Actions

Enterprise Integration

Enterprise Integration is a broad implementation of information technology to link various functional units within a business enterprise. On a wider scale, it may also integrate strategic partners in an enterprise configuration. In a manufacturing enterprise, enterprise integration (EI) may be regarded as an extension of computer integrated manufacturing (CIM) that integrates financial or executive decision-support systems with manufacturing, tracking and inventory systems, product-data management, and other information systems.[1]


Historical Background[2]
Enterprise integration has been discussed since the early days of computers in industry and especially in the manufacturing industry with computer-integrated manufacturing (CIM) as the acronym for operations integration. In spite of the different understandings of the scope of integration in CIM it has always stood for information integration across at least parts of the enterprise. Information integration essentially consists of providing the right information, at the right place, at the right time. In the 1990s enterprise integration and enterprise engineering became a focal point of discussions with active contribution of many disciplines. The state of the art in enterprise engineering and integration by the end of the 1990s has been rather confusing, according to Jim Nell and Kurt Kosanke (1997): On one hand, it claims to provide solutions for many of the issues identified in enterprise integration. On the other hand, the solutions seem to compete with each other, use conflicting terminology and do not provide any clues on their relations to solutions on other issues. Workflow modelling, business process modelling, business process reengineering (BPR), and concurrent engineering all aim toward identifying and providing the information needed in the enterprise operation. In addition, numerous integrating-platforms concepts are promoted with only marginal or no recognition or support of information identification. Tools claiming to support enterprise modelling exist in very large numbers, but the support is rather marginal, especially if models are to be used by the end user, for instance, in decision support.


Evolution of Enterprise Integration - The Consumerization of Integration[3]
For integration platforms, the way businesses work has drastically changed from the days when the traditional integration tools were created. Business apps have become consumerized, mobile-centric, self-service, and low cost, but the Integration tools to connect these apps, often remain complex, technical and, at times, more expensive than the apps they are integrating.

  • Integrations Must Be Easy to Build: It should be easy enough to use that even non-developers can build integrations. With the SaaS adoption in lines of business, it is imperative that lines of business are able to build their own integrations and workflows without any fear. For example, Sales /Marketing/Customer Support Ops should be able to integrate applications and automate their workflows without having to worry about the underlying technology, including security, scalability, error handling, duplicate record handling, being able to easily undo errors, and more. IT should still be able to govern and manage the integrations in a company, but not necessarily be the one building all the integrations in the company.
  • Integrations Must Be Agile and Flexible: Integration tools should be flexible, allowing integrations to be changed at any time. In the past, integrations were ‘built-to-last’. Once you have your Order-to-cash process going between your SAP and Salesforce, you don’t go back to change it often – the requirements were more stable and, besides, the consultants that did these integrations have moved on. This is no longer the case. Business apps and the processes and workflows around them are constantly changing as new apps are added and existing apps are changed. Constant change is the norm. An integration platform must be simpler to use, agile and flexible so that anyone, from business analysts to app admins can roll out new integrations quickly and not depend on IT / developers.
  • Integrations Must Scale to 1000s of Apps Not 10s: In the early days of integration, most of the integrations were done between a few functional app suites. Traditional integration tools are primarily optimized for a few heavy duty integrations between such suites. Today companies use more than one thousand apps, which must work together. For example, a marketing team will be using an entire stack of applications around core marketing automation apps like HubSpot - connecting to WordPress, MailChimp, EventMobi, Slack etc. To be able to integrate and automate your marketing workflows, the platform needs to support all of these apps.


Components of an Integrated Enterprise[4]

  • Interface Engine controlling data flows between applications;
  • Master Person Index: A database and rules engine that contain a unique identifier for every user in the enterprise, regardless of how many other unique application user indices exist;
  • Single Sign-on with common authentication: a process that Permits a user to enter one name and password in order to access multiple applications;
  • Context Management: a process for passing user interface (display) context (user, and item information) from one application to the other without user involvement;
  • Common Code Sets allow for the maintenance and transfer of data that can be used and “understood” in multiple systems
  • Data Warehouse: Permits access of information across the enterprise through use of a central data repository or storage system
  • Semantic Interoperability: data is not only transferred from one application to another, but its meaning is translated to match the receiver


Enterprise Integration
source: NIST


Enterprise Integration Needs[5]

  • Identify the right information: requires a precise knowledge of the information needed and created by the different activities in the enterprise operation. Knowledge has to be structured in the form of an accurate model of the enterprise operation, which describes product and administrative information, resources and organizational aspects of the operational processes and allows what-if analysis in order to optimize these processes.
  • Provide the right information at the right place: requires information sharing systems and integration platforms capable of handling information transaction across heterogeneous environments consisting of heterogeneous hardware, different operating systems and monolithic software applications (legacy systems). Environments which cross organizational boundaries and link the operation of different organizations on a temporal basis and with short set-up times and limited time horizon (extended and virtual enterprises).
  • Update the information in real time to reflect the actual state of the enterprise operation: requires not only the up-date of the operational data (information created during the operation), but adapting to environmental changes, which may originate from new customer demands, new technology, new legislation or new philosophies of the society at large. Changes may require modification of the operational processes, the human organization or even the overall scope and goals of the enterprise.
  • Coordinate business processes: requires precise modelling of the enterprise operation in terms of business processes, their relations with each other, with information, resources and organization. This goes far beyond exchange of information and information sharing. It takes into account decisional capabilities and know-how within the enterprise for real time decision support and evaluation of operational alternatives.
  • Organize and adapt the enterprise: requires very detailed and up-to-date knowledge of both the current state of the enterprise operation and its environment (market, technology, society). Knowledge has to be available a priori and very well structured to allow easy identification of and access to relevant information.


Benefits of Enterprise Integration[6]
Eliminating complicated point-to-point integration allows businesses to stay connected, innovative, and competitive. Moreover, enterprise integration provides businesses the insight to make better decisions. In order to make well-informed and incisive decisions about your company, it is vital to have real-time, up-to-date information. To do this, data from all your systems, services, and applications must be available at all times, seamlessly. Enterprise integration provides a number of benefits for business:

  • System interconnectivity: Connect people, processes, systems, and technologies within your enterprise ecosystem.
  • Improved productivity throughout the company: Enable communication across differing systems to permit everyone access to the information they need, anytime.
  • The right information when you need it: Provide accurate data regardless of what system you are deploying.
  • Real-time updates: Ensure that you are always up to date on your business needs with real-time updates.
  • Coordinate business processes: Automate business processes, further improving productivity across your company.


See Also

Service Oriented Architecture (SOA)
Enterprise Service Bus (ESB)
Enterprise Information Integration (EII)
Enterprise Application Integration (EAI)
Business Process Integration
Enterprise Architecture


References

  1. Defining Enterprise Integration Ramesh Gulati, Jerry Kahn and Robert Baldwin
  2. Historical Background Wikipedia
  3. Evolution of Enterprise Integration - The Consumerization of Integration Ross Garrett
  4. What are the Key Components of a Fully Integrated Enterprise HIMSS
  5. What does Enterprise Integration Need? Wikipedia
  6. Benefits of Enterprise Integration Mulesoft


Further Reading