Page 43 -

1.7 Outlook 25

and association rule learning. Process mining can be seen as a bridge between the
preliminaries presented in both chapters.
Part II focuses on one particular process mining task: process discovery. Chap-
ter 4 discusses the input needed for process mining. The chapter discusses different
input formats and issues related to the extraction of event logs from heterogeneous
data sources. Chapter 5 presents the α-algorithm step-by-step in such a way that the
reader can understand how it works and see its limitations. This simple algorithm
has problems dealing with less structured processes. Nevertheless, it provides a ba-
sic introduction into the topic and serves as a “hook” for discussing more advanced
algorithms and general issues related to process mining. Chapter 6 introduces more
advanced process discovery approaches. This way the reader gets a good under-
standing of the state-of-the-art and is guided in selecting suitable techniques.
Part III moves beyond process discovery, i.e., the focus is no longer on discov-
ering the control-flow. Chapter 7 presents conformance checking approaches, i.e.,
techniques to compare and relate event logs and process models. It is shown that
conformance can be quantified and that deviations can be diagnosed. Chapter 8 fo-
cuses on other perspectives: the organizational perspective, the case perspective, and
the time perspective. Chapter 9 shows that process mining can also be used to sup-
port operational processes at runtime, i.e., while cases are running it is possible to
detect violations, make predictions, and provide recommendations.
Part IV guides the reader in successfully applying process mining in practice.
Chapter 10 provides an overview of the different process mining tools. The next
two chapters are based on the observation that there are essentially two types of
processes: “Lasagna processes” and “Spaghetti processes”. Lasagna processes are
well-structured and relatively simple. Therefore, process discovery is less interest-
ing. The techniques presented in Part III are most relevant for Lasagna processes.
The added value of process mining can be found in conformance checking, detailed
performance analysis, and operational support. Chapter 11 explains how process
mining can be applied in such circumstances and provides various real-life exam-
ples. Spaghetti processes are less structured. Therefore, the added value of process
mining shifts to providing insights and generating ideas for better controlled pro-
cesses. Advanced techniques such as prediction are less relevant for Spaghetti pro-
cesses. Chapter 12 shows how to apply process mining in such less-structured en-
vironments.
Part V takes a step back and reflects on the material presented in the preced-
ing parts. Chapter 13 provides a broader vision on the topic by comparing process
modeling with cartography, and relating BPM systems to navigation systems pro-
vided by vendors such as TomTom, Garmin, and Navigon. The goal of this chapter
is to provide a refreshing view on process management and reveal the limitations
of existing information systems. Chapter 14 concludes the book by summarizing
improvement opportunities provided by process mining. The chapter also discusses
some of the key challenges and provides concrete pointers to start applying the ma-
terial presented in this book.