Appointment Requests from Multiple Channels: Characterizing Optimal Set of Appointment Days to Offer with Patient Preferences

Date/Time
Date(s) - 26/11/2021
01:30 - 03:30

Categories No Categories


Speaker: Lerzan Örmeci

Date & Time: November 26, 2021, Friday, 13:30

Zoom Link:

https://zoom.us/j/6547746234?pwd=ZENZNWtCbUlQRjVMMVFneWtxZGlzZz09

Abstract: We consider the problem of appointment scheduling for a physician or a diagnostic resource in a healthcare facility. Patients contact the facility either through a call center where patients are scheduled immediately, or through a website which produces a list of patients to be scheduled the following morning. Patients are further categorized into two types according to the revenues they generate, possibly depending on their insurance types, and their day preferences, represented through multinomial logit models. The facility aims to maximize the long-run average revenue, while ensuring that a certain service level is satisfied for patients generating lower revenue. The facility has two types of decisions: offering a set of appointment days to a patient and choosing the patient type to prioritize while contacting the website patients. We first model this system as a periodic Markov Decision Process (MDP) model, called Model 1. We characterize the structure of the optimal policy. Then we combine simulation optimization with approximate dynamic programming techniques to develop a booking limit improvement algorithm, which generates a well-performing booking limit policy. Next, we build the constrained MDP model that accommodates the service level constraint, called Model 2. We show that the optimal policy of this model is a randomized policy with a special structure. Based on this result, we develop a new algorithm that uses the booking limit improvement algorithm repeatedly to identify a feasible and well-performing policy for Model 2. The performance of this policy is evaluated in a numerical study based on the case of a university hospital through a comparison with certain benchmark policies. (This is a joint work with Feray Tuncalp.)

 

Bio: Lerzan Örmeci received her BS and MS degrees in Industrial Engineering from Middle East Technical University, Ankara, Turkey in

1990 and 1993, respectively. She completed her PhD in Operations at Case Western Reserve University in 1998. She joined the department of Industrial Engineering at Koç University in 2001. Previously, she worked as a research fellow at Eurandom (Eindhoven, Netherlands), and as an instructor at Erasmus University (Rotterdam, Netherlands). She was a visiting scholar at Wharton School of Management (University of

Pennsylvania) and HEC Paris. Her research focuses on stochastic modeling and analysis of health care systems.