Aydemir F.Pabuccu Y.U.Basciftci F.2020-03-262020-03-2620191877-0509https://dx.doi.org/10.1016/j.procs.2019.09.048https://hdl.handle.net/20.500.12395/384063rd World Conference on Technology, Innovation and Entrepreneurship, WOCTINE 2019 -- 21 June 2019 through 23 June 2019 -- 141488Business process management is an integral part of many organizations in doing their day-to-day business operations considering that they help organizations transfer their business critical information from one organizational entity to another in the form of automated state transitions. The data (or event logs) that have been generated during these transitions hold valuable insights for improving organizational business processes, highlighting problem areas and visualizing the actual vs the formal procedures. The goal of this paper is to summarize the development of a hybrid analytical approach that utilizes both SQL and NO-SQL based back-end platforms in harmony in order to carry out process mining for a participation bank in Turkey. For this purpose, first we have developed a hybrid analytical software infrastructure that is backed by MS SQL Server and Hadoop platform components in order to discover key business processes of the organization based on event data. We then established a process mining framework that visualizes the process performance indicators and proposes workflow design changes and carries out statistical tests for identifying performance fluctuations by particularly using an in-memory parallel processing framework, named Apache Spark. © 2019 The Authors. Published by Elsevier B.V.en10.1016/j.procs.2019.09.048info:eu-repo/semantics/openAccessApache SparkBusiness Process ManagementData MinigEvent LogsProcess MiningSQLConference Object158244253N/A