Inventory Optimization with Multiple Sources of Product
Speaker
Abstract
| In warranty inventory management, customers return allegedly malfunctioning products for replacement. Useful products may be recovered through testing and/or remanufacturing processes. The company must decide on the percentage of units to send to the testing and remanufacturing processes and the number of new units to purchase from a production line. These decisions depend on an array of complex relationships among stochastic demand rates, intricate cost structures, probabilistic yields from both the testing and remanufacturing processes, multiple sources of supply originating from both the stochastic reverse channel and the company’s purchasing decisions, and varying levels of information regarding reverse pipeline inventory. In this paper, we combine all of these elements to formulate a model that analyzes both strategic and tactical decisions. First, we perform a cost analysis in order to determine the optimal percentage of units to test and remanufacture. Next, we use dynamic programming to develop several models which determine the optimal ordering decisions under various levels of reverse channel visibility. “The Curse of Dimensionality” prohibits us from solving for optimal policies in most practical cases; thus, we develop heuristic dynamic programs using an aggregated state space which allow for tractable models while incorporating information gained from the pipeline visibility; we call this latter concept Advance Supply Information. |
| Contact information: |
| Dr. K.J. Roodbergen |
