An ontology-aided optimization approach to eco-efficient building design
Despite the advances in development of building performance simulation tools, their potential for performance-guided design has not been sufficiently exploited within the design community. This is in part due to the complexity of data accumulation for simulation purposes and the elaborate data entry modalities of available tools. The web undoubtedly contains extensive supplies of potentially useful information. However, the extraction and application of this data is hampered by lack of sufficient structure in the encapsulation and presentation of the information. The present paper reports on the progress of the project SEMERGY in exploring this web-based potential using semantic web technology to extract and restructure building product information into an ontology integrated within a performance based design optimization environment.
From F Shayeganfar, A. Anjomshoaa, J. Heurix, C. Sustr, N. Ghiassi, U. Pont, S. Fenz, T. Neubauer, A. Tjoa, A. Mahdavi: „An ontology-aided Optimization Approach to Eco-Efficient Building Design„; in: „Building Simulation 2013 – 13th International Conference of the International Building Performance Simulation Association.„, IBPSA (ed.); IBPSA, (2013), ISBN: 978-2-7466-6294-0; 2193 – 2199.
„Over the past few decades, many efforts have been made to tune the building design process for production of energy efficient buildings. Development of various building performance assessment tools ranging from simple normative calculation engines to highly complex performance simulation software is a major outcome of these efforts. However, the uptake of such tools in the building design process remains very limited. Traditionally, rules of thumb and simplified calculations have been used to guide thermal performance considerations during the early design stages of buildings. It is only after the design has been finalized that external energy analysts have been involved to analyze the final design solution (Hensen et al. 2004). Simulation tools are not used to support the generation of design alternatives, or to make informed choices between different design options, or to optimize building and/or systems (De Wilde 2004). Contrarily to this general tendency, studies show that decisions taken during conceptual design have a disproportionate impact on the final building performance, relative to time and effort consumed (Domeschek et al. 1994) Moreover, the cost of implementing changes during the early stages is substantially lower than in later phases of the design process (McGraw-Hill 2007). Building performance simulation is rarely used at all for supporting early design phase tasks such as feasibility studies and conceptual design evaluations (De Wilde 2004).“
„Papamichael and Pal (Papamichael et al. 2002) cite the main barriers to the development and use of BPS tools to be the low market interest and high time-cost of applying them. This is in part due to the complexity of data accumulation for simulation purposes, the elaborate data entry modalities and usability issues of available tools. The web undoubtedly contains extensive supplies of potentially useful information which can facilitate the data accumulation and entry. However, the extraction and application of this data is hampered by lack of sufficient structure in the encapsulation and presentation of the information (Mahdavi et al. 2012a). The SEMERGY project (Mahdavi et al. 2012a, Mahdavi et al. 2012b) intends to bridge this gap by providing semantic links between real world products and building model’s abstract concepts and elements. The gap between required and available AEC data sets is hypothesized to be bridgeable based on two main pillars: First, a set of compact and versatile ontologies should be created that serve as a shared standard vocabulary of AEC concepts. The real world products can be then described via these vocabularies in a uniform machine-processable format. Secondly, the building performance assessment and optimization environments should be linked to these product descriptions for exploration of design alternatives and evaluation of energy efficiency.“
„The key contribution of the research is the exploration and demonstration of the semantic web technologies toward populating the input data for building performance simulation models via the navigation of the extensive but currently ill- structured web-based information space pertaining to building materials, elements, components, and systems, as well as resources and documents concerning procedural, climatic, and financial (e.g., public funding) information that could be of value to designers and decision makers. In order to clarify this process we will first explore the major use-case of SEMERGY project. As mentioned before this project concentrates on design optimization in view of multiple criteria of investment and operation costs, energy performance and environmental impact. For this purpose, the initial design is conveyed by the user via a web-based user interface in the form of a building data model containing information such as geometry, building components and their properties, along with additional background information concerning available budget and/or desirable or intended performance objectives. In a second step, SEMERGY considers a number of semantic (non-geometric) permutations of this initial information to create valid construction