Process Mining – The hot topic in business process management

Tracking the footprints of business processes; that is the aim of process mining, the new hot topic in business process management. The overarching support of processes through business process models, integrated databases and the ERP systems built upon them, has led in the past to many successful projects resulting, in some cases, in significant increases in productivity. The hot topic developing in business process management today is process mining, a subject that addresses processes from their design through to their execution. It is only on the execution of a process that it becomes apparent whether and where the expected benefits have actually materialized.

The focus is now being placed on the behaviour of individual business process instances. Theoretically, the instances should follow the business process model, or rather the software configured with its help. But this is only then the case if the model encompasses the logic of every possible actual business instance, if the software is configured accordingly and no unforeseen deviations occur in the real world. Every procedure is then pre-configured and runs automatically. But, generally speaking, this is only ever the case in theory. By contrast, in the real world, changes happen in the intended allocation of organizational units to functions, or malfunctions occur. At this point a human intervenes and makes ad-hoc changes to procedures defined in the target model.

Buch von August-Wilhelm Scheer

Deviations of this type justify the interest in the execution of real process instances. Coming from the systems that execute them, they leave behind a data trail in the form of event messages in so-called log-files which are then available for evaluations, or process mining. The evaluation of process executions is therefore primarily data driven.

Process mining is therefore all about the capture of a business process’s data trail in a log-file during its execution and its subsequent evaluation. Deviations can be identified by comparing the existing target-state model, along with the data trail in the log-file, with an actual-state model generated for this purpose. These deviations are analysed in order to adapt the target-state model to reality and to make recommendations for organizational improvements. Process mining can give information on whether compliance rules have been adhered to or broken during the execution of a process, the points at which bottlenecks in capacity occur, whether deviations from the intended allocation of capacity have occurred and the performance of run times and quality etc. Drawing automated conclusions for the improvement of processes requires the use of artificial intelligence (AI) methodologies. A model improved upon and adapted to reality is then defined as the new target model. The comparison of target- and actual-processes must be seen as a continuing task, as ever-occurring changes will continually challenge the suitability of target processes.

Many real-world usage scenarios and the success of process mining software show that the approach taken by process mining can achieve remarkable results.

Want to know more?

Facebook Icon Twitter icon - Logo Xing Icon - Logo Linkedin Logo - Icon Icon von Youtube Contact icon