Three essays on optimization of the intensive care unit (ICU) management decisions

Bai, Jie

PDF
Bai_Dissertation.pdf - Published
Download (5MB)

URL:	https://madoc.bib.uni-mannheim.de/61639
URN:	urn:nbn:de:bsz:180-madoc-616394
Document Type:	Doctoral dissertation
Year of publication:	2022
Place of publication:	Mannheim
University:	Universität Mannheim
Evaluator:	Stolletz, Raik
Date of oral examination:	11 February 2022
Publication language:	English
Institution:	Business School > ABWL u. Produktion (Stolletz 2010-)
Subject:	650 Management
Keywords (English):	intensive care unit (ICU) , admission and discharge decisions , decision making , Markov decision process , dynamic programming , simulation , operations research
Abstract:	The intensive care unit (ICU) is one of the most crucial resources in the hospital. Improper ICU management causes many negative effects in the ICU itself and in other connected departments along the patient care path. This dissertation presents three papers on optimizing the ICU management decisions. The first paper provides the first structured and comprehensive review of ICU problems in OR/MS. The relevant papers are discussed based on a new framework. The second paper proposes a discrete-time MDP model to find admission and early discharge policies that minimize these negative consequences. By minimizing the medical consequences, the approach demonstrated significantly outperforms a myopic policy as applied by most hospitals in practice. The third paper compares eleven different management policies based on different KPIs by a simulation study. In comparison to the baseline case running on a FCFS rule, any management policy is superior regardless of the evaluation criteria.