||Today, there is a growing set of technologies being developed for information exchange in the construction industry. These range from Aspect Models in specific product areas to large scale integrated product models, to new languages such as EXPRESS-X and EXPRESS-2. The purpose of this paper is to sort out and review these various efforts, from several different perspectives: * in terms of what can be used now or in the near future in a production form; * in terms of the significant technical issues and limitations that may require generation changes in exchange technologies;* in terms of external business practices (reflecting case studies), practical benchmarks and adoption criteria, political and other externalities that are affecting these efforts.The survey will review the following issues: * current capabilities of ISO-STEP Part definitions to support information exchange in the building industry; * current efforts by IAI, BCCM in STEP, and other parallel activities and their potential contribution and pitfalls (problems to be overcome);* different current research efforts and the problems and solutions they identify, including COMBINE, EDM-2, VEGA, work at CIFE at Stanford University.Hitherto underdeveloped model aspects, such as capturing the semantics of the client's brief, or capturing design evolution (program, decisions and rationale), modeling performance assessments, and others such as relevant standards, construction site handling, etceteras will be reviewed and priorities assessed.Over the last ten years, the set of requirements that a building product model must meet in order to be accepted in practice as a significant 'productivity enhancement has incrementally expanded. That is, as various research goals have been set, then met, the true extent of the challenge for realizing production-based building product modeling has grown. We will review this expanding set of requirements and attempt to scope their final range. These requirements include, among other aspects: * 'semantic coverage', * level of interoperability across applications, * level of embedded project management control, and* maintained linkages to parallel 'unstructured' information flows, e.g. managed by Engineering Data Management and Document Management software.It will be argued that a viable growth scenario regarding the semantic coverage of building models is likely to be a determining factor in the way that CAD vendors will embrace these as the basis for developing the next generation of architectural CAD software.Priorities of development will be identified and compared with perceived market pulls. The perspective taken will emphasize the US point of view. However, we will endeavor to also weight significantly the European situation and efforts. The result of these perspectives will be to identify 2-3 scenarios of future evolution in the area of building product modeling, with an assessment of their likelihood of coming to be, and the critical issues needed to accomplish them.