Chapter 7. Process mining: from descriptive to predictive and prescriptive data-driven analysis of business processes

Business processes in modern organizations are supported by different types of IT systems. These systems often provide a large amount of data that can be exploited to monitor, analyze and improve business processes. In this chapter, we give an overview of process mining, a set of techniques that exploit process data normally available as so-called process event logs. We focus first on the more traditional use cases of process mining, which have appeared first in the literature: process discovery, i.e., discovering models of business processes from event logs, and conformance checking, i.e., assessing the conformance between process models and event logs. Then, we focus on predictive process monitoring, a discipline within process mining that has emerged relatively recently and that deals with the development of predictive models using process data, mostly through the application of machine learning techniques. Finally, we give an outlook on the future by discussing advanced aspects of the application of machine learning to process data, such as the explainability of predictive models and prescriptive process monitoring. We conclude by giving a brief overview of the open-source tools currently available to implement the process data analysis techniques presented in this chapter.