This thesis is deals with design of a Senior’s Boarding House in Lomnicka u Tisnova. Building four-storey building with one service basement floor with This thesis deals with a design of a retirement home in Lomnicka u Tisnova. This is a four storey building with one service basement floor and with a flat roof. The structural system is of a transverse wall type. Material of structural elements is a combination of reinforced concrete and masonry. The building is founded on piles. In addition, the design includes a restaurant for warming and serving food prepared in different place, laundry and a small pool. All construction drawings and schedules has been developed as building information model (BIM).

<div>Cost estimation is an integral part of any project, and accuracy in the cost estimation process is critical in achieving a successful project. Manually computing cost estimates is mentally draining, difficult to compute, and error-prone. Manual cost estimate computation is a task that requires experience. The use of automated techniques can improve the accuracy of estimates and vastly improve the cost estimation process. Two main gaps in the automation of construction cost estimation are: (1) the lack of interoperability between different software platforms, and (2) the need for manual inputs to complete quantity take-off (QTO) and cost estimation. To address these gaps, this research proposed a new systems to support the computing of cost estimation using Model View Definition (MVD)-based checking, industry foundation classes (IFC) geometric analysis, logic-based reasoning, natural language processing (NLP), and automated 3D image generation to reduce/eliminate the labor-intensive, tedious, manual efforts needed in completing construction cost estimation. In this research, new IFC-based systems were developed: (1) Modeling – an automated IFC-based system for generating 3D information models from 2D PDF plans; (2) QTO - a construction MVD specification for IFC model checking to prepare for cost estimation analysis and a new algorithm development method that computes quantities using the geometric analysis of wooden building objects in an IFC-based building information modeling (BIM) and extracts the material variables needed for cost estimation through item matching based on natural language processing; and (3) Costing – an ontology-based cost model for extracting design information from construction specifications and using the extracted information to retrieve the pricing of the materials for a robust cost information provision.</div><div><br></div><div>These systems developed were tested on different projects. Compared with the industry’s current practices, the developed systems were more robust in the automated processing of drawings, specifications, and IFC models to compute material quantities and generate cost estimates. Experimental results showed that: (1) Modeling - the developed component can be utilized in developing algorithms that can generate 3D models and IFC output files from Portable Document Format (PDF) bridge drawings in a semi-automated fashion. The developed algorithms utilized 3.33% of the time it took using the current state-of-the-art method to generate a 3D model, and the generated models were of comparative quality; (2) QTO – the results obtained using the developed component were consistent with the state-of-the-art commercial software. However, the results generated using the proposed component were more robust about the different BIM authoring tools and workflows used; (3) Extraction – the algorithms developed in the extraction component achieved 99.2% precision and 99.2% recall (i.e., 99.2% F1-measure) for extracted design information instances; 100% precision and 96.5% recall (i.e., 98.2% F1-measure) for extracted materials from the database; and (4) Costing - the developed algorithms in the costing component successfully computed the cost estimates and reduced the need for manual input in matching building components with cost items.</div>

The concept of smart cities points out the future cities, which will incorporate IoT and digitalization for facilitating the communication among people, their devices, government services, and various facilities that can provide enough services for the enormous population in the future cities. To achieve the goal of having the ideal smart cities, it is necessary to go digital and plan for having virtual imagination for every component in the cities, including the construction facilities. BIM method as a means of having a virtual vision of each element of the construction project glows in mind as one sort of assistance to reach this target. This thesis investigates the impact of BIM on the quality of communication in the future smart cities based on the literature review of the three smartest cities including Singapore, London, and Manchester.:Table of Contents List of Figures IV List of Tables V List of Abbreviations VI 1 Introduction 1 1.1 Problem Statement 1 1.2 Aim of the Study 3 1.3 Research Question 4 1.4 Methodology 4 1.5 Structure of Work 5 2 Research Methodology 6 2.1 Introduction 6 2.2 Overview of Common Research Methods 6 2.2.1 Inductive Method 6 2.2.2 Deductive Method 7 2.2.3 Inductive vs. Deductive Method 8 2.2.4 Quantitative Research 8 2.2.5 Qualitative Research 9 2.2.6 Tools for Data Collection 10 2.3 Research Scheme of This Thesis 16 3 Literature Review and Historical Background 19 3.1 Introduction to BIM 19 3.2 BIM Definition 20 3.2.1 National Institute of Building Sciences (NIBS) 20 3.2.2 Autodesk 22 3.2.3 Associated General Contractors of America (AGC) 22 3.3 BIM Levels 24 3.4 Application of BIM 25 3.4.1 Advantages of BIM 27 3.4.2 Disadvantages of BIM 29 3.5 History of employing BIM in construction projects 30 3.5.1 Before the year 2000 30 3.5.2 After the year 2000 31 3.6 Communication in construction projects 32 3.6.1 Communication, Combination of Factors 32 3.6.2 Communication Disorders in Construction Projects 33 3.7 BIM and Project Management 35 3.7.1 BIM vs. PMBOK 36 3.8 Smart Cities 36 3.8.1 Communication in smart city projects 37 3.8.2 Project Management in Smart Cities 39 3.9 Literature Review or Relevant previous studies 40 4 Case- Study 44 4.1 Singapore 44 4.1.1 BIM Use in Singapore 45 4.1.2 ITS Projects in Singapore 49 4.1.3 Intelligent Productivity and Safety System (IPASS) 50 4.1.4 Addressing Communication Challenges by BIM in the projects in Singapore 50 4.2 London, United Kingdom (UK) 52 4.2.1 Smart Projects in London (UK) 53 4.2.2 BIM Use in UK 55 4.2.3 Addressing Communication Challenges by BIM in the projects in UK 56 5 Conclusion 61 5.1 Summary of results 61 5.1.1 Findings of the Questions 62 5.2 Further Research Recommendation 63 Bibliography VIII

