Projects and Theses


Interested students have the opportunity to have a planned project or final thesis supervised by the professorship Intelligent Technical Design in German or English.

You are always welcome to present your project idea. If you have not found a topic yet, we will help to assist you in finding a topic. On this page you can already find suggestions from the project "Digital ITD Twin" or research our teaching and research content.

Topics of the chair

Bachelor's theses, Master's theses or project work that falls into the areas listed below can be supervised by our chair. The exact content of each thesis is defined with regard to the academic assessment. Details on individual topic suggestions can be found on the Moodle. The password for registration is ITD-2017.

  • Building Information Modeling
    • Structuring of building data
    • Project management
    • Workflow development
    • Model-checking
  • Bridge inspection and assessment
    • Modeling damage data
    • Damage visualization
    • Data driven assessment
  • Academic engineer education
    • Virtual learning environments
    • Application of learning theory concepts in connection with AR/VR
    • Development of learning scenarios
  • Digital Twin
    • Digitaler twins of buildings
    • Visualization of building information
    • Smart Home
  • Virtual Reality and Augmented Reality
    • Visualization of information
    • Application in teaching
    • AR/VR supported processes

Completed theses

Identification of a Workflow to integrate Non-Destructive Evaluations results into Building Information Modeling

David Oghenekaro Ovie, Master's Thesis, WiSe 2023/2024

The role of Non-Destructive Evaluations is vital to the serviceability of buildings and infrastructure as they allow for the integrity and viability of such structures to be confirmed without affecting the utilization and lifespan of those structures. However, there exists significant gaps in the integration of this techniques with modern infrastructure design and management techniques such as Building Information Modelling (BIM), hence the need for this study. Key to this issue is the lack of a standard procedure for the chosen Non-Destructive Evaluation technique namely the Pulse-Echo ultrasonic technique. This combined with the subjective nature of the analysis methods implemented for Pulse-Echo testing makes integration challenging. This study has the goal of resolving a part of the integration problem by proposing an incremental workflow based on previous research for the integration of the Pulse-Echo testing results into BIM. Proposed phases from previous research were then simplified into sequential procedures in each phase and then verified by performing each step using a sample Beam. The actors for each phase of this workflow were also identified as well as the constraints and limitations with regards to the automation of the workflow. The verification of the workflow was shown to be successful, although the automation of the workflow given the present techniques and equipment would be challenging as most processes are currently Semi-Automated or manual in nature.

Supervisors & Examiners: Prof. Dr.-Ing. Christian Koch, M.Sc. Mathias Artus.

Digitization of Structural Bridge Information from Forms

Anas Asif, Master's Thesis, WiSe 2023/2024

The digitization of infrastructure, particularly bridges, has gained significant attention in recent years due to its potential to enhance design, construction, maintenance, and performance evaluation. In the realm of civil engineering and infrastructure maintenance, the management and assessment of critical components like bridge tendons have always been of paramount importance. The condition of these vital elements can significantly impact the safety and longevity of bridges, making their precise evaluation essential. One out of several methods is taking drill cores from the structure and analyze them. These results may be available in a tabular format. This thesis outlines a method to digitize printed and handwritten forms into a structured digital format, e.g., relational databases.

Supervisors & Examiners: Prof. Dr.-Ing. Christian Koch, M.Sc. Mathias Artus.

Object Recognition in 2D Plans in the Railway Domain

Khaled Essam Abdelsamie Elahadary, Master's Thesis, SuSe 2023

Engineering drawings include significant knowledge, and unravelling the information they contain requires expertise and time. Developing a learning-based approach that saves time and effort holds the potential to utilize this knowledge effectively in the future. In pursuit of this objective, computer vision techniques have made remarkable progress, driven by the integration of machine learning across various fields. Traditionally, domain-specific engineers were responsible for component detection in engineering drawings. However, the need for efficient extraction of meaningful information within tight timeframes has led to the automation of this process. This automation allows for the seamless integration of different technologies when data is standardized. This paper presents an exploration and implementation of a comprehensive pipeline for extracting railway-specific information from engineering drawings. By employing deep learning approaches, the aim is to successfully extract relevant data.

Supervisors & Examiners: Prof. Dr.-Ing. Christian Koch, M.Sc. Judith Krischler.

