EFFICIENT SYNTACTIC PROCESS DIFFERENCE DETECTION AND ITS APPLICATION TO PROCESS SIMILARITY SEARCH

Abstract

Nowadays, business process management plays an important role in the management of organizations. More and more organizations describe their operations as business processes. It is common for organizations to have collections of thousands of business process models. The same process is usually modeled differently due to the different rules or habits of different organizations and departments. Even in the subsidiaries of the same organization, process models vary from each other, because these process models are redesigned from time to time to continuously enhance the efficiency of management and operations. Therefore, techniques are required to analyze differences between similar process models.  Current techniques can detect operations required to modify one process model to the other. However, these operations are based on activities and the syntactic meanings are limited.  In this paper, we define differences based on workflow patterns and propose a technique to detect these differences efficiently. Besides that we propose a metric that can compute process similarity based on detected syntactic differences.  To the best of our knowledge, this is the first technique that returns a list of syntactic differences while computing a similarity score between two process models. The experiment shows that these differences indeed exist in real-life process models and are useful to analyze differences between business process models; the experiment also shows that the metric for process similarity based on detected differences works well in terms of the quality of similarity search and the average precision score is 0.8.