San Diego

Wolfgang Haas (WH) gave a live demonstration of the exchange of the foundations of a building between RIBCON and Nemetschek Allplan using an AP225 compliant Part 21 file. He reported that translators for AP225 had been developed by a number of European software vendors (and CSTB) and that translators were being developed by two Japanese companies and were under consideration by Bentley Systems. A toolbox was now available to assist vendors with implementation.

He reported that AP225 had completed DIS ballot. 13 countries voted in favour; 1 country (Japan) voted against; and 3 countries (France, Switzerland and USA) abstained. The Japanese vote was a no-vote with comments. These were reviewed by BC and it was agreed that there were of a "minor" character. The issues raised included the provision of reference lines; the incorporation of relationships – such as "penetration" and "touch" – aimed (among other goals) at the reduction of computation in clash detection; and the assignment of predefined functional types to the entities fixture_equipment_element and service_element - similar to the existing types for the entity structure_enclosure_element. In addition there are three minor editorial issues.

AP225 now has its own website at:

http://www.haspar.de/Ap225/index_eng.htm

WH is presently working on ATS development with a view to entering FDIS later this year.

AP230 Building Structural Frame: Steelwork – Mike Ward

Mike Ward (MW) reported that although negotiations with Japan and the USA to raise funding for AP230 had not been successful, the US and Japan had contributed a great deal to AP230 in the form of technical inputs and that some limited money may now be available from UK sources. This money would enable AP230 to begin the interpretation process and this, in turn, should release funding from the BSI for the development of ATSs.

MW also reported that he had added object and assertion definitions (and accompanying diagrams) to the AP230 website:

http://www.leeds.ac.uk/civil/research/cae/step/ap230/ap_doc/ap_c4/ap_c4.htm

MW went on to announce that the documentation for AP230 would include a simplified ARM (probably in IDEF1X notation), an AIM, and a fully attributed ARM in EXPRESS as a separate (informative) annex. This would support ARM level implementations and be aligned with CIS/2.0.

There was a discussion of a number of issues related to the identification of standard products and to the identification of (any) data in a shared data repository. A prototype product model repository for AP230 is currently under development at the University of Leeds. MW also announced that the next version of the AP230 ARM would include resources (for representing materials) derived from part 45; a Japanese "flavour"; and resources (based on technical inputs from Japan) for dealing with seismic loading. It was also decided that Finland and the UK should work on the creation of a Finnish flavour for AP230.

106 Building Construction Core Model – Jeff Wix

Jeff Wix (JW) presented part 106, version T300 (San Diego) which is issued as a WD version document reference WG3/N599.

Changes have resulted from the comments of the STEP secretariat (concerning scope), IFC implementers, and end users; and include: adjustment of scope to accord with the requirements of STEP application resources (that part of 106 which enabled direct implementation of a sharable capability was removed); reduction of depth of type hierarchies in order to ease implementation; and responses to user requirements.

A detailed analysis of part 45 (Materials) was carried out and this part rejected as too unwieldy.

Discussion of part 106 covered following themes:

The notion of bc_characteristics had been reintroduced.
Performance had yet to be added.
The issue of identification was to be discussed with AP230.
"Edition" had been added as an optional attribute of bc_classification.
Should year of publication also be added as a relationship to a date entity?
It was suggested that the "publisher" attribute should have a different name.
The "table" attribute should be optional.
The "description" attribute should be mandatory.
Changes to bc_cost model discussed

JW also reported that all 96 issues raised against 106 had been revisited and their resolution reviewed in the light of revisions to the model since Toronto.

SteelBase-project – Kari Karstila & Raimo Pehrsson

The SteelBase-project is part of the FINNSTEEL R&D programme jointly funded by the state Technology Development Centre (TEKES) and the constructional steelwork industry. Its aim is to promote neutral data exchange using STEP technology, and the CIMsteel Integration Standards (CIS -Release 1.0.) have been adopted as the basis of such data exchange.

In order to promote deployment of the project results in Finland, SteelBase is developing a computer application for viewing, navigating and editing steel structural product models; generating BOMs; and converting between other data exchange formats used in the industry.

The software (StBrowser) is targeted at designers who generate and export CAD data and manufacturers who receive data - and wish to browse it, check it, generate reports and convert it into other formats.

Further development is planned for StBrowser - which currently uses a subset of CIS/1.0 DEP4. It is intended that the software will first provide full support for DEP4 and will later conform with CIS/2.0 and (thus) with AP230 Building structural frame: Steelwork.

Kari Karstila went on to demonstrate a prototype version of StBrowser and to show how it could be employed to browse a file of structural steel frame data.

Classification - Kees Woestenenk

Kees Woestenenk presented a proposal he had advanced in his recent publication (An Information Framework for the Construction Industry). This endeavoured to bridge the gap between "traditional" Classification and current STEP AEC developments. It proposed the establishment of a taxonomy of Object types and a set of standardised characteristics (properties and performances).

KW also presented a Resolution sent from the TC59/SC13 meeting being held at the same time in London, (3 June 1997). The resolution read: "ISO TC59/SC13/WG2 should commence a project titled: An investigation of the role of objects and object orientation in the organization of information about construction works; all interested organizations and individuals be encouraged to participate via the Internet; and the project leader be Terry Wright (Australia)".

