Title:

Hospital-wide Inpatient Flow: Optimization vs. Recommendation
Abstract:

 In an attempt to coordinate and optimize hospital operations across all services in real-time, we develop a multistage adaptive robust optimization model, informed by data and ML predictions, that unifies the entire bed assignment process while accounting for present and future inpatient flows, discharges, and bed requests. On simulations calibrated for a 600-bed institution, our optimization model was solved in seconds, reduced off-service placement by 24% on average, and boarding delays by 31%-46%.
If deployed in the hospital, however, the benefit will likely be much lower. Among others, the fact that nurses can override the recommendation made by our algorithm can negatively impact performance. In the second half of the talk, we will theoretically study the extent to which this partial adherence phenomenon (a) impacts performance, and (b) should influence the design of the algorithmic recommendation in the first place. Indeed, the best decisions are not necessarily the best advice.
Bio:

Jean is an Assistant Professor of Management Science and Operations at London Business School. His research focuses on large-scale discrete optimization, robust optimization, and machine learning, with applications to healthcare operations. His work has been published in the likes of Operations Research, Mathematical Programming, and M&SOM, and recognized by many awards, including the INFORMS Pierskalla, George E. Nicholson, and Computing Society best student paper awards. Jean received a Ph.D. in Operations Research from MIT and a Diplôme d'ingénieur from Ecole Polytechnique (Paris).