Optimizing CT Scans’ Instance Segmentation of Air Pockets in Asphalt Mix Samples via Deep Neural Network

Tehaseen Mujawar, Master's Thesis, WiSe 2022/2023

The assessment and inspection of structures such as paving roads, bridges, parking lots, and driveways are typically performed by a combination of engineers, inspectors, and other professionals. The process involves a comprehensive evaluation of the structure, materials, and condition, including visual inspections, non-destructive testing, and laboratory testing of samples taken from the structure. Some structures use asphalt mix on their deck which can deteriorate over time. This study focuses on the development of a technique to visually recognize air voids in asphalt mixtures using state-of-the-art techniques for instance segmentation of tomographic images. The data considered in this study was obtained from tomographic scans of asphalt mixtures. An automated process was used for data annotation, which reduced the efforts of annotation and did not negatively affect the performance of the model. The annotated data was used to train a deep learning model, optimized using a grid search approach to detect air voids in tomographic images. The evaluation metrics from various trained models were compared against each other to arrive to the optimal combination of hyper-parameters.

This study presents a novel approach for detecting air voids in asphalt mixtures using state-of-the-art techniques for instance segmentation of tomographic images. The proposed technique can automate the detection of air voids in asphalt mixtures, reduce the efforts of manual annotation, and increase the quality of asphalt mixtures produced. The significance of this study lies in the potential for automated quality checks and early detection of asphalt structure deterioration. The results demonstrate that the technique of semantic segmentation with post-processing was feasible for detecting air voids in asphalt mixtures. The trained model was able to differentiate between instances of the same class, which is necessary for detecting air voids. The use of automated quality checks for asphalt mixtures can lead to an increase in the quality of asphalt mixtures produced and early detection of deterioration of asphalt structures.

Supervisors & Examiners: Prof. Dr.-Ing. Christian Koch, M.Sc. Mohamed Said Helmy Alabassy.

Semantic Segmentation of Point Clouds for Open Surface Damages to Concrete Structures via PointNet++

Paul-Christian Prange, Master's Thesis, WiSe 2022/2023

The task of Structural Health Monitoring (SHM) is essential for the maintenance and longevity of infrastructure systems and buildings. The process of assessing buildings is typically resource- intensive, costly, and time-consuming. Therefore, researchers are exploring ways to simplify and support this process.

This study proposes two novel methods for automatically detecting and segmenting spalling on concrete structures in three dimensions, as a step towards the development of a more efficient and automated SHM system. The proposed methods use Machine Learning Algorithm (MLA) to process 2D and 3D data. The 2D MLA processes the image data directly, while the 3D MLA processes point clouds generated from the image data using the Structure from Motion (SfM) algorithm. The performance of both methods is evaluated, and while they yield equivalent results, there are advantages and disadvantages in the training and usage of the models. In particular, the 2D method requires less memory than the 3D method both in terms of the data required and during the training of the machine learning models. Both methods can achieve good results, and there is potential for optimizing the data pipeline further through the addition of new algorithms or modules. Overall, this study offers a promising approach to improving the assessment of building structures and highlights the potential for further research in this area.

Supervisors & Examiners: Prof. Dr.-Ing. Christian Koch, M.Sc. Mohamed Said Helmy Alabassy.

BIM and GIS Integration for Infrastructure Planning

Rogerio Antonio Mota Santos, Master's Thesis, WiSe 2022/2023

Building Information Modeling (BIM) and Geospatial Information System (GIS) concepts and their software have been widely used in different engineering fields, albeit often separately from one another. With the advancement of software tools in the last decade, bidirectional integration between these two different types of information, be it when converted, linked, or merged from one to another, has been quite challenging. This thesis developes an integrated workflow for conversion of a railway alignment from GeoJSON into a BIM model in Industry Foundation Classes (IFC) format. The implementation is mainly undertaken in Python programming language to deliver the resulting IFC model that represents an alignment to be used in the early stage of a railway project in order to build a bridge between both worlds of BIM and GIS.

Supervisors & Examiners: Prof. Dr.-Ing. Christian Koch, M.Sc. Judith Krischler.

Integration of a BIM-based Workflow in Construction Planning for a Quality Gate before the Construction of Modular Buildings

Robin Graf, Master's Thesis, SuSe 2022

This thesis deals with the creation of quality gates prior to the construction of modular buildings. A concept is presented, which milestones have to be set and which tasks have to be completed in the individual phases. From the beginning of a project, various policies and procedures need to be explained that work towards the goal of a clash-free building model during the project. The components that are important to achieve this goal are analysed.

The highest priority is to check the conformity of the model with legal requirements such as standards and guidelines for the construction industry. The conformity assessment is currently tedious and error-prone, as it is usually carried out manually and iteratively by the responsible designer for each modification to the design. Automating this process with digital methods has the potential to reduce the workload and thus the costs, to free the engineer from monotonous, iterative work.