KW also mentioned the discussion of this topic at the ICIS conference (ICIS: International Construction Information Society) which took place on 26-28 May in Norway. Members of ICIS expressed a wish to participate in the international working group mentioned in the ISO TC59/SC13/WG2 resolution.

Road & Traffic related information in STEP – Ilse van Koetsveld

Ilse van Koetsveld (IK) (on behalf of Martin van Koetsveld) announced that CIM Architects now had a contract from the Dutch government to organize meetings of a Dutch GIS/CAD group devoted to coordinating the activities of industry, the government and the STEP community in the Netherlands. The meetings will include participants from Germany and Sweden and will be held later in 1997. Interested parties should contact IK

An announcement of - and a call for participants in - a First International Workshop on Modelling of Road and Traffic Related Information was made on behalf of the Swedish National Road Administration. The workshop will be from 1997-10-07 to 1997-10-09 in Borlänge and abstracts should be submitted by 1997-06-29 to Berndt Wenzel of EuroSTEP: BG_wenzel@compuserve.com.

Both these initiatives will explore the possibility of future standardization activities for roads in STEP.

RoadRobot - Ricardo Gonçalves

Later, Ricardo Gonçalves gave a presentation devoted to the use of STEP technology in road construction industry. This work took place between 1992 and 1996 as part of the European ESPRIT III RoadRobot project (Operator Assisted Mobile Road Robot for Heavy Duty Civil Engineering Applications) and involved 6 partners from 5 European countries (Portugal - UNINOVA, Spain - Ikerlan, Germany - APS & Vogele, Netherlands - Face, and UK - UK Robotics). The RoadRobot project drew on the STEP experience gained during the European Brite-Euram CIMTOFI project (1990-1994).

The RoadRobot project employed part 21 files in the communication of instructions to automated road building systems and in general data exchange as part of an approach to the integration and (semi) automation of project planning, scheduling, excavation and paving.

The architecture of the prototype system which was developed under this project enabled a construction company to centrally control civil engineering tasks being carried out in a number of different places (cells). The platform used for the RoadRobot project was an adapted and improved version of a generic STEP based Integration Platform, called "SIP".

More information on the RoadRobot project from:

http://www.uninova.pt/CRI/GR_SSNC/RESEARCH/roadrobot.html

General Case Parametrics - Richard Wittenoom

Richard Wittenoom (RW) gave an update on General Case Parametrics - a work area originally identified in 1995/96 by John Mitchell, Howard Leslie and RW. The SC4 Parametrics group was now project ISO 14959 in WG12

RW proposed a reference model for application process - inspired by Wim Gielingh's GARM initiative (1987). This had specified the notions of functional units, functional requirements, and technical solutions - thus providing a framework for the description and realisation of project data during the design process and subsequent life-cycle stages. Such data did not only relate to geometric dimensions of building elements, but to a broad range of structural, thermal, fluid, and other characteristics which could be specified parametrically - hence "General Case" Parametrics.

The first stage of a project life cycle was the creation of a project brief which specified requirements in terms of criteria which might be satisfied by a range of design solutions. ISO 14959 would provide a framework for the encapsulation of solution processes into objects which could be instanced, exchanged or stored in libraries; and would support the modelling and exchange of information at the functional specification level, and the processes of product realization.

Ram Sriram and Robert Allen (from NIST) would be participating in this work and - whereas RW would address the vertical aspects of realization between stages of design model evolution - they would focus on horizontal communications among project team members.

A white paper would be on SOLIS by 1997-06-30 and an updated paper would be discussed at Florence. A first working draft of a Technical Report type 2 (aimed at industry) would be presented at Orlando (1998-02). An Australian project to include GCPs in release 4 of the IAI's IFCs was under consideration; and the funSTEP and CONCUR projects were other likely collaborators. It was agreed that the B&C group wished to collaborate with and be kept informed by the GCP group, and interest and inputs from other groups would be welcomed.

Further information from:

http://www.wt.com.au/~ausstep/gc_parametrics/

and from:

http://www.nist.gov/sc4/www/paradocs.htm

Collaborative Design - Ram Sriram

Ram Sriram introduced proposals for a data model which could capture information relating to collaborative design. Such information included communication levels, negotiation/conflict resolution, and design rationale; and was the sort of information which it would be appropriate to pass between designers and manufacturers.

It was proposed that the BC group should continue to keep itself informed of developments in this area.

FunSTEP - Miguel Borrás

Miguel Borrás (MB) reported that the objective of the FunSTEP project (partially based in the CIMTOFI project results) was the development of common data models and exchange mechanisms for furniture, and the development of commercial translator software.

The beneficiaries of the project would be manufacturers, designers and retailers, and these would benefit from both internal and external integration.

MB noted that the furniture industry, like the construction industry, was fragmented. Its finished products, in direct contrast to those produced by - say - the aeronautics industry, were manufactured in very large ranges from very small number of basic component types.

