@inproceedings{HoengEderSchmailzletal., author = {Hoeng, Simon Konrad and Eder, Friedrich and Schmailzl, Marc and Obergrießer, Mathias}, title = {Exploring the Potential of BIM Models for Deriving Synthetic Training Data for Machine Learning Applications, Montreal}, series = {Advances in Information Technology in Civil and Building Engineering, Proceedings of ICCCBE 2024, Volume 2, Simulation and Automation}, booktitle = {Advances in Information Technology in Civil and Building Engineering, Proceedings of ICCCBE 2024, Volume 2, Simulation and Automation}, publisher = {Springer Nature}, address = {Cham}, isbn = {9783031873638}, issn = {2366-2557}, doi = {10.1007/978-3-031-87364-5_5}, pages = {54 -- 63}, abstract = {To increase the efficiency and quality of design and construction tasks, the use of Artificial Intelligence (AI) and Machine Learning (ML) offers a way to automate both repetitive and complex tasks. Many of these ML models rely heavily on large amounts of suitable, machine-readable, and labeled training data. Therefore, a variety of conceivable use cases for ML in the Architecture, Engineering and Construction (AEC) industry are difficult to implement due to a lack of freely and directly usable training data. The process of manually structuring and labeling existing data is time-consuming and needs in some cases skilled personnel to ensure the quality of the labeled data. Due to these factors, approaches for utilizing artificially generated data, referred to as synthetic data, are becoming more prevalent. Since Building Information Models contain a large amount of information, deriving training data from these models presents an obvious route for generation of this data. There are many ML applications whose implementation is inhibited due to a lack of training data, for which model-based synthetic data offer a possible solution approach. The Industry Foundation Classes (IFC) standard provides a powerful exchange format for models independently of their authoring software. Parametric and generative approaches to model creation enable the generation of numerous different building models within a short period of time and with low effort. This paper presents a workflow for automated derivation of synthetic training data from rule-based or parametrically generated models combined with existing IFC datasets as a multimodal data repository. The method is validated by testing automated synthetically labeled image data for a plan detection task, which is carried out with the Object Detection Framework YOLOv8. The suggested workflow has the potential to enhance data accessibility, thereby contributing to the implementation of ML applications in the AEC industry.}, language = {en} } @inproceedings{EderHoengSchmailzletal., author = {Eder, Friedrich and Hoeng, Simon Konrad and Schmailzl, Marc and Linner, Thomas and Obergrießer, Mathias}, title = {Towards improving data interoperability for the reconstruction of existing buildings}, series = {The 20th conference of the International Society for Computing in Civil and Building Engineering (ICCCBE 2024), August 25 to 28, 2024, Montreal}, booktitle = {The 20th conference of the International Society for Computing in Civil and Building Engineering (ICCCBE 2024), August 25 to 28, 2024, Montreal}, abstract = {Digital representations of buildings are the supporting structures of various use-cases in the emerging field of data-driven decision making. From large scale applications in the context of city planning to the detailed evaluation of critical infrastructure they enable specialists to observe problems, interpret relationships, test solutions virtually and apply them in the real world. This is only feasible if the individual underlying digital model meets the requirements imposed by the analysis at hand. In practice, especially models of existing buildings are not easy to come by as the information describing the existing structure is often scattered across multiple different data sources in various formats. Previous research efforts have outlined methodologies which leverage machine learning, computer vision and subsequent semantic enrichment in order to achieve the (re)construction of such building models. However, these methods are generally not integrated with each other, nor do they consider being able to interface with a shared repository of building related data. In this paper we present a methodology which focuses on establishing a common context for all building related data by utilizing the Industry Foundation Classes (IFC) schema. In particular we focus on utilizing readily available geometric and semantic data originating from geographic information systems as a basis, subsequently referencing additional data sources in their corresponding context and finally outlining interfaces with downstream enrichment processes in both directions. Through incorporating contextualized (IFC) data into the early stages of the remodeling workflow, we outline an end-to-end process from the initial component-based data-acquisition to the as-built building information model. In establishing a standardized foundation for data exchange and collaboration it enables all stakeholders to work more seamlessly across different stages of the remodeling project.