Therefore, one of the most important components is the automatic detection of clashes. This happens at various defined milestones in the project. Single or multiple specialist models are checked for clashes. For the individual checks, certain prerequisites must be fulfilled. The definition of rule sets is necessary for each quality gate. For this reason, the created sets are closely examined for their usefulness and applicability. The goal is that it should be possible to check the models automatically with the rule sets created and that the previously exclusively manual clash detection is no longer necessary.

There are many programmes on the market for this purpose. Three software applications from the leading manufacturers, namely KUBUS, Solibri and Autodesk, were selected and evaluated on the basis of various features. These include user-friendliness, applicability and practicality for the purpose of clash detection at specific milestones in the project. After evaluating specific scenarios, an evaluation matrix is used to determine which software is most suitable.

The clashes found in the selected software are not the final goal, because they must also be communicated to the responsible user. For this reason, one possibility of communication via a Common Data Environment (CDE) is examined in more detail, the workflow is analysed and the most effective approach explained. The identified workflow with all quality gates will be tested and evaluated in detail at the end of the work on a pilot project. The potential and the possibilities for improvement as well as future developments are shown.

Supervisors & Examiners: Prof. Dr.-Ing. Christian Koch, M.Sc. Judith Krischler.

Assembly of Mechanical Components under Robot Vision System

Yamini Konka, Master's Thesis, WiSe 2021/2022

Industry 4.0 is making revolutionary improvements in Robotics and Automation. Research institutions and manufacturing companies are working together to bring intelligent machineries in production lines, construction sites, medical surgeries, and home appliances, etc. Manipulation of objects is a common task performed by robots in manufacturing industries. With the help of the pre-programmed motion plans of a robot, the machines or robots are automatised to perform mundane jobs in factory. In addition to the automatic movements of the robot, vision sensor integration, developing path trajectories and grasping strategies based on extracted features from the gathered images like human cortex processing system while manipulating the objects, provide a way to make today's machines intelligent. Based on aforementioned concept, this thesis work proposes clear description on methods to implement the robot vision system and utilization of image processing and machine learning algorithms to assemble the two mating parts in semi-unstructured environment.

A video with an example use case can be seen at

Supervisors: Prof. Dr.-Ing. Christian Koch, Prof. Dr.-Ing. Frank Schrödel, M.Sc. Mohamed Said Helmy Alabassy.
Examiners: Prof. Dr.-Ing. Christian Koch, Prof. Dr.-Ing. Frank Schrödel.

Integration and Visualization of Material Data in BIM - Inspection Information Modeling with IFC Approach

Jason Lai Poh Hwa, Master's Thesis, SuSe 2021

Building Information Modeling (BIM) is a virtual representation of physical asset. A digital BIM model is intended to cover a facility’s entire life cycle from planning, construction, maintenance and finally destruction. According to the current state of art, BIM editor software tools are available to create BIM as-designed model. Nowadays, the application of BIM as-built model with the point cloud Three-Dimension (3D) model reconstruction is actively researched and developed. While for as-damaged model, Damage Information Modeling (DIM) is yet to be fully defined and finalised in order to be align with existing BIM technology. The data from regular and period inspection activities on the facility structure is obliged to be systematically archived. The massive data amount is a challenge to be reviewed and evaluated. Hence, it is vital for the incorporation of raw material data to a coherent model of building information. Integration of such data into BIM environment requires necessary file parsing and rendering software tool. The goal is to have a tool which allows users to make material condition assessment, defect detection, and quality assurance in BIM environment. The core of the tool is built around a data entry system for documenting material diagnosis information from Non-Destructive Evaluation (NDE) in relation to the respective inspected element in asset’s BIM model. The implementation will create the tie of raw material data with BIM which connects NDT material data to the corresponding measured building element. The measurement information and the association is then visualized either graphically or non-graphically in the view port. A follow-up discussion will be on which representations are suitable for material information when is presented to users. The prototype application allows rendering of different measured material data along with its spatial relationship with BIM geometry in terms of position, orientation, and scale. The inspection information modelling product can be stored and exchanged as as-inspected BIM model while complying Industry Foundation Classes (IFC). Feasibility of concept in modelling material data enriched BIM will be evaluated with the prototype. The relativity of material data representation with BIM model will be further discussed in the thesis.

Supervisors & Examiners: Prof. Dr.-Ing. Christian Koch, M.Sc. Mathias Artus.

