ABSTRACT
Operating room inventories typically involve hundreds of surgical items. Managers require very high service levels since the cost of shortages can be excessive up to and including cancelling surgery. From our experience, many inventory managers rely on manual approaches and intuition to set inventory control parameters. Although there are classical theoretical methods available for deriving “optimal” inventory policies, these classical methods rely on assumptions that do not accurately represent typical operating room inventories. Using 5 years of data from a large Canadian hospital, we use simulation and a simple search heuristic to find the optimal (s, S) ordering policy and show that 1) current hospital methods dramatically underperform with respect to service level and 2) there are significant savings to be realised over the best available classical theoretical models. An example case testing potential lead time changes is discussed.
Acknowledgments
Thank you to Prof. Vahid Sarhangian for his advice at the beginning of this project and input when preparing this manuscript. Thank you to the team at London Health Sciences Centre including Susan Reiber, Cheryl Churcher, and Brenda Maxwell for providing access to data and contextual information.
Disclosure statement
No potential conflict of interest was reported by the author(s).