Outsourcing warranty repairs: Dynamic allocation

Abstract

In this paper we consider the problem of minimizing the costs of outsourcing warranty repairs when failed items are dynamically routed to one of several service vendors. In our model, the manufacturer incurs a repair cost each time an item needs repair and also incurs a goodwill cost while an item is awaiting and undergoing repair. For a large manufacturer with annual warranty costs in the tens of millions of dollars, even a small relative cost reduction from the use of dynamic (rather than static) allocation may be practically significant. However, due to the size of the state space, the resulting dynamic programming problem is not exactly solvable in practice. Furthermore, standard routing heuristics, such as join-theshortest-queue, are simply not good enough to identify potential cost savings of any significance. We use two different approaches to develop effective, simply structured index policies for the dynamic allocation problem. The first uses dynamic programming policy improvement while the second deploys Whittle's proposal for restless bandits. The closed form indices concerned are new and the policies sufficiently close to optimal to provide cost savings over static allocation. All results of this paper are demonstrated using a simulation study.

Comments on: Dynamic priority allocation via restless bandit marginal productivity indices

“…3.2) and wireless communication (Sects. 4.3, 4.4), but also in, e.g., machine maintenance ) and outsourcing of warranty repairs (Opp et al 2005). Numerical studies like his (2007b) paper show that his MPI policy is often near-optimal and considerably better than existing policies.…”

“…As the author remarks at the end of his survey, many interesting issues remain to be explored. Issues that might be added are (i) a systematic comparison with other approaches to develop effective, simply structured index policies such as dynamic programming one-step policy improvement (see, e.g., Opp et al 2005;Sassen et al 1997) and (ii) a comparison with the 'more traditional' static or dynamic optimization of polling systems (see e.g., Borst 1994;Weststrate 1992;Yechiali 1991).…”

Allocation Models and Heuristics for the Outsourcing of Repairs for a Dynamic Warranty Population

“…Opp et al (2003) quote a Merill Lynch report (Serant 2001) to the effect that this trend represents a one hundred billion dollar opportunity for subcontractors and service vendors. Such outsourcing enables manufacturers to focus on their core business and saves the costs involved in maintaining in-house repair facilities.…”

“…In response to the first-stage problem above, Opp et al (2003) and Ding and Glazebrook (2005) have formulated simple static allocation models in which it is supposed that there is a fixed number (K) of items under warranty for all time. These are to be divided between the (V ) vendors, vendor v receiving a fixed allocation (k v ) of the items which will be under the vendor's care.…”

“…However, the decision maker is not able to observe the current repair queue at each vendor which is the locus at which costs are actually incurred. Continuous observation of all vendor repair queues would involve a substantial administrative overhead but results in a simpler fully observable stochastic dynamic optimization problem (Opp et al 2005). That analysis is of a relatively conventional model concerning the dynamic routing of incoming items to alternative service stations to minimize average cost rates over an infinite horizon.…”