Detecting potential weaknesses in business processes : An exploration of semantic pattern matching in process models

Purpose – Automating the task of identifying process weaknesses using process models is promising, as many organizations have to manage a large amount of process models. The purpose of this paper is to introduce a pattern-based approach for automatically detecting potential process weaknesses in semantic process models, thus supporting the task of business process improvement. Design/methodology/approach – Based on design research, combined with a case study, the authors explore the design, application and evaluation of a pattern-based process weakness detection approach within the setting of a real-life case study in a German bank. Findings – Business process weakness detection can be automated to a remarkable extent using pattern matching and a semantic business process modeling language. A case study provided evidence that such an approach highly supports business process analysts. Research limitations/implications – The presented approach is limited by the fact that not every potential process weakness detected by pattern matching is really a weakness but just gives the impression to be one. Hence, after detecting a weakness, analysts still have to decide on its authenticity. Practical implications – Applying weakness patterns to semantic process models via pattern matching allows organizations to automatically and efficiently identify process improvement potentials. Hence, this research helps to avoid time- and resource-consuming manual analysis of process model landscapes. Originality/value – The approach is not restricted to a single modeling language. Furthermore, by applying the pattern matching approach to a semantic modeling language, the authors avoid ambiguous search results. A case study proves the usefulness of the approach.