|
|
|
|
|
Data Integration – An effective approach to address PLM implementation challenges and interoperability problems Interoperability problems are of concern in most companies, as they impact cost, performance and the ability to make sound business decisions. Interoperability issues also feature in any PLM implementation programme. Daista Software, a Datamation sister company, has identified the need to translate data from one proprietary data model to another as the root cause of interoperability problems. Yet, all current interoperability solutions involve translation of data in one way or another, be it automated (via interfaces) or not. The process of translating and moving information between disjointed applications is error prone - particularly where manual processes are present. These errors may impact data quality i.e. data accuracy, currency, integrity and consistency - any of which can make data not fit for purpose. A further source of data quality problems originates in the customisations and bespoke applications that are needed to deliver capabilities that commercial-off-the-shelf applications do not deliver. Many of these, particularly the bespoke applications, are needed to meet urgent business requirements, and therefore are often developed on the fly with little, or no, controls. As a result such applications lack proper data integrity and consistency checks causing data quality problems down the line. Current solutions are failing to solve interoperability problems because they are not addressing the underlying cause of the problems. They focus on facilitating the movement of data from one application to another through interfaces or data exchange processes. This is not sufficient. The data must also be of sufficient quality and integrity for its intended use and that any change must be managed, controlled and made traceable. Failure to address data quality issues at source results in the need to introduce non-value added activities to ensure that data is fit for purpose at point of use. Therefore, the often quoted role of Chief Information Officers (CIOs): “To make information available to the right people at the right place at the right time and in the right format”. Should be qualified to: “To make fit for purpose information available to the right people at the right place at the right time and in the right format, in a controlled and traceable way”. The Daista universal Integration Platform (uIP) To deliver fit for purpose information in a controlled and traceable way at point of use, it is necessary to remove the disjointedness between applications. This is only possible through “data integration” – the consolidation of data to form a consistent and integrated whole. Data integration is more than moving data from one application to another, or collating information from different sources and presenting them on one screen. Daista has developed its universal Integration Platform (uIP) to:
The Daista Approach As mentioned earlier, current approaches to solving interoperability problems focus on facilitating the movement of data between applications while leaving their databases disjointed. The Daista approach focuses on making data fit for purpose. Not only does this ensure that every activity adds value to the business, but also ensures that data is used within its business context. Furthermore, introducing the business context offers the advantage that the solution becomes application independent which in turn makes implementation easier and simpler. Within the Daista Approach, implementation issues are broadly grouped into three categories:
Implementation strategy Past experience has shown that the “big bang” approach of developing over-arching strategic solutions rarely succeed because it is difficult to execute in the fast changing environment that is typical in everyday business. The Daista approach focuses on developing localised solutions that fit within an overall framework. That is, rationalising local environments by the integration of “islands of automation” to form “continents of automation”, then the integration of the “continents of automation”. This approach offers maximum flexibility while keeping within the bounds of a predefined overall framework or “big picture”. Implementing a solution that satisfies current requirements Since not all users need access to all information all the time, delivering the objectives would only require the identification of the various stakeholders and addressing their individual needs. In practice, stakeholders may be grouped into two main categories namely “Specialists” and “Generalists”. The “Specialists” spend most of their time using specific commercial-off-the-shelf (COTS) applications such as PLM, ERP, etc. As such, their needs are effectively served by the respective COTS applications they use. The “Generalists” on the other hand, require access to information from more than one business application. Their needs are therefore typically catered for by custom capabilities via customisations to the COTS solutions, development of bespoke applications and integrations between the COTS and/or the bespoke applications. The Daista uIP has been designed to deliver such custom capabilities effectively and flexibly by creating a consistent integrated set of information that supports the custom capabilities while providing single points of integrations to the major COTS applications. Enabling the solution to evolve with changing business needs and priorities In the Daista approach, investment is directed to the creation and maintenance of an information infrastructure and not in the custom applications. This is because custom capabilities are typically required to address short term needs. Therefore new custom capabilities can be added and old ones retired with ease while the information assets continue to exist. Furthermore, building the custom capabilities on the Daista uIP rather than as customisations to the COTS solutions has the added benefit that new releases of the COTS solutions can be implemented speedily and without the cost of updating the customisations. Click here for further information about the Daista approach.
|
|