Simulating Patient Flows in a Cancer Clinic

Abstract

Planning resources in cancer care is difficult due to system complexity, varying
resource availability and uncertain demand. It has become so complex that the
analysis of treatment processes in cancer clinics cannot be done analytically [12].
Therefore, in this thesis a model of a cancer clinic is developed using discrete
event simulation. The goal of the thesis is develop the model so that it can be
used as research tool to investigate bottlenecks, allocation of resources and system
properties.The model will also serve as a base plate for further investigations.
Most other studies on this subject have not in detail presented how the simulation
handles the discrete effect of weekends and servers only being active during certain
part of the day, which is addressed in thesis. In this thesis a sensitivity analysis
is carried out for a number of different parameters. First we investigate the effect
of increasing the load and analyse how schedule utilization and patient system
time (PST) are affected. Secondly, the system sensitivity to disturbances such as
sick doctors and patient rebooks are investigated. Furthermore, bottlenecks and
schedule alignment is also investigated.

Popular Abstract

During the past 50 years health care and cancer treatment have experienced a
tremendous development. One major side effect of this is that the treatment
process and healthcare clinics in general have become much more complicated.
Therefore, it has become difficult to analyse inefficiencies such as bottlenecks in
the treatment process. It has also become harder to study cause and effect relationships
with conventional tools such as spreadsheet and organization charts.
Furthermore, the complexity level also makes it difficult to study the system with
analytical tools such as optimization or queuing theory. To be able to study the
system a computer simulation model is developed. The model basically tries to
mimic and take in all the aspects, rules and processes of a real world clinic. Of
course this is not possible so therefore there are some limitations set on the model.
The model is built upon on the theory of discrete event simulation. Discrete event
simulation is a convenient simulation tool to use when a discrete process is to
be modeled, in this model we are dealing with patient and time schedules which
are discrete and therefore discrete event simulation is suitable to use for this model.
All patients that come into a clinic go through a workflow, depending on the
cancer type, stage and many other factors. A workflow is a treatment process with
some predefined stations such as a doctor consultation, CT-scan and radiation
treatment [6]. In this thesis we only study one workflow in a radiation therapy
clinic. The main goal of this thesis is to explore how the system behaves when
certain parameters are changed. One of the properties investigated is how the time
a patient spends in the system changes when either more or less patients enter the
clinic. Another factor that is investigated is how the system reacts to disturbances.
A disturbance can be that a doctor becomes sick or that a patient reschedules his
or her appointment. Finally, all these interactions and results are analyzed to see
if the system has some general properties that might apply to other workflows and
clinic setups as well.

• author: Lejdström, Christian
• supervisor: Emma Fitzgerald
• organization: Department of Electrical and Information Technology
• course: EITM01 20201
• year: 2020
• type: H2 - Master's Degree (Two Years)
• subject: Medicine and Health Sciences
• keywords: Discrete Event Simulation Radiotherapy
• report number: LU/LTH-EIT 2020-765
• language: English
• id: 9019983
• date added to LUP: 2020-06-24 16:23:17
• date last changed: 2020-06-24 16:23:17