Operations Improvement on Integrated Health and Social Care System

綜合醫療及社會服務系統的運營改進

Student thesis: Doctoral Thesis

View graph of relations

Author(s)

  • Meiyan LIN

Detail(s)

Awarding Institution
Supervisors/Advisors
Award date22 Dec 2016

Abstract

The Integrated Health and Social Care (IHSC) system plays a crucial role in affecting the health of the population and its quality of life. However, the media and the internet almost daily shed a bad light on IHSC services because of bad health outcomes, such as medical errors, poor quality, etc., poor patient experience, such as, long waiting time, insufficient access, etc., and high costs. One essential reason for these complaints is weak operations management in the IHSC system. Therefore, in order to achieve the ‘‘Triple Aim’’, namely, improving health outcomes, enhancing patient experience, and reducing costs, efforts should be made to improve different aspects of the IHSC system such as the interplay of its different parts (i.e., functions, structures, and professions) and the role of operations management in the system. The latter requires each structure/care provider to pay attention to operational issues that hinder the achievement of the ‘‘Triple Aim’’.

This thesis has a macro aim of drawing more attention to improvement on planning efficiency and satisfaction of care participants in the IHSC system from different planning horizons and different IHSC structures. This thesis proposes a three-step approach to these different cases: (1) to identify the complex operational problems that several IHSC structures face in this system; (2) to develop models and algorithms for these problems hence leading to improvement on planning efficiency; and (3) to offer observations for further improvement on satisfaction of different care participants by analyzing the experimental results.

The first chapter of this thesis presents a qualitative analysis of operations management in the IHSC system. More specifically, it identifies the complexities within the system which are related to nature of services, criteria of satisfactory quality, and costs. Thereafter, operational issues over different planning horizons within the IHSC system are investigated and surveyed. This investigation aims at identifying the complexity of IHSC operations management and unveiling unexplored opportunities for researchers.

The second chapter of this thesis mainly discusses the operational issue facing a daily EDCC, that is, its timetabling problem. Firstly, a mathematical formulation based on the maximum constraints satisfaction problem (Max-CSP) framework is presented for the problem which considers such factors as clashing, availability, and capacity constraints. To solve the Max-CSP-based model, an effective genetic-based discrete particle swarm optimization (GDPSO) algorithm is proposed using the combination of min-conflict strategy, random walk, genetic operators, and the one-way information-sharing mechanism of PSO. By validating GDPSO with six real-life instances, analysis of the experimental result reveals that (1) the proposed GDPSO excels, in terms of stability and efficiency, in solving the Max-CSP-based EDCC timetabling problem; that (2) operational improvement for the EDCC can be achieved by such strategies as extending its room opening hours or purchasing more equipment for activities at acceptable costs. In other words, timetabling efficiency can be achieved by effective Max-CSP model and GDPSO and satisfaction of EDCC could be improved by proper strategies based on analyzed results.

The third chapter of this thesis investigates and tackles the operational issue that occurs in the rehab department of an IHSC structure, that is, the weekly therapist assignment problem (TAP). To do that, a generic mix-integer programming (MIP) model is formulated for the TAP. Thereafter, based on the model, five weight-allocation criteria are discussed as to whether they match the current operational process of service providers. A numerical analysis of the computational results obtained by the Gurobi Optimizer is then carried out for the TAP. Especially, the effect of characteristics, such as patients’ preferred time periods (PTPs), continuity of care (COC), and patient priority (PP) on performance, is discussed to offer the department ideas for improvement. The analysis reveals that (1) characteristics, including COC, PP, and PTPs, turn out to have important effects on the rate of patient satisfaction; (2) the proposed MIP model for the TAP is valid and can be extended to integrate variations; (3) the Gurobi Optimizer is an effective tool for instances of acceptable sizes in a few seconds. That is, MIP model helps to improve the assignment efficiency and observations on key characteristics provide directions for further improvement on satisfaction of patient and HHC structure.

The fourth chapter of this thesis analyzes the operational issue that occurs when providing Meals-On-Wheels (MOW) service to patients, that is, the MOW service districting (MOWSD) problem at strategic level. To solve the problem, an integrated MIP model is proposed to formulate it by considering criteria such as capacity and delivery duration limitation, accessibility, compactness, and the individuality of customers. A greedy heuristic method is also developed to quickly solve the MOWSD problem. Finally, several computational experiments are conducted to evaluate the performance of the heuristic method and the Gurobi Optimizer and to analyze the sensitivity of many key parameters of the MOWSD problem. The results demonstrate that (1) the greedy heuristic method can quickly find good feasible solutions; (2) improvement can be achieved by the proposed MIP model and the greedy heuristic method on the basis of comparison between the resulting and existing districts; (3) key parameters of the MOWSD problem, such as the capacity of meal carts, the total time for MOW service delivery, and the maximum travel time between any two customers in a district, have significant effect on the end results. In a word, the effective MIP model and greedy heuristic method achieve improvement on both districting efficiency and human resource utilization; while observation on key parameters offers proper decisions for making sure satisfaction of patient and care worker.

The contribution of this thesis lies in three areas. First, this is a timely research, focusing on improvement in current IHSC system from different planning horizons and structures. Second, effective models and algorithms are proposed for efficiency improvement. Third, characteristics of these problems are analyzed with proposed methods to provide tips for satisfaction improvement.