Discovering Business Processes in Legacy Systems Using Business Rules and Log Mining

This paper shows a method to semi-automate the discovery of business processes implemented implicitly in the source code of legacy systems. The method identifies the encoding of the business activities through the business rules implemented in legacy source code. In this paper we propose a tool to executes the source code instrumentation, identifying the business rules implemented in the legacy system and enabling the creation of event logs. This allows using log mining techniques to discover the partial order of execution of the business rules. Thus it is possible to discover business processes implemented implicitly in the source code of legacy systems. This paper also presents a case study conducted in a medical information system.