In order to meet the challenge posed by the creation of a product data standard in this environment, the project had obtained ESPRIT funding, contributions from a spectrum of European furniture manufacturers and retailers, and commitment to implement from a range of software vendors.

The data model under development was parametric in character - reflecting the need to support product behaviour requirements in the exchange of customers' projects. There was a need for graphics in the furniture sector and a need for end user control of product libraries in the manufacturing sector.

FunSTEP was also committed to the development of compliance testing and certification methods for commercial CAD systems and, in pursuit of a common model for data exchange, was considering inputs from the IAI and from AP225.

Further information from:

http://cupido.uninova.pt/funstep/

CSTB - Patrice Poyet & Jean-Luc Monceyron

Patrice Poyet (PP) presented work done by Jean-Luc Monceyron on a SDAI C++ late binding Toolbox by CSTB. This provided a persistent storage capability with transactional services. As an example of the capability of the toolbox, there was an Interoperability Demonstration between two CAD applications. A model of CSTB offices created using Nemetschek’s Allplan software was imported by the toolbox as an AP225 compliant Part 21 file and output to AutoCAD as IFC 1.0 objects. PP explained the browsing, editing and conversion functionality in between.

International Alliance for Interoperability - Richard See

In view of the impending vote at SC4 on formal liaison between IAI and STEP, international representatives of the IAI were invited to present on the background, objectives, organisation and plans of IAI.

Richard See (RS) first presented a view of the building construction industry and the needs foreseen for greater process integration through data exchange and sharing. There was a growing industry requirement for change arising from increased competition and the need to achieve cost and time savings whilst improving product quality. There were new technologies becoming available such as O-O and web technologies which were relevant to industry and there was a need to provide freedom of choice for users to select "best of class" software from different vendors. New types of application software were foreseen (including simulation, code checking, construction, operations etc.) and there was a perception that industry requires an intelligent, shared AEC project model that is supported by multiple vendors. The new types of application possibilities were cited as examples of the benefits accruing form such a shared project model.

The IAI had started in 1995 as a closed alliance of a few companies in US who had a shared vision for the interoperability of software tools and the interoperability of information within the construction industry. Having proved the concept, it was decided to open the alliance to wider participation, including vendors other than Autodesk and to other countries. Most of the well known vendors and many third party software developers were now supporting IAI. Many industry and client organisations were in membership in the seven IAI international chapters (North America, German speaking, UK, French speaking, Nordic, Singapore and Japan). The IAI were cognisant of the work in STEP but had chosen to establish themselves independently because of the need to drive developments with greater urgency and to provide mechanisms for involving the many small as well as large organisations in the highly fragmented construction industry. Closed IAI really began in late '94. Open Alliance began in September '95. The primary motive for establishing an independent organization (distinct from STEP) was the wish to pursue a pragmatic an incremental approach to problem solving. In was the view of the IAI that it would take several iterations before they were ready to consider formal standardization of its work.

The production of the Industry Foundation Classes (IFCs) would be incremental so that industry could see results and gain benefits at as early a stage as possible. The first release of IFC was in November 1996 and the enhanced release 1.5 was due for release imminently. Thereafter, an annual release cycle of IFC was anticipated The IAI had a Five Year vision, with Implementation Phases leading from Model Exchange (via Part 21 files), to Model Servers (client-server via SDAI) and on to Distributed Interoperability. Release 1.5 for member comments would be in late July '97, and Final Release 1.5 would be in late August '97.

RS described the structure of the IAI organisation. IAI chapters were autonomous with regard to their operation and funding, but the chapters were under the umbrella of an International Council. Each chapter had Domain Committees where industry members developed the requirements which were then passed to an international team of modelling experts to produce the IFCs. IFCs are modelled in EXPRESS.

RS finally presented the "roadmaps’ for release 2.0 and 3.0, showing the priority topics selected by IAI, constrained by the available budgets.

Note: RS is the IAI International Technical Director

IFCs - Thomas Liebich

Thomas Liebich (TL) gave a walkthrough of Release 1.5 represented as an EXPRESS-G model. The IFC Architecture is modular and layered:-

First layer: Independent Resources

Second layer: Core layer: Kernel (non AEC specific), Core Extensions (AEC specific, abstract part)

Third layer: Interoperability layer: commonly used concepts

Fourth layer: Domain Model Layer: May provide a mapping mechanism for the "plugging in" of non IAI compliant domain models - an "adapter" layer.

It was explained that this architecture should facilitiate interoperability within and between AEC domain models and supporting applications. Selected parts of the IFC model - from the IFC Kernel and Product Extension - were displayed and discussed, and the prototype HTML based documentation for IFCs was previewed.

Note: TL is a member of the IAI Spec Task Force

Minutes Graham Storer and Mike Ward (with major inputs from Kari Karstila and Kees Woestenenk).

This page is maintained by M A Ward

Last updated 1997-06-18

To the STEP Home Page at Leeds

To the AP230 Project Home Page

To the Civil Engineering Home Page

Minutes of the STEP WG3 / T12 / AEC / Building Construction Sub-Group Meeting in San Diego 1997-06-01 to 1997-06-06

Contents

Attendance

AP225 Building Elements by Explicit Shape Representation – Wolfgang Haas