BIM-basiertes Facility Management (BIM-to-FM)

Markus Friedrich Boden, Bacherlorarbeit, SoSe 2021

In den Projekten der Drees & Sommer SE geht der Trend in Richtung Building Information Modeling (BIM). Dabei werden die Vorteile der Methodik besonders in der Planungs- und Errichtungsphase der Gebäude genutzt. Bei einem Blick auf die Lebenszykluskosten eines Gebäudes wird jedoch deutlich, dass rund 80% der Gesamtkosten auf die Nutzungsphase fallen. In dem Gebäudebetrieb ist die integrale Nutzung der Daten jedoch noch nicht ver-breitet.Es besteht in diesem Bereich ein beträchtliches Potential, Prozesse zu optimieren und Kosten einzusparen. Durch die Analyse der Vorteile von BIM im Facility Management (FM) soll für Bauherren ein zusätzlicher Anreiz geschaffen werden, Gebäude mit BIM zu planen und zu betreiben.Bei mit BIM geplanten Bauwerken sind die Voraussetzungen, für eine Bewirtschaftungsphase mit einem modellbasierten Computer-Aided-Facility-Management (CAFM), zu erfüllen. Es stellt sich hierbei die Frage, welche Funktionen ein derartiges CAFM-System bieten kann und welche Prozesse für dessen Herstellung von Nöten sind.In der Bachelorarbeit wird dafür die Schnittstelle BIM-To-FM anhand eines Beispielprojektes betrachtet. Mithilfe der erlangten Erfahrungen im Prozess BIM-To-FM sollen Anforderungen an ein BIM-Modell für die Eingliederung in ein BIM-basiertes CAFM festgehalten werden.

Betreut durch: Prof. Dr.-Ing. Christian Koch, M.Eng. Anna Nast.

BIM-basiertes Risikomanagement

Viet Linh Nguyen, Masterarbeit, SoSe 2021

Risikomanagement spielt in der Baubranche eine wichtige Rolle. Gefahren sollen minimiert und Sicherheit sowie Qualität im Lebenszyklus eines Bauwerkes verbessert werden. Risikomanagement ist zumeist ein erfahrungsbasiertes, manuelles Verfahren, welches auf multidisziplinärem Wissen beruht. Die Verwendung von Building Information Modeling (BIM) zur Unterstützung des Risikomanagements von Bauprojekten ist zu einem anwachsenden Forschungstrend geworden. Es wurde jedoch auch festgestellt, dass das auf BIM-basierende Risikomanagement in der Praxis noch nicht so weit verbreitet ist.

Das Ziel dieser Masterarbeit ist die Erarbeitung eines Konzepts für eine Umsetzung von modellbasiertem Risikomanagement. Zu Beginn eines Bauvorhabens werden Kosten, Termine und Qualitäten festgelegt, die es während des gesamten Bauprozesses, bis zur Fertigstellung, einzuhalten gilt. Eine zuverlässige Termin- und Kostenprognose über die gesamte Projektdauer ist es insbesondere interessant für Bauherren. Es wird der Risikomanagementprozess während Planungs- und Bauphasen unter Anwendung der Methode BIM aufgestellt und die Potenziale durch die entstehende Informationsverknüpfung herausgearbeitet. Hier erfolgt eine detailliertere BIM-basierte Betrachtung der Risiken bis auf Bauteilebene im 4D/5D-Modell. Das BIM-basierte Risikomanagement ermöglicht durch die digitale Erfassung von Informationen zu Risiken am Gebäudedatenmodell eine qualifizierte und aktuell gehaltene Risikobetrachtung. Zum Abschluss dient die Arbeit der Validierung und Verifizierung der Ergebnisse in der Praxis, sodass ermöglicht ein vorgeschlagenes Tool die Visualisierung verschiedener Zeit- und Kostenrisiken während Planung- und Bauphasen, indem eine Farbkarte basierend auf den Auswirkungen der Risiken in Bezug auf Bauteilebene des BIM-Modells präsentiert wird. Die verschiedenen Bauaufgaben können hinsichtlich ihres Risikos analysiert werden, indem sie eine Fuzzy-Fehlermöglichkeits-Einfluss-Analyse in einer Schnittstelle anwenden, wo sie eine
Risikofarbe erhalten, die sie nach ihrer Beurteilung kategorisiert. Diese Farben werden dann in ein 3D-BIM-Modell und ein 4D-Modell übertragen, wo sie einfach analysiert, dokumentiert und mit allen relevanten Stakeholdern aufklare und effiziente Weise diskutiert werden können. Danach wird das 4D-Modell mit dieser Informationsverknüpfung in 5D-Modell importiert, um die Visualisierung der Kosten
zu erstellen.

