- Presentation by Michel Koeken (thesis student supervised by Pieter Van Gorp. Michel should graduate within one Month).
- Abstract and Focus Question for the session:
First I would like to introduce my master thesis. I am performing an analysis on the European tenders for the procurement of a HIS/EHR system in the Dutch academic hospitals. Such a European tender consists of 2000 questions which are made up by the academic hospital. Each supplier has to fill in all questions and for each question a supplier can gain points if they answer it correctly. In the end the academic hospital adds all gained points per supplier and the winner is the supplier with the highest score.
Each tender is divided into several subtopics and each subtopic consists of several questions. As you can gain points per questions and it is known how many questions belong to one specific subtopic it’s also known how many points can be gained per subtopic.
I have this information of three tenders and the aim is to identify the most important subtopics in order of points and how these most important subtopics changed over time. Unfortunately each tender used slightly different subtopics. To analyze how the most important subtopics changed over time we had to take one tender as reference and match the subtopics of the other two tenders with the subtopics of the reference tender. I used the Delphi method (in the first round the two experts matched the subtopics individually, and in the second round the experts and I sat together to find the correct matching) to match every subtopic with a subtopic of the reference tender.
The example in the presentation is a simplified example of three tenders, but with only 5, 4 and 3 subtopics per tender. The matching that is made is fictional and it’s only purpose is as an example. In this example Tender A is the reference tender.
The question that we have is how we should present the results? Which trend analysis can be used and how should this be performed?
- Draft slides: IS-HEART MK
Author: pvgorp
Master Thesis defense by Robert Janssen
Title
Increasing accessibility and reproducibility of process mining research in healthcare
Abstract
The goal of this thesis was to develop a set of best practices for process mining research in healthcare. Process mining in healthcare has been studied by several researchers already, but these studies suffer from disappointing results, which may be partly to blame to the low accessibility and reproducibility of the process mining research methodologies. Therefore, in this thesis we aimed to synthesize the most common and best research steps, or best practices, of previous process mining research by means of a multiple case study and expert interviews into one process mining methodology for healthcare which would show increased accessibility, reproducibility and results that are better, or at least comparable, to present process mining research. Pattern language was identified as the most suitable method to report the best practices and as a result, 22 best practices patterns that cover a variety of process mining research steps have been developed. Subsequently, 11 patterns were selected as recommendations for use in healthcare, as they showed the most promising results and highest accessibility. In addition, we have introduced the concepts of virtual machines and screencasts to the field of process mining as means to increase accessibility and reproducibility of the best practices pattern language. The results of the best practices and recommendations have been validated by experts at two Dutch healthcare institutions and showed that the process mining best practices pattern language is indeed able to produce process models which are of similar or superior quality compared to process models created
with conventional methodologies.
Online Access
Get the PDF from the TU/e library
Master Thesis defense Guyon (J.M.) Muijres
Title
Comparing BPM approaches in the healthcare Domain: Case Handling vs. Model Driven Engineering (BPMone vs. Mendix)
Abstract
In a world of continuously changing business environments, organizations are forced to deal with, and adjust to the demands for such level of change. To handle this level of change, many companies use different approaches and tools which include modeling and measuring business processes. Business Process Modeling provides support on business operations to improve business efficiency and to find new opportunities for the company. Not all organizations have recognized the added value of this approach. Especially healthcare organizations do not fully use the benefits of IT support. This thesis presents the pros and cons of two promising BPM approaches in the healthcare domain, with the Eye Care Network in Rotterdam as case study. The Eye Care Network is a collaborating healthcare business network of different players within the ophthalmology domain. There is often a network manager within a healthcare business network that wants to standardize the way the different players work. This can improve efficiency within the network and it can improve communication between the different players. The Eye Care Network, as a network manager, designed different treatment plans, as desired processes for treating patients in the hospitals connected to the network. One important aspect of these treatment plans is dealing with exceptions. Flexibility is required with this, to ensure the network members can continue providing their services, even when they want or need to deviate from the standardized process. This thesis presents the further investigated business process that describes the different steps a patient will endure during the treatment plan for the eye disease glaucoma. This thesis discusses how well two promising BPM approaches meet with the requirements from the Eye Care Network, based on the construction and evaluation of two prototypes. To be able to objectively compare the two approaches, the comparison framework of the thesis consists of seven principles divided over four different levels; case study level, design level, language level, and tool level. Model Driven Engineering (MDE) and Case Handling are the two approaches used to deal with the required flexibility by design. BPMone from Pallas Athena represents a specialized case handling tool, and Mendix represents a MDE tool. After creating a first version of both applications, BPMone could be seen as the best option out of the two tools. Mendix’s primary weakness is the lower level of abstraction, compared to BPMone. This results in the need of a higher specification effort, which leads to the risk of making bad design decisions. Decisions that are already taken implicitly in BPMone. However, this limitation can be overcome by changing the Mendix model with a number of novel patterns, based on the apparently stronger BPMone principles. Similar options are not seen for BPMone (nor is there a need for it). An improved version of the Mendix model resulted in a comparing situation where BPMone was no longer the clear winner. This thesis not only gives a description of the abstraction patterns it also shows how they could be used on other cases as well. The suitability is not limited to this specific healthcare setting, since the generalizability of the thesis conclusions are demonstrated by considering a case study from a completely different domain as well.
