Joint Workshop for the Use of Models that Define the Data and Processes for Information Systems

TC184 SC5 WG1 and CEN TC310 WG1 Presentation, David Shorter.

I am David Shorter, from IT Focus in the UK. I'm an independent consultant and have been asked to talk about ISO TC184 SC5 WG1. Having talked to its chairman, Jim Nell, I'm also going to cover briefly some of the parallel European work, in particular that of a CEN WG which I convene.

The ISO WG1 work was originally concerned with the development of a reference model and methodology that would allow the identification of areas where shop-floor production standards were required. This was done in two stages and completed in 1990/1991. The first resulted in TR 10314 Part 1, which proposed the model itself, in terms of a number of levels of responsibility, to each of which a Generic Activity Model could be applied. The second stage used the model to identify areas of standards and was published as Part 2. While the work made some contribution to the thinking of ISO TC184 it is difficult to point to any specific new work item (NWI) that was proposed as a direct result. Both parts of the TR were recently reviewed by the ISO WG1 and felt to have some continuing relevance although not enough to warrant the revision of the material.

In 1992, the ISO WG1 changed its focus and started to work on how enterprise models might be applied to support enterprise integration in manufacturing. The specific standards-making activity was targeted at producing a Framework for Enterprise Integration, which would set out the needed concepts and describe how they could be applied. While the main customer for the work was still seen as the standards-makers themselves, there was considerable feeling within the group that the standard should also be directly useful and relevant to industry, in particular by showing how enterprise modelling could be supported by systems theory.

In June 1994, after internal debate and some compromise, the ISO WG1 produced a CD for ballot. This attracted considerable criticism, mainly on the grounds that it was a reworking of previous system material which did not adequate reflect the substantive inputs that had been made and that it had insufficient normative content.

A new structure was defined for the CD, new contributions gathered and the ISO WG1 has now produced a substantially revised version which is to be balloted by ISO in Q3 1996 as CD 14258 "Industrial automation systems - Concepts and rules for enterprise models". A major shift in emphasis has been the shift from "Concepts" to "Concepts and rules" - the intention being to constrain the use of concepts in the standards and implementations that use them.

The main elements of CD are:

From a data modelling perspective, the CD makes several normative statements about what data modelling and representation functionality is required but no statements about how that is to provided. It does not describe what are to be the components of enterprise models, nor how these are to be represented. So for this workshop, the CD could be seen as a high level statement of data modelling requirements in a particular modelling domain.

ISO TC184 SC5 WG1 has also taken up a NWI proposal resulting from the work of the IFAC/IFIP Task Force on Enterprise Integration, which has developed a Generic Reference Architecture for Manufacturing (GERAM) under Ted Williams' co-ordination. The NWI is concerned with what is to be considered in an enterprise architecture, and is intended to provide a methodology and checklists to help a company in developing and maintaining its own modelling architecture.

Further information about the ISO WG1 is at http://elib.cme.nist.gov/sc5wg1/

In Europe, standards are developed by committees and working groups under the umbrellas of CEN, CENELEC and "ETSI". I am currently convenor of CEN TC310 WG1, Systems Architecture for CIM (call it CEN WG1 for short). This group is tasked with producing standards for enterprise modelling to meet the needs of European industry, partly by adopting international standards where appropriate, and where necessary developing new standards which we hope will be pre-cursors and contributions to international activity.

In 1990, CEN WG1 produced its Framework for Enterprise Modelling, ENV 40003 which used a three-dimensional framework to describe modelling concepts. The WG then worked on requirements for the environment in which models are to be developed and later executed (published as CEN CR 1831&1832:1995) and is now hoping to take that work forward into another ENV providing the necessary resources can be found.

In 1995 the CEN WG1 produced ENV 12204, Constructs for Enterprise Modelling, which contains definitions and descriptions of the common constructs necessary for computer-based modelling of enterprises, focusing on Discrete Parts Manufacturing. It describes thirteen frequently used modelling concepts and a representation structure which together provide a foundation and a common pro forma for the descriptions of the constructs themselves.

Two issues for CEN WG1 are the need to provide (i) a formal description of the constructs (using a language to be decided) and (ii) guidance on construct-based modelling. Both are under active consideration for future work. Further infomation about CEN TC310 WG1 is at

http://www.demon.co.uk/tvtp/tc310wg1/wghome1.html

David Shorter, 28/9/96


Send message to: david@itfocus.demon.co.uk , (David Shorter), or nell@nist.gov, (Jim Nell) Workshop convener.
Return to: JSW Home Page.