國立虎尾科技大學 |

Stochastic and Empirical Models in Support of Managing Patient Flow from Acute to Rehabilitation Care.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Stochastic and Empirical Models in Support of Managing Patient Flow from Acute to Rehabilitation Care./
作者:	Gorgulu, Berk.
面頁冊數:	1 online resource (172 pages)
附註:	Source: Dissertations Abstracts International, Volume: 85-05, Section: B.
Contained By:	Dissertations Abstracts International85-05B.
標題:	Industrial engineering. -
電子資源:	click for full text (PQDT)
ISBN:	9798380834650

Stochastic and Empirical Models in Support of Managing Patient Flow from Acute to Rehabilitation Care.
Gorgulu, Berk.

Stochastic and Empirical Models in Support of Managing Patient Flow from Acute to Rehabilitation Care. - 1 online resource (172 pages)

Source: Dissertations Abstracts International, Volume: 85-05, Section: B.

Thesis (Ph.D.)--University of Toronto (Canada), 2023.

Includes bibliographical references

The continuing growth of the older population has led to a significant increase in demand for rehabilitation over the past two decades and subsequently in long delays in admission to rehabilitation care. These delays have two important implications: Firstly, delayed patients continue occupying their beds in acute care, blocking new patients from being admitted and incurring additional costs of operating acute beds. Secondly, admission delays could adversely impact rehabilitation outcomes. In this thesis, we propose new empirical and stochastic models of patient flow and use them to develop insights and operational interventions aimed at reducing admission delays. In Chapter 2, we investigate the sources of admission delays in Low Tolerance Long Duration (LTLD) rehabilitation. Admission delays can be caused by capacity constraints or extra processing requirements to plan rehabilitation activities. However, standard data only includes a single (combined) measure of delay, and the bed allocation decisions in practice can be quite complicated with multiple determinant factors. To address these challenges, we propose a Hidden Markov Model to estimate the status-quo bed allocation policy with unobserved processing times. The estimation results are then combined with a high-fidelity queueing model of the system to evaluate the effectiveness of various interventions to reduce admission delays. In Chapter 3, we focus on High Tolerance Short Duration (HTSD) rehabilitation and investigate the causal impact of delays on rehabilitation length-of-stays (LOSs) and outcomes. By utilizing an instrumental variable approach, we find that admission delays increase the rehabilitation LOS and negatively impact the functional outcomes of the patients in rehabilitation. Finally, motivated by the empirical findings of Chapter 3, in Chapter 4, we develop approximately optimal scheduling policies for multiclass queues with wait-dependent service times. We propose a simulation-based approximate policy iteration algorithm, wherein the policy improvement step is guided through Machine Learning-based classifiers and the policy evaluation is accelerated using a coupling approach. Using extensive numerical experiments, we investigate the scalability of the proposed algorithm and further apply it in a Case Study calibrated using HTSD rehabilitation data.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798380834650Subjects--Topical Terms:

679492
Industrial engineering.
Subjects--Index Terms:

Approximate dynamic programmingIndex Terms--Genre/Form:

554714
Electronic books.

Stochastic and Empirical Models in Support of Managing Patient Flow from Acute to Rehabilitation Care.
LDR:03831ntm a22004097 4500 001 1143648
005 20240517104621.5
006 m o d
007 cr mn ---uuuuu
008 250605s2023 xx obm 000 0 eng d
020 $a 9798380834650
035 $a (MiAaPQ)AAI30637031
035 $a AAI30637031
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Gorgulu, Berk. $3 1468396
245 1 0 $a Stochastic and Empirical Models in Support of Managing Patient Flow from Acute to Rehabilitation Care.
264 0 $c 2023
300 $a 1 online resource (172 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 85-05, Section: B.
500 $a Advisor: Sarhangian, Vahid.
502 $a Thesis (Ph.D.)--University of Toronto (Canada), 2023.
504 $a Includes bibliographical references
520 $a The continuing growth of the older population has led to a significant increase in demand for rehabilitation over the past two decades and subsequently in long delays in admission to rehabilitation care. These delays have two important implications: Firstly, delayed patients continue occupying their beds in acute care, blocking new patients from being admitted and incurring additional costs of operating acute beds. Secondly, admission delays could adversely impact rehabilitation outcomes. In this thesis, we propose new empirical and stochastic models of patient flow and use them to develop insights and operational interventions aimed at reducing admission delays. In Chapter 2, we investigate the sources of admission delays in Low Tolerance Long Duration (LTLD) rehabilitation. Admission delays can be caused by capacity constraints or extra processing requirements to plan rehabilitation activities. However, standard data only includes a single (combined) measure of delay, and the bed allocation decisions in practice can be quite complicated with multiple determinant factors. To address these challenges, we propose a Hidden Markov Model to estimate the status-quo bed allocation policy with unobserved processing times. The estimation results are then combined with a high-fidelity queueing model of the system to evaluate the effectiveness of various interventions to reduce admission delays. In Chapter 3, we focus on High Tolerance Short Duration (HTSD) rehabilitation and investigate the causal impact of delays on rehabilitation length-of-stays (LOSs) and outcomes. By utilizing an instrumental variable approach, we find that admission delays increase the rehabilitation LOS and negatively impact the functional outcomes of the patients in rehabilitation. Finally, motivated by the empirical findings of Chapter 3, in Chapter 4, we develop approximately optimal scheduling policies for multiclass queues with wait-dependent service times. We propose a simulation-based approximate policy iteration algorithm, wherein the policy improvement step is guided through Machine Learning-based classifiers and the policy evaluation is accelerated using a coupling approach. Using extensive numerical experiments, we investigate the scalability of the proposed algorithm and further apply it in a Case Study calibrated using HTSD rehabilitation data.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Industrial engineering. $3 679492
650 4 $a Mechanical engineering. $3 557493
653 $a Approximate dynamic programming
653 $a Causal inference
653 $a Healthcare operations
653 $a Hidden Markov models
653 $a Patient flow
653 $a Structural estimation
655 7 $a Electronic books. $2 local $3 554714
690 $a 0546
690 $a 0796
690 $a 0548
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a University of Toronto (Canada). $b Mechanical and Industrial Engineering. $3 845502
773 0 $t Dissertations Abstracts International $g 85-05B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30637031 $z click for full text (PQDT)