Author:

Abstract:

One of the main objectives of this doctoral dissertation is to provide insights on how to improve operating room (OR) planning. These insights are derived from real data with the help of data science and simulation modeling techniques. In our models we aim to increase the number of patients that are served within the medically advised time interval, while taking into account the performance measures that are important for the other stakeholders (OR staff and hospital management). In Chapter 2, we discuss the existing models found in the literature on the OR planning problem. In order to help researchers and practitioners to select new relevant articles, we classify the recent OR planning and scheduling literature into tables regarding patient type, used performance measures, decisions made, OR up- and downstream facilities, uncertainty, research methodology and testing phase. Based on these classifications, we identify trends and promising topics. In Chapter 3, we turn our attention to a specific group of patients: the non-elective patients. As these patients have to be served on the day of their arrival, they cannot be planned according to the algorithms or policies that are used for elective patients. In this chapter, we review the different possibilities to incorporate non-elective patients in the schedule with the goal of maximizing both patient-, staff- and hospital-related performance measures. We discuss and compare three main policies: the dedicated, the flexible and the hybrid policy. In Chapter 4, we tackle the need for policies that guide the OR manager in the trade-offs that arise when OR time is shared between elective and non-elective patients. We examine whether and in which cases it is reasonable to install capacity buffers and we investigate which type of buffer (i.e., a dedicated OR or a buffer per OR) results in the best performance. We also show the impact of different access policies for non-elective patients and discuss the applicability of both existing and newly developed policies. Based on both an extensive analysis of the inpatient surgical department of the UZ Leuven and the collaboration with the hospital management and staff, we identify the aspects of the real hospital setting that need to be included in the simulation model (e.g., time-varying non-elective arrivals, different break-in options). Although the main attention in both practice and literature is on the question whether or not to dedicate buffer capacity to non-elective patients in general, we discuss in this chapter why it is not advisable to deal with this problem in isolation of other hospital policies. In Chapter 5, we focus on another patient group: the pediatric patients. The need for specialized anesthesia equipment for these patients and the preference for audiovisual separation of pediatric and adult patients can be addressed by planning the pediatric surgeries in one or more fully equipped pediatric ORs. Although this separation offers benefits concerning equipment use and the planning of anesthesiologists, its impact on patient scheduling has barely been studied. In this chapter, we assess the feasibility of allocating OR sessions to pediatric patients. We therefore investigate the resulting change in the access times for both adult and pediatric patients as well as the impact the separation would have on other operational performance measures. Finally, while in the previous chapters we took the different sources of uncertainty on the day of surgery as given, in Chapter 6 we look at several aspects that can play a key role in reducing this uncertainty. First, we present an exploratory analysis on how the organization of the overtime capacity affects the realized overtime hours, the elective patient's cancellation rates and non-elective access times. These insights can support the hospital in its search for personnel shift types and overtime rules that aim to increase the satisfaction of the OR staff. Secondly, we discuss the insights derived from an analysis of the real surgery durations. These insights can contribute to future research on the estimation of the surgery duration, which is a popular topic in the literature and an ongoing challenge in practice.