alternatives for different building components. The underlying reasoning interface responsible for the identification of these alternatives holds a set of rules derived from an analysis of common building constructions based on different product properties. Product libraries play an important role for providing appropriate products that match the user/design requirements. For this purpose, SEMERGY system employes semantic web-technologies in order to capture and formulate the required information in a dynamic and machine- processable format. Finally, a comprehensive evaluation process will be executed. Thereby, both simplified calculation routines (e.g., those necessary to generate energy certificates for building projects or perform life-cycle analyses) and numeric simulation applications (e.g., thermal performance simulation tools) could be deployed (Mahdavi et al. 2012a).“
Background: „Nowadays, the BIM methods are being used intensively in different processes of building design, construction, operation, and maintenance. In this context, BIM uses a schema to define the main concepts and a number of local names and descriptions. In order to achieve the interoperability at global level, the schema concepts should be separated from local names and descriptions. This separation could be realized via dedicated dictionaries of construction terminologies that cross the regional and language borders and uniformly define the meaning of terms such as product names, properties, etc. The International Framework for Dictionaries (IFD) is aiming to establish such a standard library, where concepts and terms are semantically described and given a unique identification number (Bell et al. 2006). More explicitly the IFD is an ISO standard (ISO 12006-3) that is described using an EXPRESS model with a short explanation of its purpose and use. IFD libraries are more than a simple mapping of words among various languages and provide an abstraction layer that contains the conceptualization of entities. This abstraction layer will then facilitate connecting the entities via shared library concepts. For instance the word “Tür” in German is mapped to the same library concept as the English word “door”. So manufacturers may introduce their products in the international market without being hampered due to language issues. The IFD library will handle the representation of products in other languages by aligning the product specifications to the IFC reference model. Moreover an entity might be interpreted differently in different countries and again an IFD can address this issue by providing an abstract reference library. An important role of IFD library is the separation of a concept from its local names and descriptions that define that concept. In IFD this is achieved by separating the concepts from the names and descriptions that are used to name and describe it. As a matter of fact, the IFD library let the concept be both described by multiple name and descriptions and also its relation to other concepts.“
„Several countries have started building dictionaries based of IFD. The most important libraries for building smart are BARBi (Bell et al. 2004) and LexiCon (Woestenenk 2002), which are defined for a better communication between construction partners and for better information handling by computers. Some research works have tried to harmonize IFC with IFD structure (Jansen et al. 2003). By using globally Unique ID (GUID), all the information in the IFC format can be tagged and the concepts may be defined in any language and can be processed by computers. It means that these GUID are used by machines to process the data and textual descriptions by humans. In 2009, the IFD library group has become a part of buildingSMART International (bSI) and since then, the integration of IFD Library into bSI has progressed with plans underway to transfer the IFD Library Intellectual Property to bSI. Later on in September 2010, the IFD Library was renamed to buildingSMART Data Dictionary to fit in with the renaming of the IFC standard to buildingSMART Data Model, and the IDM standard to buildingSMART Processes (NBIMS 2012). Several other research projects have been done over the years on IFC model in order to develop modeling and implementation of AEC objects and to provide more integrated, interoperable and intelligent AEC objects (Halfawy et al. 2002). Some of them have tried to provide online product libraries for AEC industries based on IFC and present the architecture for implementation with aim to support industry practices in the production and consumption of product information (Owolabi et al. 2003). Another trend in product libraries follows the Semantic Web and ontology principles to define the required concepts and their properties. Having a semantically enriched, searchable library of building products can support designers to efficiently select those products that best match design constraints and criteria pertaining to specific projects (Shayeganfar et al. 2008, Beetz et al. 2009, Issa 2012). The SkyDreamer project (Shayeganfar 2009) provided a prototypical combination of BIM and semantic technologies for the selection of energy- efficient products (skylight components). This research provided a proof of concept in view of the potential of elaborate semantic technologies toward bridging the knowledge gap between manufacturers‘ data, building information models, and simulation web services.“