Engineering analysis can be currently characterised as a collection of specific tasks, the solution methods of which vary depending on the available analysis tools. End users are, however, primarily interested in getting their problems solved, without needing to reformulate the data for each type of tool. Considerable time, effort and cost is involved in passing date between the activities in the design process.
An ideal solution to the end user’s requirement would be a generic description of the analysis task which can be mapped, automatically by a converter, to the appropriate analysis tool. In this way a flexible generic representation would be created which would allow reuse of results from one analysis as an input to any subsequent analyses, even if this analysis requires different discretisations or formulations from the previous stage.
It is this concept which is central to GEM , a Generic Engineering analysis Model,which encompasses, in a more generic manner, a greater range of analysis problems than any existing methods, including the current parts of STEP.
Increasingly engineers are using a greater number of analysis tools throughout the product design process. Furthermore there is more emphasis on design optimisation. The direct use of CAD data in the analysis environment is also required. However these combined factors inevitably create problems in the re-use, sharing and, in the longer term, archiving the date and results at the various analyses stages. The aim of the GEM model was to enable engineering analysis methods such as Finite Element Analysis or Computer Aided Design to be used more effectively within the design process and throughout a product’s lifetime.
By addressing these problems GEM will allow further reductions in product-lead-times and the flexibility to choose best-in-class solutions without losing integration benefits. To do this GEM represents properties and results independent of the analysis method or discretisation used, in such a way that they are associated with the underlying geometry, or product component.
To ensure that GEM was sufficiently generic, a survey of end users industrial problems in the types of engineering analysis which was supported was commissioned. Careful consideration was also given to the need to interface with CAD generated data. As a result GEM is capable of supporting the following types of analysis and solution techniques.
Thermodynamics and Heat Transfer
Transmission Line Ray Tracing
GEM will allow faster re-use of results and significantly increase the number of “what-if” studies leading to an optimised product deign.
Currently the opportunities for industrial organisations to implement GEM are limited to the software packages offered by the consortium. As support for GEM gains further momentum this list will grow. A future development being investigated is the feasibility of developing a toolkit which would allow industrial users to build a GEM Database on top of any system which supported STEP.
The GEM consortium has, however, been very active in promoting the GEM model, both through the international standards process and directly, to software vendors and to a number of large industrial organisations.
All GEM modelling is based on ISO 10303, known as STEP, the international standard for product modelling data. GEM uses the same methodology as STEP, in particular the data modelling language Express, standardised as ISO 10303-11, and the representation of data on a physical file, standardised as ISO 10303-21, but GEM is a data model, whose application is far more generic than that of STEP. Work is at an advanced stage in getting GEM recognised as an international standard.
References to a set of three complementary documents (ref. 1, 2, 3) which provide greater detail for those wishing to investigate implementation of GEM in their systems, are given at the end of this document.
|Date||1st January 1997|
|Order Ref||R0047 Book|
|Member Price||£25.00 | $31.04 | €27.68|
|Non-member Price||£200.00 | $248.29 | €221.39|