metis - Knowledge-based search and query methods for accessing information of semantic models (BIM) to support searching in early design stages.

metis - 基于知识的搜索和查询方法，用于访问语义模型（BIM）信息以支持早期设计阶段的搜索。

• 批准号: 235841221
• 负责人: Professor Dr. Andreas Dengel
• 金额: --
• 依托单位: Lehrstuhl für Architekturinformatik
• 依托单位国家: 德国
• 项目类别: Research Grants
• 财政年份: 2013
• 资助国家: 德国
• 起止时间: 2012-12-31 至 2017-12-31
• 项目状态: 已结题
• 来源: https://gepris.dfg.de/gepris/projekt/235841221?language=en
• 关键词: metis; Knowledge; based; search; query

Developments in IT led to a restructuring of design and planning processes in the construction sector. Today's design and planning systems are increasingly based on Building Information Modeling methodologies. Data is stored and archived in semantically organized building models that serve as a knowledge base. Architectural and technical knowledge are stored as an integral part of such models, however methods for managing the stored knowledge and informing the sketching and conceptual design phase are still lacking. Pre-existing storage and retrieval strategies are of only limited help as they are based on keyword taxonomies and cannot adequately represent spatial configurations.The aim of the research project is to develop research strategies for planning-relevant information as a knowledge basis and reference source in the early design phases based on semantic models (BIM). The intention is to develop methods and prototypical implementations that instead of using text-based research methods employ graphical input metaphors and can recognize formal structures on the one hand as well as looser, sketchy inputs (functional or conceptual sketches) on the other. To exploit the potential of modern digital technologies, the project links artificial intelligence methods (computer science) with design research (architecture).

信息技术的发展导致了建筑部门设计和规划程序的重组。当今的设计和规划系统越来越多地基于建筑信息建模方法。数据以语义组织的建筑模型作为知识库进行存储和存档。建筑和技术知识作为此类模型的组成部分进行存储，但管理存储的知识并为绘制草图和概念设计阶段提供信息的方法仍然缺乏。现有的存储和检索策略基于关键字分类，不能很好地表示空间形态，因此帮助有限.研究项目的目的是在基于语义模型(BIM)的早期设计阶段，为规划相关信息开发研究策略，作为知识基础和参考源.其目的是开发方法和原型实现，而不是使用基于文本的研究方法，使用图形输入隐喻，一方面可以识别形式结构，另一方面可以识别更松散、粗略的输入(功能或概念草图)。为了挖掘现代数字技术的潜力，该项目将人工智能方法(计算机科学)与设计研究(建筑)联系起来。

项目成果

Comparative Evaluation of Rule-Based and Case-Based Retrieval Coordination for Search of Architectural Building Designs

• DOI: 10.1007/978-3-319-47096-2_2
• 发表时间: 2016-10
• 作者: Viktor Ayzenshtadt;C. Langenhan;Johannes Roith;S. S. Bukhari-S.;K. Althoff;F. Petzold;A. Dengel

Supporting the building design process with graph-based methods using centrally coordinated federated databases

使用集中协调的联合数据库通过基于图形的方法支持建筑设计过程

• DOI: 10.1186/s40327-017-0057-y
• 发表时间: 2016
• 期刊: Visualization in Engineering
• 作者: Roith J;Langenhan C;Petzold F.
• 通讯作者: Petzold F.

Migrating the Classical Pen-and-Paper Based Conceptual Sketching of Architecture Plans Towards Computer Tools - Prototype Design and Evaluation

将传统的基于笔和纸的建筑规划概念草图迁移到计算机工具 - 原型设计和评估

• DOI: 10.1007/978-3-319-52159-6_4
• 发表时间: 2015
• 作者: Bukhari S;Dengel A;Bayer J;Langenhan C;Althoff K.-D;Petzold F;Eichenberger-Liwicki M.
• 通讯作者: Eichenberger-Liwicki M.

Thinking With Containers: A Multi-Agent Retrieval Approach for the Case-Based Semantic Search of Architectural Designs

用容器思考：基于案例的建筑设计语义搜索的多代理检索方法

• DOI: 10.5220/0005683801490156
• 发表时间: 2016
• 作者: Ayzenshtadt V;Langenhan C;Bukhari S;Althoff K-D;Petzold F;Dengel
• 通讯作者: Dengel