Design and Development of a Process Mining System for the Optimization of a Loan Approval Process

Eduardo Varas. (2017). Design and Development of a Process Mining System for the Optimization of a Loan Approval Process. Trabajo Fin de Titulación (TFG). ETSI Telecomunicación, Universidad Politécnica de Madrid.

Nowadays information systems play a crucial role in supporting business processes. Thus, it is possible to record and analyze business events with the aim of understanding, discovering and optimizing processes from their logs. The challenge now is to exploit those events data in a meaningful way. The process of extracting meaningful processes from event logs is known as Process Mining. This final degree project applies different techniques in order to analyze, understand and optimize a loan approval process from its recorded event data. This data is provided by a Dutch Bank. Firstly, from the logs provided, and using some Process Mining tools, the process will be modelled, and analyzed following some Process Mining guidelines. This mentioned Process Mining discipline is a group of techniques that allows business process analysis from event logs. Nowadays almost all the modern BPM record every event in a log form, the process mining techniques goal is to use the logs in order to extract the needed information e.g., automatically discover a process model. This project objective is to correctly apply the Process Mining techniques in order to deeply understand how the process works, find inefficiencies, bottlenecks and any other issues that appear during the analysis. And finally if possible reorganize it trying to drive efficiency to the process. To achieve this objective different guidelines are going to be followed, in the first place, some of the paths and methodologies described in the Process Mining: Discovery, Conformance and Enhancement of Business Processes by Wil M. P. van der Aalst are going to be applied. BPI challenge 2012 will also be taken into account as a guideline since this 2012 challenge used the same process as a study subject, turning it into the actual process. The main tools that will be used are Disco, a tool developed by Fluxicon, which already contains many Process Mining utilities simplifying the analysis. Celonis, another Process Mining tool, will be used too thanks to its additional features not included in disco. This project will start with the modelling of the process, trying to obtain a model as real as possible. Then the process will be analyzed in order to find the inefficiencies sources and the bottlenecks, and finally where it is possible to reorganize the process looking for the process optimization.