Betreut durch: Prof. Dr.-Ing. Christian Koch, M.Sc. Judith Krischler.

Charakterisierung von Streckenabschnitten der Verkehrsinfrastruktur auf Grundlage von Deskriptoren

Eric Czerwonka, Studienarbeit, WiSe 2020/2021

Die Verkehrsinfrastruktur spielt in Deutschland eine elementare Rolle zur Sicherung des Wohlstands. Die Planung von Straßen (die sich im Bundesbesitz befinden) nimmt dabei eine besonders große Bedeutung ein, da sie beispielsweise überregionale Mobilität ermöglichen und Transportwege schnell miteinander verknüpfen. Eine möglichst effiziente und fehlerfreie Verkehrswegeplanung ist somit unerlässlich.

Diese Masterthesis befasst sich mit dem Ziel Deskriptoren zu identifizieren, um Streckenabschnitte der Verkehrsinfrastruktur bereits in sehr frühen Planungs- und Projektphasen, zum Beispiel während der Bedarfsfeststellung im Rahmen der Bedarfsplanung, charakterisieren zu können. Die methodische Vorgehensweise sieht zunächst einen theoretischen Teil vor, der den Leser:innen den Einstieg in die Bundesverkehrswegeplanung und die Phase der Bedarfsplanung, erleichtern soll. Im Konzeptionellen Teil werden die jeweiligen Deskriptoren identifiziert und beschrieben. In einem Fallbeispiel, das entsprechend der Aufgabenstellung durch den Einsatz der Trassierungssoftware ProVI durchgeführt werden soll, werden die Deskriptoren hinsichtlich ihres praktischen Nutzen evaluiert.

Ein zentrales Ergebnis dieser Arbeit war, dass sich die beschriebenen Deskriptoren grundsätzlich dazu eignen, Streckenabschnitte zu charakterisieren. Eine digitale Umsetzung mit der Trassierungssoftware ProVI war jedoch nicht möglich. Bereits auf dem Markt existierende CAD- und GIS-Infrastrukturprogramme (wie z.B. AutoCAD Map 3D und QGIS) zeigen jedoch, dass eine Umsetzung theoretisch möglich ist. Wird eine Trasse hinsichtlich der identifizierten Deskriptoren beschrieben, besteht das Potenzial, die Bedarfsfeststellung effizienter zu gestalten.

Betreut durch: Prof. Dr.-Ing. Christian Koch, M.Sc. Judith Krischler.

Automated Approach for Building Information Modelling of Crack Damages via Image Segmentation and Image-based 3D Reconstruction

Mohamed Said Helmy Alabassy, Master's Thesis, WiSe 2020/2021

As machine vision-based inspection methods in the field of Structural Health Monitoring (SHM) continue to advance, the need for integrating resulting inspection and maintenance data into a centralised building information model for structures notably grows. Consequently, the modelling of found damages based on those images in a streamlined automated manner becomes increasingly important, not just for saving time and money spent on updating the model to include the latest information gathered through each inspection, but also to easily visualise them, provide all stakeholders involved with a comprehensive digital representation containing all the necessary information to fully understand the structure’s current condition, keep track of any progressing deterioration, estimate the reduced load bearing capacity of the damaged element in the model or simulate the propagation of cracks to make well-informed decisions interactively and facilitate maintenance actions that optimally extend the service life of the structure. Though significant progress has been recently made in information modelling of damages, the current devised methods for the geometrical modelling approach are cumbersome and time consuming to implement in a full-scale model. For crack damages, an approach for a feasible automated image-based modelling is proposed utilising neural networks, classical computer vision and computational geometry techniques with the aim of creating valid shapes to be introduced into the information model, including related semantic properties and attributes from inspection data (e.g., width, depth, length, date, etc.). The creation of such models opens the door for further possible uses ranging from more accurate structural analysis possibilities to simulation of damage propagation in model elements, estimating deterioration rates and allows for better documentation, data sharing, and realistic visualisation of damages in a 3D model.

DOI: 10.25643/bauhaus-universitaet.6416

Supervisors: Prof. Dr.-Ing. Christian Koch, Univ.-Prof. Dr.-Ing. habil. Volker Rodehorst.
Examiners: Prof. Dr.-Ing. Christian Koch, M.Sc. Mathias Artus.