}, language = {en} } @inproceedings{SchmailzlSaffertKaramaraetal., author = {Schmailzl, Marc and Saffert, Anne-Sophie and Karamara, Merve and Linner, Thomas and Eder, Friedrich and Hoeng, Simon Konrad and Obergriesser, Mathias}, title = {Enhancing Decision-Making for Human-Centered Construction Robotics: A Methodological Framework}, series = {Proceedings of the 41st International Symposium on Automation and Robotics in Construction (ISARC), Lille, France}, booktitle = {Proceedings of the 41st International Symposium on Automation and Robotics in Construction (ISARC), Lille, France}, publisher = {International Association for Automation and Robotics in Construction (IAARC)}, isbn = {978-0-6458322-1-1}, issn = {2413-5844}, doi = {10.22260/ISARC2024/0083}, pages = {637 -- 644}, abstract = {While the Architecture, Engineering, and Construction (AEC) industry is increasingly aware of the rising demands for productivity and human-centered construction improvements, the holistic adoption of robotics as a fundamental strategy to address these challenges has not yet reached comprehensive fruition. This paper therefore introduces a methodological framework aiming to address the industry's pressing need for a systematic approach for assessing the feasibility of integrating robotics into human-centered construction processes. It aims to enhance decision-making regarding the degree of automation in human-centered construction processes, ranging from partial to full robotization or non-robotization. The framework is characterized by a more holistic end-to-end data-/workflow and therefore adopts a multifaceted approach, leveraging BIM-based planning methodologies and integrating new technologies [e.g., Motion Capturing (MoCap), work process simulation software incorporating Digital Human Models (DHM), self-developed conversion/interfacing software and more] that have not been widely used in the industry to date. Subsequently, the framework is evaluated in a real-life bricklaying construction process to ensure a more application-based approach. Overall, the framework advances current construction processes with a more inclusive and conscious technology infill to empower construction professionals with the workflow and corresponding tools necessary for the practical integration of robotics into human-centered construction processes.}, language = {en} } @inproceedings{SchmailzlSaffertKaramaraetal., author = {Schmailzl, Marc and Saffert, Anne-Sophie and Karamara, Merve and Linner, Thomas and Eder, Friedrich and Hoeng, Simon Konrad and Obergrießer, Mathias}, title = {Enhancing Decision-Making for Human-Centered Construction Robotics: A Methodological Framework}, series = {Proceedings of the 41st International Symposium on Automation and Robotics in Construction (ISARC 2024), 2024, Lille, France}, booktitle = {Proceedings of the 41st International Symposium on Automation and Robotics in Construction (ISARC 2024), 2024, Lille, France}, publisher = {IAARC}, isbn = {978-0-6458322-1-1}, doi = {10.22260/ISARC2024/0083}, pages = {637 -- 644}, abstract = {While the Architecture, Engineering, and Construction (AEC) industry is increasingly aware of the rising demands for productivity and human-centered construction improvements, the holistic adoption of robotics as a fundamental strategy to address these challenges has not yet reached comprehensive fruition. This paper therefore introduces a methodological framework aiming to address the industry's pressing need for a systematic approach for assessing the feasibility of integrating robotics into human-centered construction processes. It aims to enhance decision-making regarding the degree of automation in human-centered construction processes, ranging from partial to full robotization or non-robotization. The framework is characterized by a more holistic end-to-end data-/workflow and therefore adopts a multifaceted approach, leveraging BIM-based planning methodologies and integrating new technologies [e.g., Motion Capturing (MoCap), work process simulation software incorporating Digital Human Models (DHM), self-developed conversion/interfacing software and more] that have not been widely used in the industry to date. Subsequently, the framework is evaluated in a real-life bricklaying construction process to ensure a more application-based approach. Overall, the framework advances current construction processes with a more inclusive and conscious technology infill to empower construction professionals with the workflow and corresponding tools necessary for the practical integration of robotics into human-centered construction processes.}, language = {en} }