Probability distributions of cost, revenue and profit over a warranty cycle

A warranty cycle starts with the sale of a new item and terminues when the warranty servicing obligation of the manufacturer created by this sale ends. Cost and profit measures over a warranty cycle are of interest with regards to accounting, pricing, product quality, warranty policy design, and other issues. For most predictive and decision making purposes, these variables cannot be described adequately by their means alone. In this paper, we obtain the probability distributions of the manufacturer's rebate, cost, revenue and profit during a cycle, under a (combination) free-replacement/pro-rata warranty policy, with the incorporation of the customer repurchase behavior under warranty. We first derive the joint distribution of the numbers of free and pro-rata replacements during a cycle. We then construct and compute the distributions of interest by conditioning on the numbers of such replacements. In so doing, we take into account the warranty policy, in terms of the lengths of the free and pro-rata periods, the cost and rebate profiles, customer repurchase probabilities during a pro-rata period, and the failure time distribution of the product. We provide several numerical examples, using data that are typical of battery warranties.