運用萬用啟發式演算法求解病人轉檢問題

Applying Meta-heuristics to Solve Patient Referral Problems

王宣尹(Xuan-Yin Wang)

指導教授 ： 陳平舜

中原大學/電機資訊學院/工業與系統工程研究所/碩士(2015年)

https://doi.org/10.6840/cycu201500333

摘要

運用萬用啟發式演算法求解病人轉檢問題
研究生：王宣尹 指導教授：陳平舜
中原大學工業與系統工程學系 (所)
摘 要
近年來民眾對於醫療資源的需求逐漸增加，而醫療資源的利用率將影響醫院的醫療服務品質；因此，如何將醫療資源進行分配便成為一個重要的問題。民眾既定的印象認為大醫院能提供較好的醫療服務，此認知造成大醫院湧入大量病患。而這樣的情形將使醫院不堪負荷並增加病人的等待時間，進而降低醫療服務品質。
本研究所探討的個案醫院由於兩院區的核磁共振影像檢查之病人等待時間相差甚鉅，為因應此問題，個案醫院對其兩院區進行病人轉檢。而轉檢的病人數量多寡將會對於病人的等待時間有所影響，因此本研究之目的為求解個案醫院之病人轉檢數量及評估個案醫院目前轉檢人數是否適當。
本研究使用系統模擬及萬用啟發式演算法，建立符合個案醫院之模擬模型，並求解出個案醫院適當的轉檢人數。本研究採用之蝙蝠演算法，藉由求解多組情境，以證明其演算法之穩定性；而從求解所得之轉檢人數及模擬結果可發現，不僅使個案醫院的病人等待時間大幅下降，也可提供個案醫院作為訂定轉檢人數標準之參考。

關鍵字： 模擬最佳化、病人轉檢、蝙蝠演算法、萬用啟發式演算法

關鍵字

蝙蝠演算法 ； 模擬最佳化 ； 萬用啟發式演算法 ； 病人轉檢

並列摘要

Applying Meta-heuristics to Solve Patient Referral Problems
Student: Xuan-Yin Wang Advisor: Dr. Ping-Shun Chen
ABSTRACT
In recent years, the demands of people for medical service increase significantly. The utilization of medical resources has an impact on the service quality of hospitals. Therefore, how to allocate medical resources becomes an important issue. In general, people believe that large hospitals are capable to provide the better service, and people are willing to go to large hospitals instead of small ones. This phenomenon will lead to not only overcrowd in larger hospitals, but also lower the service quality in larger hospitals.
The case hospital has a large gap of the number of Magnetic Resonance Imaging (MRI) patients between two districts. To deal with this problem, the case hospital began to transfer its patients from one district to another, but the question is how many patients should be transferred in order to shorten the average patient waiting time. The purpose of this thesis is therefore to determine the appropriate number of referring patients for the case hospital and compare to its current referral mechanism.
This research applied system simulation and combined with meta-heuristics to obtain the appropriate number of referring patients. After solving twenty instances, the results showed that the robustness of the bat algorithm which was adopted by this study. Furthermore, the outcomes showed that there was significant improvement on the average patient waiting time of the case hospital. The findings of this research can provide as a reference for other hospitals to develop their referral standard procedures.
Keywords: Simulation optimization, patient referral, bat algorithm (BA), meta-heuristics

並列關鍵字

patient referral ； Simulation optimization ； meta-heuristics ； bat algorithm (BA)

參考文獻

張馨予，郭美宏，劉佩芬，康春梅，(2012)，改善心臟科門診病人跨院際轉診檢查之時效，長庚護理，23(4)，488-501。

楊淳懿，(2011)，運用系統模擬探討醫院協同環境之轉院機制，中原大學工業與系統管理學系，碩士論文。

楊智堯，(2013)，醫學中心急診室單元工程佈置設計與分析，國立成功大學製造資訊與系統研究所，碩士論文。

Arora, S., & Singh, S. (2013). A conceptual comparison of firefly algorithm, bat algorithm and cuckoo search. Control Computing Communication & Materials (ICCCCM), 2013 International Conference.

Aungkulanon, P. (2014). A comparative study of global-best harmony search and bat algorithms on optimization problems. Applied Mechanics and Materials, 464, 352-357.