The goal of this thesis is to examine which BPM approach is best suiting the requirements given by the Eye Care Network, for their business process. Furthermore, the possibility to generalize these findings outside of this case study will be checked.
Poster
See attached: Poster Master Thesis_G.J.M.Muijres
Master Thesis Defense Erwin van den Broek
Thesis Title
Transforming Unstructured Process Models: Translating Implicit and Scattered Parallel Structures into Structured Hierarchies
Abstract
The usage of process models in healthcare is low. Even though this industry could benefit greatly from an increased usage of process models, since these models help with the understanding of processes taking place. We therefore set out to find a way for increasing the understandability of these models. The goal is to transform existing models into better understandable models. The focus of this research is not on the construction of a good model, but on improving an existing model without changing its behavior. As suggested by literature, hierarchy is used to achieve this improvement in understandability.
For the research PN2HSC was used, a transformation implementation in the GrGen software suite. This transformation tool transforms petri nets to state charts with hierarchy. Though PN2HSC can transform structured process models without a problem, it is not able to transform unstructured models correctly.
Unfortunately, models from healthcare and other industries are often unstructured, thereby severely limiting the usage of PN2HSC. This research aimed to improve PN2HSC by extending it to handle unstructured models. Four patterns of unstructured behavior where found in literature, which served as the basis for developing rules for finding violations of structure in process models.
Edges that cause this unstructuredness are temporary hidden from the model, so a structured model remains, that can be transformed using PN2HSC (or another transformation tool). Afterwards the hidden edges are made visible again.
Applying these rules to models showed that they do not take multi‐level hierarchies into account: parallelism as part of another parallel path is incorrectly marked as a violation. A filter is created to prevent this incorrect marking. Additionally, resource structures and structures where there is a crossover between parallel paths, require additional rules to be handled correctly. Thus the four existing patterns were extended with two additional patterns. Subsequently, additional rules are created as well.
The results were validated using real‐life models from the healthcare field, to see how these patterns perform in real situations. It was noticed that the degree of success varied: the rules created can uncover latent hierarchy in models, but they will not find hierarchy where there is originally none. It was noted that finite models, with a start and end node, performed better in contrast to continuing (infinite) models. Also, black token nets outperformed colored nets. There should not be any conclusions connected to these observations, since the number of models tested was too low to substantiate any claims. Though it seems possible that certain types of models contain a higher degree of hierarchy than other model types.
The validation did show that the developed rules are very successful in uncovering latent hierarchy, and some complicated models benefitted greatly from the additional transformation rules. These models became much easier to understand, due to hierarchical presentation of the model.
Online Access
Get the PDF from the TU/e library.
Master Thesis defense by Tom van der Molen
Title
Maintaining consistency between business process diagrams and textual documentation using the EPSILON Model Management Platform
Abstract
This study focuses on the demonstration how a generic model transformation tool can be used to manage inconsistencies between business process diagrams and textual documentation. It is tried to prove that a generic model transformation tool is more suitable for maintaining consistency in comparison to the already existing tools. Therefore, first literature is used to provide a framework for managing inconsistencies. The state-of-the-art in commercialized tools that can contribute in
consistency management is explored and rated against the provided framework. In a final stage a prototype of an inconsistency management tool, build with a generic model transformation tool, is used and expanded to show the improvements in inconsistency management based on the framework and compared to the available commercialized tools.
Online Access
Get the PDF from the TU/e library