Generatives Design - Multikriterielle Optimierung von Skeletttragwerken auf Basis evolutionärer Algorithmen

Marvin Reisener, Masterarbeit, SoSe 2020

Das interdisziplinäre Zusammenspiel von Planungsbeteiligten, ausführenden Gewerken und Entscheidungsträgern ist für den Erfolg einer Baumaßnahme entscheidend. In konventionellen Planungsprozessen ist die Erstellung, Anpassung und Leistungsfähigkeitsuntersuchung einer einzigen Entwurfsvariante mit einem erheblichen Arbeitsaufwand verbunden. Vor dem Hintergrund der Tragweite von Entscheidungen in frühen Projektphasen mangelt es häufig an alternativen Entwurfsvarianten, Flexibilität sowie objektiven Entscheidungsparametern.

Generatives Design (GD) ist der Prozess, in dem Ziele und Einschränkungen definiert werden, um mit Hilfe von automatischen Berechnungen eine Reihe von Lösungsvarianten zu untersuchen, die den gewünschten Anforderungen entsprechen.

In Bezug zu Building Information Modeling (BIM) und der Digitalisierung der Baubranche im Allgemeinen dienen Generative Prozesse der Beschleunigung früher Entwurfsphasen und erhöhen die Anpassungsflexibilität durch parametrisches Modellieren.

Gegenstand der zugrundeliegenden Arbeit ist es, in Grasshopper ein Skript zu entwickeln, das unter Verwendung von Einschränkungen wie Gebäudespezifika, Lastannahmen und Vorbemessung eine Vielzahlan Lösungsvarianten für ein Tragwerk in Skelettbauweise generiert und anhand gegebener Zieldefinitionen analysiert. Die Ziele können beispielsweise ein minimaler Verbrauch an Ressourcen, eine minimale Anzahl an Tragwerkselementen oder minimale Kosten sein. Die Ergebnisse sollen für eine multikriterielle Entscheidungsanalyse herangezogen und CAD-Modelle entsprechend der BIM-Methodik ausgetauscht werden.

Die Machbarkeit sowie praktische Anwendungsmöglichkeiten generativer Designprozesse in der Baubranchesollen in der vorliegenden Arbeit nachgewiesen werden.

Betreut durch: Prof. Dr.-Ing. Christian Koch, M.Sc. Judith Krischler.

Linking parametric design and structural analysis for the optimization of bridge cross sections

Juan Gonzales Cavero, Master's Thesis, SuSe 2020

In the early steps of the current digital era, software solutions, which are continuously being created and developed, make new strategies possible to significantly affect the efficiency of almost any process. Building Information Modeling (BIM) represents this philosophy for the construction industry. Despite BIM has yield to huge developments regarding vertical building structures for the last decade, the growth for the civil infrastructure domain has been rather scarce. In this thesis the link of parametric design and structural analysis regarding bridges is investigated and different potential workflows are evaluated. In consequence, a bidirectional framework to exchange models between BIM modelers and a structural analysis application is presented. Furthermore, taking advantage of the flexibility parametric design provides an optimizer tool for for bridge cross sections based on genetic algorithms. In order to present the workflow, a real-word railway bridge is used as an example.

Supervisors & Examiners: Prof. Dr.-Ing. Christian Koch, M.Sc. Mathias Artus.

Mixed Reality Interface Prototype for Smart Home Systems

Liselot Ramirez, Master's Thesis, WiSe 2019/2020

Smart Homes are living spaces equipped with lighting, heating, and electronic devices that can be remotely controlled by digital devices. The Chair of Intelligent Technical Design in the Faculty of Civil Engineering has a model project installed in their office. This document reviews the convergence of MR technologies in Building Information Modeling (BIM), and how users benefit from the usability experience during the construction process. Thereafter, usability guidelines for interface design are followed to create a design concept for an interface in Mixed Reality (MR) to interact with the aforementioned Smart Home system. The challenge relies on four user cases: Visualizing Consumption,Comparing Rooms, Automation of Tasks, and Maintenance and Repair. The design proposal is based on the Atomic Design Methodology, in which the components were rendered and exported to a Microsoft HoloLens. The design process for MR is more challenging to implement in the HoloLens than in a mobile application, nevertheless, the result is a design system suited for immersive and mobile applications.

Supervisors & Examiners: Prof. Dr.-Ing. Eva Hornecker, Prof. Dr.-Ing. Christian Koch.

