Cw Christian Günther

Process Mining Manifesto

Process mining techniques are able to extract knowledge from event logs commonly available in today’s information systems. These techniques provide new means to discover, monitor, and improve processes in a variety of application domains. There are two main drivers for the growing interest in process mining. On the one hand, more and more events are being recorded, thus, providing detailed information about the history of processes. On the other hand, there is a need to improve and support business processes in competitive and rapidly changing environments. This manifesto is created by the IEEE Task Force on Process Mining and aims to promote the topic of process mining. Moreover, by defining a set of guiding principles and listing important challenges, this manifesto hopes to serve as a guide for software developers, scientists, consultants, business managers, and end-users. The goal is to increase the maturity of process mining as a new tool to improve the (re)design, control, and support of operational business processes.

Process mining : A two-step approach to balance between underfitting and overfitting

Process mining includes the automated discovery of processes from event logs. Based on observed events (e.g., activities being executed or messages being exchanged) a process model is constructed. One of the essential problems in process mining is that one cannot assume to have seen all possible behavior. At best, one has seen a representative subset. Therefore, classical synthesis techniques are not suitable as they aim at finding a model that is able to exactly reproduce the log. Existing process mining techniques try to avoid such “overfitting” by generalizing the model to allow for more behavior. This generalization is often driven by the representation language and very crude assumptions about completeness. As a result, parts of the model are “overfitting” (allow only for what has actually been observed) while other parts may be “underfitting” (allow for much more behavior without strong support for it). None of the existing techniques enables the user to control the balance between “overfitting” and “underfitting”. To address this, we propose a two-step approach. First, using a configurable approach, a transition system is constructed. Then, using the “theory of regions”, the model is synthesized. The approach has been implemented in the context of ProM and overcomes many of the limitations of traditional approaches.

Trace clustering in process mining

Process mining has proven to be a valuable tool for analyzing operational process executions based on event logs. Existing techniques perform well on structured processes, but still have problems discovering and visualizing less structured ones. Unfortunately, process mining is most interesting in domains requiring flexibility. A typical example would be the treatment process in a hospital where it is vital that people can deviate to deal with changing circumstances. Here it is useful to provide insights into the actual processes but at the same time there is a lot of diversity leading to complex models that are difficult to interpret. This paper presents an approach using trace clustering, i.e., the event log is split into homogeneous subsets and for each subset a process model is created. We demonstrate that our approach, based on log profiles, can improve process mining results in real flexible environments. To illustrate this we present a real-life case study.

Using process mining to learn from process changes in evolutionary systems

Traditional information systems struggle with the requirement to provide flexibility and process support while still enforcing some degree of control. Accordingly, adaptive Process Management Systems (PMSs) have emerged that provide some flexibility by enabling dynamic process changes during runtime. Based on the assumption that these process changes are recorded explicitly, we present two techniques for mining change logs in adaptive PMSs; that is, we do not only analyse the execution logs of the operational processes, but also consider the adaptations made at the process instance level. The change processes discovered through process mining provide an aggregated overview of all changes that happened so far. Using process mining as an analysis tool we show in this paper how better support can be provided for truly flexible processes by understanding when and why process changes become necessary.

Change mining in adaptive process management systems

The wide-spread adoption of process-aware information systems has resulted in a bulk of computerized information about real-world processes This data can be utilized for process performance analysis as well as for process improvement In this context process mining offers promising perspectives So far, existing mining techniques have been applied to operational processes, i.e., knowledge is extracted from execution logs (process discovery), or execution logs are compared with some a-priori process model (conformance checking) However, execution logs only constitute one kind of data gathered during process enactment In particular, adaptive processes provide additional information about process changes (e.g., ad-hoc changes of single process instances) which can be used to enable organizational learning In this paper we present an approach for mining change logs in adaptive process management systems The change process discovered through process mining provides an aggregated overview of all changes that happened so far This, in turn, can serve as basis for all kinds of process improvement actions, e.g., it may trigger process redesign or better control mechanisms.

A generic import framework for process event logs

The application of process mining techniques to real-life corporate environments has been of an ad-hoc nature so far, focused on proving the concept. One major reason for this rather slow adoption has been the complicated task of transforming real-life event log data to the MXML format used by advanced process mining tools, such as ProM. In this paper, the ProM Import Framework is presented, which has been designed to bridge this gap and to build a stable foundation for the extraction of event log data from any given PAIS implementation. Its flexible and extensible architecture, adherence to open standards, and open source availability make it a versatile contribution to the general BPI community.

Process Mining Applied to the Test Process of Wafer Scanners in ASML

Process mining techniques attempt to extract nontrivial and useful information from event logs. For example, there are many process mining techniques to automatically discover a process model describing the causal dependencies between activities. Several successful case studies have been reported in literature, all demonstrating the applicability of process mining. However, these case studies refer to rather structured administrative processes. In this paper, we investigate the applicability of process mining to less structured processes. We report on a case study where the process mining (ProM) framework has been applied to the test processes of ASML (the leading manufacturer of wafer scanners in the world).This case study provides many interesting insights. On the one hand, process mining is also applicable to the less structured processes of ASML. On the other hand, the case study also shows the need for alternative mining approaches.

The need for a process mining evaluation framework in research and practice

Although there has been much progress in developing process mining algorithms in recent years, no effort has been put in developing a common means of assessing the quality of the models discovered by these algorithms. In this paper, we motivate the need for such an evaluation mechanism, and outline elements of an evaluation framework that is intended to enable (a) process mining researchers to compare the performance of their algorithms, and (b) end users to evaluate the validity of their process mining results.

Activity Mining by Global Trace Segmentation

Process Mining is a technology for extracting non-trivial and useful information from execution logs. For example, there are many process mining techniques to automatically discover a process model describing the causal dependencies between activities . Unfortunately, the quality of a discovered process model strongly depends on the quality and suitability of the input data. For example, the logs of many real-life systems do not refer to the activities an analyst would have in mind, but are on a much more detailed level of abstraction. Trace segmentation attempts to group low-level events into clusters, which represent the execution of a higher-level activity in the (available or imagined) process meta-model. As a result, the simplified log can be used to discover better process models. This paper presents a new activity mining approach based on global trace segmentation. We also present an implementation of the approach, and we validate it using a real-life event log from ASML’s test process.

Using minimum description length for process mining

In the field of process mining, the goal is to automatically extract process models from event logs. Recently, many algorithms have been proposed for this task. For comparing these models, different quality measures have been proposed. Most of these measures, however, have several disadvantages; they are model-dependent, assume that the model that generated the log is known, or need negative examples of event sequences. In this paper we propose a new measure, based on the minimal description length principle, to evaluate the quality of process models that does not have these disadvantages. To illustrate the properties of the new measure we conduct experiments and discuss the trade-off between model complexity and compression.