Recently, process mining emerged as a new scientific discipline on the interface between process models and event data. Conventional Business Process Management (BPM) and Workflow Management (WfM) approaches and tools are mostly model-driven with little consideration for event data. Data Mining (DM), Business Intelligence (BI), and Machine Learning (ML) focus on data without considering end-to-end process models. Process mining aims to bridge the gap between BPM and WfM on the one hand and DM, BI, and ML on the other hand. The challenge is to turn torrents of event data (“Big Data”) into valuable insights related to performance and compliance. The results can be used to identify and understand bottlenecks, inefficiencies, deviations, and risks. Process mining helps organizations to “mine their own business”: they are enabled to discover, monitor and improve real processes by extracting knowledge from event logs. In his keynote, Wil van der Aalst will provide an overview of this exciting field that will become more and more important for the Information Systems (IS) discipline.

• 10-ecis2013keynote-mineyourownbusiness.pdf

Wil van der Aalst

Operational processes leave trails in the information systems supporting them. Such event data are the starting point for process mining - an emerging scientific discipline relating modeled and observed behavior. Whereas an event log describes example behavior of the underlying process, a process model aims to describe an abstraction of the same process. Models may be descriptive or normative. Descriptive models aim to describe the underlying process and are used for discussion, performance analysis, obtaining insights, and prediction. Normative models describe the desired behavior and are used for workflow management, system configuration, auditing, compliance management, and conformance checking. Differences between modeled and observed behavior may point to undesirable deviations or inadequate models. In this paper, we discuss challenges related to finding the “right” process, i.e., the process model that describes the real underlying process or a process that behaves as desired.

• 9-rcis2013keynotequestforthe_rightprocess-wvda.pdf

The keynote reports on a journey through the history of process mining. From the roots that date back as much as to the 1950s, through the early beginnings of process mining as we know it, and all the way up to today. An overview of the milestones and progress in the field is given and process mining is related to neighboring fields.Slides

• 8-process-mining-camp2013-history-of-process-mining.pdf

Today's hospital information systems contain a wealth of data. Typically, these systems record information about (business) processes in the form of so-called event logs. These logs can be used as input for process mining such that process-related information can be extracted. In a healthcare context, process mining can be used to provide insights into how healthcare processes are really executed. People involved in these processes tend to have only an idealized version of the real process in mind. However, based on event data, process mining can be used to extract real process knowledge (e.g. process models) in order to discover, monitor, and improve care processes. Prof. Wil van der Aalst is well-known for his process mining research and he and his team applied process mining in over 100 organizations (www.processmining.org). Given the characteristics of care processes and available data, process mining can be of great value for hospitals (www.healthcare-analyticsprocessmining.org). In his talk, prof. Van der Aalst will introduced the concept of process mining, provide an overview of available analysis techniques, and show real-life applications of process mining in healthcare.

• 7-qut-ihbi-procmin-in-hc.pdf

Overview of process mining techniques with some emphasis on new types of analyses enabled by genetic process mining techniques and process trees. Operational processes leave trails in the information systems supporting them. Such event data are the starting point for process mining - an emerging scientific discipline relating modeled and observed behavior. Whereas an event log describes example behavior of the underlying process, a process model aims to describe an abstraction of the same process. Models may be descriptive or normative. Descriptive models aim to describe the underlying process and are used for discussion, performance analysis, obtaining insights, and prediction. Normative models describe the desired behavior and are used for workflow management, system configuration, auditing, compliance management, and conformance checking. Differences between modeled and observed behavior may point to undesirable deviations or inadequate models. In his talk prof. Van der Aalst discusses challenges related to finding the “right” process, i.e., the process model that describes the real underlying process or a process that behaves as desired. Concrete topics that will be addressed are: the representational bias in process mining (covering different process representations ranging from C-nets to process trees), divide-and-conquer approaches (splitting logs and models), the importance of alignments, and advanced forms of process discovery (e.g., finding compromises between a reference model and observed behavior and mining configurable process models).

• 6-pais-lab-wvda-april-2013.pdf

Introduction to operational support, including techniques for real-time prediction.

• 5-siks_lets-play-2013-operational-support-wvda.pdf

Introduction to process discovery focusing on region-based process mining techniques.

