Operations & Logistics Division

Research seminars

Current seminars

Date: Friday, August 23rd, 2019

Speaker: Reza Skandari, Imperial College Business School, London
Title:  "Patient-Type Bayes-Adaptive Treatment Plans: the Case of Chronic Kidney Disease"
Time: 12:30 PM - 1:30 PM
Place: HA 966

Abstract: Patient heterogeneity in disease progression is prevalent in many settings. Treatment decisions that explicitly consider this heterogeneity can lower the cost of care and improve outcomes by providing the right care for the right patient at the right time. In this paper, we analyze the problem of designing ongoing treatment plans for a population with heterogeneity in disease progression and response to medical interventions. We create a model that learns the patient type by monitoring the patient health over time and updates a patient's treatment plan according to the gathered information. We formulate the problem as a multivariate state-space, partially observable Markov decision process (POMDP) and provide structural properties of the value function, as well as the optimal policy. We extend this modeling framework to a general class of treatment initiation problems where there is a stochastic lead-time before a treatment becomes available or effective. As a case study, we develop a data-driven, decision-analytic model to study the optimal timing of vascular access surgery for patients with progressive chronic kidney disease, and we establish policies that consider a patient's rate of disease progression in addition to the kidney health state. To circumvent the curse of dimensionality of the POMDP, we develop several approximate policies, as well as simpler heuristics, and evaluate them against a high-quality lower-bound. Through a numerical study and several sensitivity analyses, we establish the high quality and robustness of an approximate policy that we develop. We provide further policy insights that sharpen existing guidelines for the case-study problem.

 

Date: Friday, September 13th, 2019
Speaker: Javad Nasiry, McGill University
Title: "TBA"
Time: 9:30 AM - 10:30 AM
Place: HA 967

Abstract: TBA

 

Date: Friday, September 20th, 2019
Speaker: Gah-Yi Ban, London Business School, UK
Title: "TBA"
Time: 9:30 AM - 10:30 AM
Place: HA 133

Abstract: TBA

 

Date: Friday, September 27th, 2019
Speaker: Mona Imanpoor Yourdshahy, UBC Sauder School of Business
Title: "TBA"
Time: 9:30 AM - 10:30 AM
Place: HA 967

Abstract: TBA

 

Date: Monday, September 30th, 2019
Speaker: Ruxian Wang, Johns Hopkins Carey Business School
Title: "TBA"
Time: 9:30 AM - 10:30 AM
Place: HA 966

Abstract: TBA

 

Date: Friday, October 4th, 2019
Speaker: Moshen Bayati, Stanford University
Title: "TBA"
Time: 9:30 AM - 10:30 AM
Place: HA 967

Abstract: TBA

 

Date: Monday, October 7th, 2019
Speaker: Tamar Cohen, MIT
Title: "TBA"
Time: 12:00 PM - 1:00 PM
Place: HA 967

Abstract: TBA

 

Date: Friday, October 11th, 2019
Speaker: Sergei Savin, Wharton School of Business
Title: "TBA"
Time: 9:30 AM - 10:30 AM
Place: HA 967

Abstract: TBA

 

Date: Friday, October 18th, 2019
Speaker: Ed Kaplan, Yale School of Management
Title: "TBA"
Time: 9:30 AM - 10:30 AM
Place: HA 967

Abstract: TBA

 

Date: Friday, October 25th, 2019
Speaker: Yanchong (Karen) Zheng, MIT
Title: "TBA"
Time: 9:30 AM - 10:30 AM
Place: HA 967

Abstract: TBA

 

Date: Friday, November 1st, 2019
Speaker: David Brown, Duke University
Title: "TBA"
Time: 9:30 AM - 10:30 AM
Place: HA 967

Abstract: TBA

 

Date: Friday, November 15th, 2019
Speaker: Alex Belloni, Duke University
Title: "TBA"
Time: 9:30 AM - 10:30 AM
Place: HA 967

Abstract: TBA

 

Past seminar – July 2019

Date: Thursday, July 11th, 2019
Speaker: Mark Lewis, Cornell University
Title: "A Constrained Optimization Approach to Scheduling Patient Flow in Health Care"
Time: 12:30 PM - 1:30 PM
Place: HA 967

Abstract: Congestion in emergency departments continues to be an issue. We model patient flow for low acuity patients using a constrained Markov decision process formulation. In doing so, we consider the problem of scheduling a single-server when there are multiple parallel stations. A classic result in scheduling says to create a station dependent index consisting of the product of the holding cost (per customer, per unit time) times the rate at which the service can be completed at that station. The scheduler then prioritizes work in the order of the indices from highest to lowest. Preferences are captured by the various holding costs. A more natural method for modeling preferences is to assign constraints to the highest priority customers (guaranteeing a fixed quality of service level) and to provide best effort service for the other classes. We consider this formulation, present conditions for optimality and show how to construct an optimal control. We initially focus on the two station model, then explain where the results can be extended.

Image of woman with a headset talking