Scheduling policies for a repair shop problem

Liang, William Kun and Balcıoğlu, Ahmet Barış and Svaluto, Robert (2013) Scheduling policies for a repair shop problem. Annals of Operations Research, 211 (1). pp. 273-288. ISSN 0254-5330 (Print) 1572-9338 (Online)

PDF (This is a RoMEO green journal -- author can archive pre-print (ie pre-refereeing) and post-print (ie final draft post-refereeing)) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader

Official URL: http://dx.doi.org/10.1007/s10479-013-1412-6


In this paper, we analyze a repair shop serving several fleets of machines that fail from time to time. To reduce downtime costs, a continuous-review spare machine inventory is kept for each fleet. A spare machine, if available on stock, is installed instantaneously in place of a broken machine. When a repaired machine is returned from the repair shop, it is placed in inventory for future use if the fleet has the required number of machines operating. Since the repair shop is shared by different fleets, choosing which type of broken machine to repair is crucial to minimize downtime and holding costs. The optimal policy of this problem is difficult to characterize, and, therefore, is only formulated as a Markov Decision Process to numerically compute the optimal cost and base-stock level for each spare machine inventory. As an alternative, we propose the dynamic Myopic(R) policy, which is easy to implement, yielding costs very close to the optimal. Most of the time it outperforms the static first-come-first-served, and preemptive-resume priority policies. Additionally, via our numerical study, we demonstrate that repair shop pooling is better than reserving a repair shop for each fleet.

Item Type:Article
Uncontrolled Keywords:Spare part inventory control; Multiple finite-population queueing systems; Static and dynamic repair scheduling; Repair shop pooling
Subjects:Q Science > Q Science (General)
ID Code:24033
Deposited By:Barış Balcıoğlu
Deposited On:14 Apr 2014 10:45
Last Modified:02 Aug 2019 09:59

Repository Staff Only: item control page