Abstract
Purpose - In this paper a solution is proposed for automating the task of matching business process models and searching for correspondences with regard to the model semantics, thus improving the efficiency of such works.Design/methodology/approach - A method is proposed based on combining several semantic technologies. The research follows a design-science oriented approach in that a method together with its supporting artifacts has been developed. Its application allows for reusing legacy models and automatedly determining semantic similarity.Findings - The method has been applied and the first findings suggest the effectiveness of the approach. The results of applying the method show its feasibility and significance. The suggested heuristic computing of semantic correspondences between semantically heterogeneous business process models is flexible and can directly support domain users.Research limitations/implications - Even though a solution can be offered that is directly usable, so far the full complexity of the natural language as given in model element labels is not yet completely resolvable. Here further research could contribute to the potential optimizations and refinement of automatic matching and linguistic procedures. However, an open research question could be solved.Practical implications - The method presented is aimed at adding to the methods in the field of business process management and could extend the possibilities of automating support for business analysis. Originality/value - The suggested combination of semantic technologies is innovative and addresses the aspect of semantic heterogeneity in a holistic way, which is novel to the field.
Purpose - In this paper a solution is proposed for automating the task of matching business process models and searching for correspondences with regard to the model semantics, thus improving the efficiency of such works.Design/methodology/approach - A method is proposed based on combining several semantic technologies. The research follows a design-science oriented approach in that a method together with its supporting artifacts has been developed. Its application allows for reusing legacy models and automatedly determining semantic similarity.Findings - The method has been applied and the first findings suggest the effectiveness of the approach. The results of applying the method show its feasibility and significance. The suggested heuristic computing of semantic correspondences between semantically heterogeneous business process models is flexible and can directly support domain users.Research limitations/implications - Even though a solution can be offered that is directly usable, so far the full complexity of the natural language as given in model element labels is not yet completely resolvable. Here further research could contribute to the potential optimizations and refinement of automatic matching and linguistic procedures. However, an open research question could be solved.Practical implications - The method presented is aimed at adding to the methods in the field of business process management and could extend the possibilities of automating support for business analysis. Originality/value - The suggested combination of semantic technologies is innovative and addresses the aspect of semantic heterogeneity in a holistic way, which is novel to the field.