On systematic approaches for interpreted information transfer of inspection data from bridge models to structural analysis

Jan Fröhlich, Master's Thesis, WiSe 2019/2020

In conjunction with the improved methods of monitoring damage and degradation processes, the interest in reliability assessment of reinforced concrete bridges is increasing in recent years. Automated image based inspections of the structural surface provide valuable data to extract quantitative information about deteriorations, such as crack patterns. However, the knowledge gain results from processing this information in a structural context, i.e. relating the damage artifacts to building components. This way, transformation to structural analysis is enabled. This approach sets two further requirements: availability of structural bridge information and a standardized storage for interoperability with subsequent analysis tools. Since the involved large datasets are only efficiently processed in an automated manner, the implementation of the complete workflow from damage and building data to structural analysis is targeted in this work. First, domain concepts are derived from the back-end tasks: structural analysis, damage modeling, and life-cycle assessment. The common interoperability format, the Industry Foundation Class (IFC), and processes in these domains are further assessed. The need for user controlled interpretation steps is identified and the developed prototype thus allows interaction at subsequent model stages. The latter has the advantage that interpretation steps can be individually separated into either a structural analysis or a damage information model or a combination of both. This approach to damage information processing from the perspective of structural analysis is then validated in different case studies.

Supervisors & Examiners: Prof. Dr.-Ing. Guido Morgenthal, Prof. Dr.-Ing. Christian Koch.

BIM-based Framework for Bridge Design and Analysis

Uma Shankar Bharathala, Master's Thesis, WiSe 2019/2020

Enormous engineering work is required to design and construct bridges. Hence, several engineering companies are included in the bridge design workflow from conceptual design to the construction phase. However, many engineering companies are still using traditional design methods to design bridge structures. The initial design phase is crucial in the bridge design phase to avoid problems during the later construction phase. To deliver the project within the required time and budget, massive coordination among the involved teams is necessary. Traditional workflows lack adequate interfaces for such coordination tasks. Building Information Modeling (BIM) is a possible solution to overcome this issue. This study aims to apply a BIM-based framework to design a bridge from the conceptual design phase up to construction plans. The focus is mainly on the bridge design workflow, and on how the traditional methods and BIM methods perform in the stages of conceptual design, preliminary design and main design.

Supervisors & Examiners: Prof. Dr.-Ing. Christian Koch, Univ.-Prof. Dr. Dipl.-Ing. Guido Morgenthal.

Virtual Reality Interfaces for 3D Modeling

Anil Chandra Kavi, Master's Thesis, WiSe 2019/2020

This project focuses on the interfaces for modeling of a classroom in a virtual environment. The Interfaces refers to different mediums through which interactions are possible. Three main interfaces available in the virtual environment are 2D menu, speech recognition, and gesture recognition. In a speech recognition based interface, Microsoft speech application is used, and in gesture recognition based interface, convolutional neural network (CNN) based recognition from 2D drawing to 3D objects. These three interfaces are compared individually and also verified after the integration of three interfaces.

Supervisors & Examiners: Prof. Dr.-Ing. Christian Koch, M.Sc. Mathias Artus.

Technologien der automatisierten Erzeugung und Verarbeitung von Punktwolken zu as-built-Bauwerksinformationsmodellen im Hochbau

Martin Lindner, Bachelorarbeit, SoSe 2019

Im Zuge der Weiterentwicklung der BIM-Methode und einer steigenden Anzahl von Anwendungsfällen, deren Fokus auf der Betriebsphase eines Gebäudes liegt, ist die Modellierung des Gebäudebestands von immer größerer Bedeutung. Der erstrebenswerte Workflow besteht dabei in einem weitestgehend automatischen digitalen Aufmaß und anschließender softwaregestützter Verarbeitung der Daten zu einem "intelligenten" Bauwerksmodell.

Betreut durch: Prof. Dr.-Ing. Christian Koch, M.Sc. Judith Krischler.

BIM-basierte Energiebedarfsberechnung von Gebäuden

Wei Zhao, Bachelorarbeit, SoSe 2019

In der Thesis wird eine Energieanalyse auf Basis eines gegebenen BIM-Modells mittels Autodesk Insight erstellt, mit der Prämisse einen möglichst ressourcenschonenden und -effizienten Einsatz von Energie während der Betriebsphase eines Gebäudes unter Berücksichtigung des Komforts zu erreichen.

Betreut durch: Prof. Dr.-Ing. Christian Koch, M.Sc. Thomas Behnke.

