An integrated enterprise is one in which the enterprise's business processes, organizational structure and information systems are all aligned to optimize the business objectives of the enterprise. However this goal is not easily achieved as enterprises have traditionally grown in a siloed manner where each functional unit has its own systems, processes and organizational structure. This is especially the case in enterprises that have grown inorganically through mergers and acquisitions. As a result, an enterprise does not have a unified view of its processes and systems, and how they align with each other and with the policies and objectives of the enterprise. In many enterprises this is a source of great inefficiency and, at systems level, a source of many integration problems.
Our goal is to investigate and come up with a set of architectures, models and methods that will help an enterprise achieve an integrated view of its processes and systems in a systematic manner. Among other things, this involves investigating how data, processes and information systems should be modeled to clearly bring out their business intent (semantic intent) and how they link up with business policies and objectives. We can then envisage an integration architecture where we use these modeling notations to build a set of reference conceptual models at the enterprise level, and map existing systems to these conceptual models, providing a semantic foundation for their integration. We can develop model driven techniques to process these models and automate the generation of platform specific integration solutions.
As a first step, we have focused on the data aspect, i.e. how data should be modeled to achieve an integrated view. As part of this, we have developed a conceptual modeling framework using which we can specify a unified conceptual model of enterprise data and map various application specific conceptual and physical models to this unified model. We have also developed tools to process these models: a query processing tool that can translate enterprise conceptual model queries into platform specific implementations, an ETL generation tool, and a tool to discover mappings between different models. A few more tools are in the pipeline.
Going forward we will be focusing on modeling of processes and services in order to unify all these aspects into an integrated view of the enterprise.
This group is headed by Sreedhar Reddy