Organizing Documented Processes

A frequent hurdle in applying Business Process Management(BPM) to large enterprises is that, since business processes are not only numerous but also documented in an engagement in multiple representations,it is difficult to work with the documented \´as-is\´ or \´to-be\´ state of the business, leave aside make accurate transformation decisions. In this paper, we consider the problem of how to reconcile and organize documented information about processes into groups that convey inter-process similarity. The discovered knowledge can be used for many applications like search, e.g., find all requirements across all available processes that are similar to those of the "Account Receivable" process. The method has been tested on a dataset consisting of hundreds of processes documented in Word and Visio, wherefrom it could find process clusters that significantly boosted search performance.