Evaluation of Sensor-Based Registration Accuracy using Head Mounted Augmented Reality Devices

Felix Kremp, Master's Thesis, SuSe 2019

This thesis will propose different use cases for registration of virtual content in real space. The implemented use cases split up in a 2Dmarker, a user placed and a 3D marker based on a BIM model approach. Furthermore, an evaluation of registration accuracy within the context of head mounted augmented reality is realized.

Supervisors & examiners: Prof. Dr.-Ing. Christian Koch, M.Sc. Mathias Artus.

Utilizing BIM-Based Model Checking (BMC) in the early stages of design to examine the vulnerability of masonry structures to earthquakes

Mohamed Matook, Master's Thesis, SuSe 2019

The aim of this thesis is to check the possibility of utilizing the concept of BIM-based model checking (BMC) to automatically check the compliance of masonry structures with the design standards. The rules, implemented using Revit and Dynamo, concerning masonry structures from EN 1996-1-1 and EN 1998-1 are assessed and analyzed. 

Supervisors & Examiners: Prof. Dr.-Ing. Christian Koch, Dr.-Ing. Lars Abrahamczyk.

BIM-based Facility Management using COBie

Yuzhe Wu, Master's Thesis, SoSe 2019

Based on the theory of facilities management and BIM technology, this article first analyzes the research status of building facilities management and BIM thorough literature research method. Then sets forth the disadvantages of traditional facility management workflow and what BIM brings to facility management as well as BIM-COBie based facility management software. At last this article evaluates three pieces of facility software through comparative study method and makes a relevant case study to prove the feasibility and superiority of BIM-COBie based facility management, meanwhile provides references for the use of BIM-COBie according to this.

Supervisors & Examiners: Prof. Dr.-Ing. Christian Koch, M.Sc. Thomas Behnke.

Digitaler ITD Zwilling - Design und Implementierung eines Webinterface für nutzerorientierte Datenvisualisierung und Interraktion mit Smart-Home-Systemen

Xi Chen, Master Thesis, 2019

Diese Masterarbeit behandelt die Interaktion von Internet of Things (IoT), Digital Twin und Smart Home über eine responsive Weboberfläche für einen Heimautomatisierungsserver (FHEM), die sich intuitiv bedienen lässt und auch für Endanwender ohne technischem Verständnis optimiert ist. 

Prüfer: Prof. Dr.-Ing. Christian Koch, M.Sc. Mathias Artus

Entwicklung von Prüfstrukturen zur Qualitäts- und Planungskontrolle von Bauwerksinformationsmodellen

Das Ziel der Arbeit bestand darin, Baubeschreibungen zu analysieren und Konzepte zu erarbeiten, die eine Überprüfung des digitalen Modells bezüglich der Baubeschreibung und der darin festgelegten Qualitäten ermöglichen.

Sebastian Seiß, Master Thesis, 2018

Prüfer: Prof. Dr.-Ing. Hans-Joachim Bargstädt M.Sc.; Prof. Dr.-Ing. Christian Koch; M.Sc. Judith Ponnewitz

Spezifizierung einer Datenschnittstelle zur Integration von digitalen Oberleitungsplanungen in Building Information Modeling

Das vordergründige Ziel der vorliegenden Arbeit bestand darin, Untersuchungen zu Datenübergabeszenarien zwischen Planungssoftware für Oberleitungsanlagen und anderen im BIM-Prozess verwendeten Programmen anzustellen und die Datenschnittstelle mit der besten Eignung anschließend zu spezifizieren.

Justus Boeck, Master Thesis, 2018

Prüfer: Prof. Dr.-Ing. Arnd Stephan (TU Dresden); Dipl.-Ing. Fritjof Aufschläger (TU Dresden); Prof. Dr.-Ing. Christian Koch (Bauhaus-Universität Weimar); Dipl.-Ing. Matthias Krippner (DB Engineering & Consulting GmbH, Leipzig)  

Design and Implementation of Structural Dynamics Problems in a Serious Gaming Environment

Tarek Ahmed Mohamed, Master's Thesis, SuSe 2018

The aim of this project was to develop a serious game to teach engineering students some structural dynamics problems which they can use to interact and get more involved with the theory behind these problems. The chosen problems were calculating the natural frequency and the mode shapes of a single span beam for different cases, and the Dynamic Amplification Factor (DAF). 

Supervisors & Examiners: Prof. Dr.-Ing. Christian Koch, Dr.-Ing. Volkmar Zabel.


Templates for project reports, bachelor thesis and master thesis. These template can be used. However, they are not mandatory.

Templates Word:

Templates Latex: