Publicado em: 13/12/2012

Proposta de Tese em Engenharia de Software

UNIVERSIDADE FEDERAL DO RIO GRANDE DO SUL
INSTITUTO DE INFORMÁTICA
PROGRAMA DE PÓS-GRADUAÇÃO EM COMPUTAÇÃO
———————————————-
DEFESA DE PROPOSTA DE TESE

Aluno: Gleison Samuel do Nascimento
Orientador: Prof. Dr. Cirano Iochpe
Coorientador: Prof. Dr. Álvaro Freitas Moreira

Título: A Semi-Automatic Method to Discovery Business Processes Underlying the Source Code of Legacy Systems

Linha de Pesquisa: Engenharia de Software

Data: 19/12/2012
Horário: 14h
Local: Sala 220 (conselhos). Prédio 43412 – Instituto de Informática

Banca Examinadora:
Prof. Dr. Jano Moreira de Souza (UFRJ)
Prof. Dr. Marcelo Soares Pimenta (UFRGS)
Profa. Dra. Renata de Matos Galante (UFRGS)
Prof. Dr. Stanley Loh (ULBRA)

Presidente da Banca: Prof. Dr. Cirano Iochpe

Abstract:
Legacy systems perform useful tasks for the organization, but were developed with obsolete technologies. Normally, these systems include information’s which are fundamental to the organization and often are not documented. Consequently, organizations have difficulty in updating or rewriting their legacy systems.
To reduce these problems, many organizations are investing in modernizing their legacy systems, using technologies that ensure the rapid, efficient and continuous business evolution. In this context the technologies of Business Process Management (BPM) and Service Oriented Architecture (SOA) are alternatives for modernizing legacy systems.
Currently, many researches propose the integration of BPM and SOA technologies as the basis for the modernization of legacy systems. BPM provides tools for building and running business processes and SOA provides a standard interface that connects business processes and legacy systems (web services), which can reuse parts of the legacy source code.
However, modernization of legacy through BPM and SOA is not a simple task. Usually, business processes are programmed implicitly in the source code of legacy systems, without any external documentation. Thus, to identify and map the legacy system behavior to business processes is the main problem for the modernization of legacy systems through BPM and SOA.
There are no tools to support developers in the extraction of knowledge about the legacy system behavior.
Thus, this thesis proposes the building of a method to semi-automate the discovery of business processes implemented implicitly in the source code of legacy systems.
The method combines techniques of static and dynamic analysis to extract business processes implemented in legacy systems. Static analysis techniques identify business activities implemented in legacy source code. While dynamic analysis techniques allow the discovery of the partial order of the activities during the execution of the legacy system.
Thus, the method presented in this thesis, transferring the behavior of legacy source code for one or more business process models. Subsequently, these business processes models can be implemented in a Business Process Management System.