Jcam Joos Buijs

XES, XESame, and ProM 6

Process mining has emerged as a new way to analyze business processes based on event logs. These events logs need to be extracted from operational systems and can subsequently be used to discover or check the conformance of processes. ProM is a widely used tool for process mining. In earlier versions of ProM, MXML was used as an input format. In future releases of ProM, a new logging format will be used: the eXtensible Event Stream (XES) format. This format has several advantages over MXML. The paper presents two tools that use this format - XESame and ProM 6 - and highlights the main innovations and the role of XES. XESame enables domain experts to specify how the event log should be extracted from existing systems and converted to XES. ProM 6 is a completely new process mining framework based on XES and enabling innovative process mining functionality.

On the role of fitness, precision, generalization and simplicity in process discovery

Process discovery algorithms typically aim at discovering process models from event logs that best describe the recorded behavior. Often, the quality of a process discovery algorithm is measured by quantifying to what extent the resulting model can reproduce the behavior in the log, i.e. replay fitness. At the same time, there are many other metrics that compare a model with recorded behavior in terms of the precision of the model and the extent to which the model generalizes the behavior in the log. Furthermore, several metrics exist to measure the complexity of a model irrespective of the log.

A genetic algorithm for discovering process trees

Existing process discovery approaches have problems dealing with competing quality dimensions (fitness, simplicity, generalization, and precision) and may produce anomalous process models (e.g., deadlocking models). In this paper we propose a new genetic process mining algorithm that discovers process models from event logs. The tree representation ensures the soundness of the model. Moreover, as experiments show, it is possible to balance the different quality dimensions. Our genetic process mining algorithm is the first algorithm where the search process can be guided by preferences of the user while ensuring correctness.

Quality dimensions in process discovery : the importance of fitness, precision, generalization and simplicity

Process discovery algorithms typically aim at discovering process models from event logs that best describe the recorded behavior. Often, the quality of a process discovery algorithm is measured by quantifying to what extent the resulting model can reproduce the behavior in the log, i.e. replay fitness. At the same time, there are other measures that compare a model with recorded behavior in terms of the precision of the model and the extent to which the model generalizes the behavior in the log. Furthermore, many measures exist to express the complexity of a model irrespective of the log. In this paper, we first discuss several quality dimensions related to process discovery. We further show that existing process discovery algorithms typically consider at most two out of the four main quality dimensions: replay fitness, precision, generalization and simplicity. Moreover, existing approaches cannot steer the discovery process based on user-defined weights for the four quality dimensions. This paper presents the ETM algorithm which allows the user to seamlessly steer the discovery process based on preferences with respect to the four quality dimensions. We show that all dimensions are important for process discovery. However, it only makes sense to consider precision, generalization and simplicity if the replay fitness is acceptable.

Mining configurable process models from collections of event logs

Existing process mining techniques are able to discover a specific process model for a given event log. In this paper, we aim to discover a configurable process model from a collection of event logs, i.e., the model should describe a family of process variants rather than one specific process. Consider for example the handling of building permits in different municipalities. Instead of discovering a process model per municipality, we want to discover one configurable process model showing commonalities and differences among the different variants. Although there are various techniques that merge individual process models into a configurable process model, there are no techniques that construct a configurable process model based on a collection of event logs. By extending our ETM genetic algorithm, we propose and compare four novel approaches to learn configurable process models from collections of event logs. We evaluate these four approaches using both a running example and a collection of real event logs.

Towards cross-organizational process mining in collections of process models and their executions

Variants of the same process may be encountered in different organizations, e.g., any municipality will have a process to handle building permits. New paradigms such as Software-as-a-Service (SaaS) and Cloud Computing stimulate organizations to share a BPM infrastructure. The shared infrastructure has to support many processes and their variants. Dealing with such large collections of similar process models for multiple organizations is challenging. However, a shared BPM infrastructure also enables cross-organizational process mining. Since events are recorded in a unified way, it is possible to cross-correlate process models and the actual observed behavior in different organizations. This paper presents a novel approach to compare collections of process models and their events logs. The approach is used to compare processes in different Dutch municipalities.

Improving business process models using observed behavior

Process-aware information systems (PAISs) can be configured using a reference process model, which is typically obtained via expert interviews. Over time, however, contextual factors and system requirements may cause the operational process to start deviating from this reference model. While a reference model should ideally be updated to remain aligned with such changes, this is a costly and often neglected activity. We present a new process mining technique that automatically improves the reference model on the basis of the observed behavior as recorded in the event logs of a PAIS. We discuss how to balance the four basic quality dimensions for process mining (fitness, precision, simplicity and generalization) and a new dimension, namely the structural similarity between the reference model and the discovered model. We demonstrate the applicability of this technique using a real-life scenario from a Dutch municipality.

Towards Improving the Representational Bias of Process Mining

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. Process discovery—discovering a process model from example behavior recorded in an event log—is one of the most challenging tasks in process mining. A variety of process discovery techniques have been proposed. Most techniques suffer from the problem that often the discovered model is internally inconsistent (i.e., the model has deadlocks, livelocks or other behavioral anomalies). This suggests that the search space should be limited to sound models. In this paper, we propose a tree representation that ensures soundness. We evaluate the impact of the search space reduction by implementing a simple genetic algorithm that discovers such process trees. Although the result can be translated to conventional languages, we ensure the internal consistency of the resulting model while mining, thus reducing the search space and allowing for more efficient algorithms.

Comparing business process variants using models and event logs

Organizations realize that benefits can be achieved by closely working together on the design of their business processes. But even when there is a joint design for a particular business process, the way individual organizations carry out that process may differ – either wittingly or unwittingly. This paper proposes an analytical approach that helps to compare how different organizations execute essentially the same process. This comparison is based on the alignment of recorded process behavior with explicitly defined process models. The distinctive feature of the proposed approach is that it supports the comparison of the actual execution of a process within a particular organization with its intended design, as well as with the variants of that design by other organizations. In this way, organizations can develop a better understanding of how they can work together and further standardize a process of common interest. We include an industrial case study from the context of the CoSeLoG project to demonstrate the value of this comparison approach.

Mining Process Performance from Event Logs

In systems where process executions are not strictly enforced by a predefined process model, obtaining reliable performance information is not trivial. In this paper, we analyzed an event log of a real-life process, taken from a Dutch financial institute, using process mining techniques. In particular, we exploited the alignment technique [2] to gain insights into the control flow and performance of the process execution. We showed that alignments between event logs and discovered process models from process discovery algorithms reveal insights into frequently occurring deviations and how such insights can be exploited to repair the original process models to better reflect reality. Furthermore, we showed that the alignments can be further exploited to obtain performance information. All analysis in this paper is performed using plug-ins within the open-source process mining toolkit ProM.