Sağlık Süreçlerini İyileştirmede Süreç Madenciliği Tekniğini Kullanan Bir Performans Analiz Metodu
Özet
A PERFORMANCE ANALYSIS METHOD FOR HEALTHCARE PROCESS IMPROVEMENT USING PROCESS MINING TECHNIQUE
In the last decade, as an emerging technique for business processes management, process mining has been applied in many domains including manufacturing, supply-chain, government, healthcare, and software engineering. The application of this technique is growing yet challenging, particularly in healthcare, where most processes are complex, variable, dynamic, and multi-disciplinary in nature. In addition to the body of evidence on the applicability and usefulness of process mining techniques in the healthcare domain, there are also significant challenges to be addressed. The basic challenges observed during literature review are related to healthcare data, process mining techniques, performance, and data visualization.
In the field of healthcare processes, a systematic mapping study has been conducted in this area in order to be able to create a solution proposal to better understand the nature of process mining applications and cope with the challenges. Systematic mapping is a well-accepted method to identify and categorize research literature, in which the number of primary studies is rapidly growing. In the electronic digital libraries of scientific literature, the studies between 2005 and 2017 were searched, and 172 primary studies out of the 2428 initially found on the topic of process mining in healthcare were identified. A concept map of the information provided by primary studies on process mining applications in healthcare was systematically developed and then this map was used to classify the studies based on a number of attributes; namely, the types of research and contribution, application context, mining activity, process modeling type and notation, and mining algorithm. To the best of our knowledge, this is the only systematic mapping study that provides a broad and general overview of the studies on this topic. The results of mapping showed that, despite the challenges related to healthcare data and technique, the field is rapidly growing and open for further research and practice. The researchers who are interested in the field could use the results to elicit opportunities for further research. The practitioners who are considering applications of process mining, on the other hand, could observe the most common aims and specialties that process mining techniques are applied.
The systematic mapping study also showed that process mining community in the healthcare domain is lacking a comprehensive methodology that describes how to accomplish process mining activities and techniques in practice. Based on the need stated above, this thesis proposes a novel, distributed process mining based process performance evaluation method for monitoring, controlling and improving healthcare processes. This goal-oriented method aims to: (1) provide a broad overview of processes in a short time, (2) create improvement opportunities for processes by incorporating domain knowledge into process enactments, (3) analyze unstructured, multidisciplinary processes like healthcare processes, (4) implement process mining activities and techniques in process mining methodology, and (5) provide a practical guidance for traceability, usability and functionality of process mining implementations using a goal-driven approach.
The proposed method was applied in case studies on the surgery process and emergency service process of a university hospital in Turkey, which revealed for quantitative insights into the process. Bottlenecks and deviations that were crucial for determining measures (e.g., data and performance information) were identified to improve the efficiency of these processes. The results of case study showed that the proposed method is effective for initiating process improvements by guiding the use of process mining techniques in the healthcare domain.
Keywords: Software Engineering, Process Improvement, Process Modelling, Process Mining, Systematic Mapping, Goal-Question-Metric Framework
