Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.742241
Title: Modelling and solving healthcare decision making problems under uncertainty
Author: Gökalp, Elvan
ISNI:       0000 0004 7227 7586
Awarding Body: University of Warwick
Current Institution: University of Warwick
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
The efficient management of healthcare services is a great challenge for healthcare managers because of ageing populations, rising healthcare costs, and complex operation and service delivery systems. The challenge is intensified due to the fact that healthcare systems involve various uncertainties. Operations Research (OR) can be used to model and solve several healthcare decision making problems at strategic, tactical and also operational levels. Among different stages of healthcare decision making, resoure allocation and capacity planning play an important role for the overall performance of the complex systems. This thesis aims to develop modelling and solution tools to support healthcare decision making process within dynamic and stochastic systems. In particular, we are concerned with stochastic optimization problems, namely i) capacity planning in a stem-cell donation network, ii) resource allocation in a healthcare outsourcing network and iii) real-time surgery planning. The patient waiting times and operational costs are considered as the main performance indicators in these healthcare settings. The uncertainties arising in patient arrivals and service durations are integrated into the decision making as the most significant factors affecting the overall performance of the underlying healthcare systems. We use stochastic programming, a collection of OR tools for decision-making under uncertainty, to obtain robust solutions against these uncertainties. Due to complexities of the underlying stochastic optimization models such as large real-life problem instances and non-convexity, these models cannot be solved efficiently by exact methods within reasonable computation time. Thus, we employ approximate solution approaches to obtain feasible decisions close to the optimum. The computational experiments are designed to illustrate the performance of the proposed approximate methods. Moreover, we analyze the numerical results to provide some managerial insights to aid the decision-making processes. The numerical results show the benefits of integrating the uncertainty into decision making process and the impact of various factors in the overall performance of the healthcare systems.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.742241  DOI: Not available
Keywords: R Medicine (General) ; TA Engineering (General). Civil engineering (General)
Share: