Energy and Buildings 36 (2004) 737–748 An interoperability workbench for design analysis integration Godfried Augenbroe a,* , Pieter de Wilde b , Hyeun Jun Moon a , Ali Malkawi c a Doctoral Program, Georgia Institute of Technology, College of Architecture, Atlanta, GA 30332, USA b Sustainable Energy and Buildings, TNO Building and Construction Research, Delft, 2600AA, The Netherlands c Department of Architecture, School of Design, University of Pennsylvania, Philadelphia, PA 19104-6311, USA Abstract Past design analysis integration efforts have addressed the need to achieve seamless data integration between design and analysis software applications. These efforts have been dominated by an interoperability focus. Recently, it has been recognized that these solutions suffer from several limitations. For one, they assume a ‘perfect world’ in which all information is structured and all mappings between design and analysis representations exist on a generic level. Secondly, most interoperability solutions have a data focus, whereas true design analysis integration requires a “language” to express both analysis requests and the answers that are generated by experts responding to these requests. An intrinsic part of this language is the logic of the design analysis process. Hence, the integration effort requires a strong process rather than data focus. The Design Analysis Integration (DAI)-Initiative aims to steer towards new solutions for design analysis integration that may overcome the limitations of current data-centric interoperability approaches. This paper reports on the first phase of the development, which has produced a first-generation ‘workbench’ prototype for managing a process driven design analysis dialogue. The workbench is meant to enable a more robust use of existing building models such as IFC for the mapping to simulation tools. The paper presents the underlying theories, prototype development, and findings from the research and concludes with a discussion of future work, targeting extension and benchmarking of the current prototype. © 2004 Published by Elsevier B.V. Keywords: Building design; Performance analysis; Building model; Interoperability 1. Introduction This article deals with the integration of building perfor- mance analysis tools and the building design process. While a large number of analysis tools are available (for instance in the field of building energy tools the US Department of En- ergy tool directory [1] currently lists more than 270 tools), the uptake of these tools in building design practice does not live up to expectations [2–6]. Therefore, many past research and development efforts have dealt with this integration (e.g. [7–19]); yet existing solutions still suffer from several limi- tations (see e.g. [6,20–23]). Most of the focus in past efforts has been on the data aspects of design analysis interaction. The main preoc- cupation has been with enabling seamless data exchange between CAD applications and analysis applications. These efforts aim to achieve some level of interoperability between design and analysis applications based on common shared representations that serve as a neutral exchange format [24]. * Corresponding author. Tel.: +1-404-894-1686; fax: +1-404-894-1629. E-mail address: fried.augenbroe@arch.gatech.edu (G. Augenbroe). The Design Analysis Interface (DAI)-Initiative [25,26] was launched to develop credible solutions for the integra- tion of building performance analysis tools in the building design process. These solutions are driven by the apparent need to enable a more effective and efficient use of exist- ing and emerging building performance analysis tools by collaborating building engineering teams. The longer term objectives are better functional embedding of performance analysis tools in the design process, increased quality control for building analysis efforts, and exploitation of the oppor- tunities provided by the Internet. The latter refers to the pos- sibilities for collaboration in loosely coupled teams where the execution of specific building performance analysis tasks is delegated to (remote) domain experts). It is obvious that in such teams process coordination is the critical factor with interoperability as a support act rather than the main objective. The DAI-Initiative starts from the premise that available solutions for integration based on building product modeling and standardization efforts alone will not be able to meet the longer term objective for a number of reasons [20]. • Current product models and standards are focused on data exchange; they do not take process context into account 0378-7788/$ – see front matter © 2004 Published by Elsevier B.V. doi:10.1016/j.enbuild.2004.01.049