• 4-siks-lets-play-2013-discovery-wvda.pdf

Operational processes leave trails in the information systems supporting them. Such event data are the starting point for process mining: an emerging scientific discipline relating modeled and observed behavior. Whereas an event log describes example behavior of the underlying process, a process model aims to describe an abstraction of the same process. Models may be descriptive or normative. Descriptive models aim to describe the underlying process and are used for discussion, performance analysis, obtaining insights, and prediction. Normative models describe the desired behavior and are used for workflow management, system configuration, auditing, compliance management, and conformance checking. Differences between modeled and observed behavior may point to undesirable deviations or inadequate models. In his talk, prof. Wil van der Aalst will discuss challenges related to finding the “right” process, i.e., the process model that describes the real underlying process or a process that behaves as desired. He will show how to align reality and models using techniques developed in the context of ProM. Moreover, he will discuss different metrics (fitness, precision, generalization, and simplicity) used to judge process mining results and to quantify conformance. All of this relates to one of the main research challenges in information science: What makes a good process model?

• 3-icwe-2013-keynote-wvda.pdf

Operational processes leave trails in the information systems supporting them. Such event data are the starting point for process mining - an emerging scientific discipline relating modeled and observed behavior. The relevance of process mining is increasing as more and more event data become available. The increasing volume of such data (“Big Data”) provides both opportunities and challenges for process mining. In this paper we focus on two particular types of process mining: process discovery (learning a process model from example behavior recorded in an event log) and conformance checking (diagnosing and quantifying discrepancies between observed behavior and modeled behavior). These tasks become challenging when there are hundreds or even thousands of different activities and millions of cases. Typically, process mining algorithms are linear in the number of cases and exponential in the number of different activities. This keynote proposes a very general divide-and-conquer approach that decomposes the event log based on a partitioning of activities. Unlike existing approaches, this paper does not assume a particular process representation (e.g., Petri nets or BPMN) and allows for various decomposition strategies (e.g., SESE- or passage-based decomposition). Moreover, the generic divide-and-conquer approach reveals the core requirements for decomposing process discovery and conformance checking problems.

• 2-fedcsis2013-keynote-wvda.pdf

Recent breakthroughs in process mining research make it possible to discover, analyze, and improve business processes based on event data. More and more events are recorded by a wide variety of systems ranging from embedded systems to enterprise information systems. The growth of event data provides many opportunities for process mining. However, current approaches cannot deal well with heterogeneous processes that change over time while emitting torrents of event data. Process mining is typically done for an isolated well-defined process in steady-state in an offline and centralized fashion. In his talk prof. Wil van der Aalst proposes the notion of process cubes where events and process models are organized using different dimensions. The idea is related to the well-known OLAP (Online Analytical Processing) data cubes and associated operations such as slice, dice, roll-up, and drill-down. However, there are also significant differences because of the process-related nature of event data. For example, process discovery based on events is incomparable to computing the average or sum over a set of values. Prof. Van der Aalst will also relate the process cube notion to distributed process mining techniques where event data are partitioned to improve performance and scalability.

• 1-ap-bpm2013-process-cube-keynote.pdf

• composing-workflow-activities-on-the-basis-of-data-flow-structures1.pdf

• declarative-modeling-an-academic-dream-or-the-future-for-bpm1.pdf

• presentation_bpm13.pptx

• declarative-modeling-an-academic-dream-or-the-future-for-bpm1.pdf

• bpi.pptx

• 2013-08-26_bpi_buijs_pareto.pdf

• petri_nets.pptx

The 'Aan de Dommel' series of lectures presents the work of the four new university professors. Prof. dr. ir. Maarten Steinbuch (W), prof. dr. ir. Wil van der Aalst (W&I), prof. dr. ir. Anthonie Meijers (IE&IS) and prof. dr. ir. René Janssen (ST and TN) . Prof. Van der Aalst provides an overview of the process mining research done at TU/e. The presentation relates process mining to the incredible growth of even data (Big Data) and also introduces the new Data Science Center Eindhoven (DSC/e). DSC/e strives to become the internationally leading expertise center for data science research and education. DSC/e's launch symposium will take place on December 2nd 2013.

• dommellezing-sept2013-wvda.pdf