An Outlook on Semantic Business Process Mining and Monitoring

Abstract

Abstract. Semantic Business Process Management (SBPM) has been proposed as an extension of BPM with Semantic Web and Semantic Web Services (SWS) technologies in order to increase and enhance the level of automation that can be achieved within the BPM life-cycle. In a nutshell, SBPM is based on the extensive and exhaustive conceptualization of the BPM domain so as to support reasoning during business processes modelling, composition, execution, and analysis, leading to important enhancements throughout the life-cycle of business processes. An important step of the BPM life-cycle is the analysis of the processes deployed in companies. This analysis provides feedback about how these processes are actually being executed (like common control-flow paths, performance measures, detection of bottlenecks, alert to approaching deadlines, auditing, etc). The use of semantic information can lead to dramatic enhancements in the state-of-the-art in analysis techniques. In this paper we present an outlook on the opportunities and challenges on semantic business process mining and monitoring, thus paving the way for the implementation of the next generation of BPM analysis tools.

Ontology-based metrics computation for business process analysis

“…We have previously argued for the use of semantic technologies, namely ontologies and Problem-Solving Methods [17], as a means to enhance the state of the art in BPA [2]. In the light of this vision, we have defined Core Ontology for Business pRocess Analysis (COBRA) [14] which provides a core terminology where business practitioners can map domain-specific knowledge in order to analyse their business processes.…”

A Core Ontology for Business Process Analysis

“…Future work will thus be devoted to extending our work along the vision previously presented in [11]. In particular, next steps will be devoted to the definition of a metrics computation engine that will support the computation of both generic and user defined metrics, and the implementation of a classification ProblemSolving Method for detecting process deviations.…”

“…On the other hand, bottom-up approaches integrate some sort of light-weight automated reasoning machinery with existing BPM solutions, see for instance [9,10,11]. These efforts are mainly dominated by researchers from the BPM area, where knowledge-based technologies have not been widely used so far.…”

“…Analysis Results are Instantaneous Entities of a Quantitative or Qualitative nature 3 . Being part of the core ontology for analysing business process, this allow us to reuse results across different types of BPA which paves the way for enhancing current analysis techniques [11]. For instance, metrics computed at runtime can be reused when performing RBE, mining results can be applied during monitoring, etc.…”

A stage-based model for enabling decision support in process mining

“…The Process Mining Manifesto [92] indicates that process mining is classified into four perspectives, whereas other researchers [27,64] Other perspectives were discussed in other studies, namely, the semantic perspective [23], data or information perspective [70] and quality perspective [86].…”