Manufacturing perspective of enterprise application integration: the state of the art review

Manufacturing enterprise today has become a matter of effective and efficient application of information technology and knowledge-based engineering. Several new manufacturing paradigms such as virtual enterprise (or extended enterprise) and mass customization have resulted in a highly distributed and autonomous manufacturing system. On the one hand, this will increase the competitiveness of a firm in terms of quickly meeting dynamic changes in the market; on the other hand, this will also increase the difficulty of integrating different information and knowledge systems residing in each member firm. This integration is also called enterprise application integration (EAI) (here the term application means information systems or software systems for supporting manufacturing or service activities). The methodology for EAI has been studied for at least a decade; but no satisfactory solution has been found from a practical viewpoint. EAI is becoming even more difficult due to the boom in various enterprise information and knowledge systems (and also to ever increasing competition in the technical software market). The study presented in this paper conducted a critical analysis of existing solutions to EAI. We consider EAI as having two generic issues: semantic integration and syntactic integration. The main problem of semantic integration reduces to the general problem of enterprise or business modelling. The main problem of syntactic integration reduces to the general problem of software architecture of enterprise applications which enables interoperability between any two EAs. While the first issue is studied in the manufacturing field, the second issue is studied in the computing engineering field. In this paper, both issues are put together under the context of EAI and studied. The result of the study, together with our experience with one Canadian manufacturing firm, has led to the identification of several issues to be addressed in the future. We also outline possible ways to approach these issues.