The main aim of this task deals with analysis of the possibility of valuing building budgets in the building information model (BIM). The work is divided into two parts. The first, research part, presents the necessary definitions and concepts, such as definitions of BIM, LOD, CDE, which are needed for working with (BIM) software. The second, practical part, analyses possible ways to determine building budgets using the building information model on one building, which was modeled in the student version of Graphisoft ArchiCAD 24.

Digitalization is a hot topic in the construction industry today, with more and more ICT-tools being used. Web-based project platforms, digital meetings and BIM have all existed for quite a while, however the progress in using them and exploring the potential takes time. ICT-tools are usually aimed to make the information flow and communication more efficient, which directly relates to the project manager in a construction project. Therefore, the aim of this thesis is to understand how digital tools has affected the role of the client’s construction project manager over the last 5 years. The purpose is to gain knowledge on what consequences the use of digital tools such as BIM, digital project platforms and digital meetings has had on the role of the project manager and how they deal with the consequences. In this thesis, a qualitative study has been performed, where construction project managers were interviewed about their experiences of ICT-tools and the changes that has occurred over time. The findings were analyzed with a theoretical framework and discussed together with previous research and knowledge. The theoretical framework includes key aspects of leadership in project management and four main categories explaining how actors use ICT. The use of BIM was in particular analyzed using the Bew-Richards BIM maturity model. The study shows that there has been an increase in use of ICT-tools in construction projects over the last 5 years, and that there are both benefits and challenges for the project manager with all three types of digital tools that were in this study. BIM, digital project platforms and digital meetings. The primary benefits were shown to be that the tools increase timeefficiency in the project manager role and it facilitates communication. However, there is less personal contact with project team members and other challenges with getting project participants to the same use of the tools. To increase the significance of the benefits and decrease the negative consequences, the construction project manager should implement a leadership style focused on the human side, with more follow-ups, structure and routines. Technical knowledge about the ICT-tools is also beneficial to the project manager to have since it increases the possibility to guide the project team members in a good way. / Digitalisering är ett hett ämne i byggbranschen, med mer och mer ICT-verktyg som används. Webb-baserade projektplattformar, digitala möten och BIM har alla existerat ett tag nu, men utvecklingen av användandet och utforskningen av deras potential tar tid. ICT-verktyg är ofta avsedda för att effektivisera informationsflödet och kommunikationen, vilket relaterar till projektledaren i ett byggprojekt. Därför är ändamålet med det här examensarbetet att förstå hur digitala verktyg har påverkat byggprojektledarollen på beställarens sida, under de senaste 5 åren. Syftet är att få kunskap om vad för konsekvenser användningen av digitala verktyg så som BIM, digitala projektplattformar och digitala möten har medfört för en projektledare och hur de hanterar konsekvenserna. I det här examensarbetet har en kvalitativ studie gjorts där byggprojektledare intervjuades om deras erfarenheter av ICT-verktyg och de förändringar som har skett över tid. Resultaten analyserades med hjälp av ett teoretiskt ramverk och diskuterades tillsammans med tidigare forskning och kunskap. Det teoretiska ramverket inkluderar vissa nyckelaspekter av ledarskap inom projektledning och 4 huvudsakliga kategorier som beskriver hur aktörer använder ICT. Användningen av BIM var särskilt analyserad genom Bew-Richards model för BIM mognadsgrad. Studien visar att det har varit en ökning av användandet av ICT-verktyg i byggprojekt de senaste 5 åren, och att det är både fördelar och utmaningar för projektledaren med de tre typerna av digitala verktyg som var med i denna studie. BIM, digitala projektplattformar och digitala möten. De främsta fördelarna visade sig vara att verktygen ökar tidseffektiviteten i projektledarrollen och underlättar kommunikationen. Det blir dock mindre personlig kontakt med projektdeltagare och andra utmaningar med att få med projektdeltagare på samma användande av verktygen. För att öka betydelsen av fördelarna och minska de negativa konsekvenserna, borde projektledaren implementera en ledarskapsstil som är mer fokuserad på den mänskliga sidan, med mer uppföljningar, struktur och rutiner. Teknisk kunskap om ICT-verktygen är också fördelaktiga för projektledarna att ha eftersom det ökar möjligheten att vägleda medlemmarna i projektet in på rätt väg.

Digitalization has become something of a buzzword in today's society and rightly so as it brings multiple benefits and opportunities. The AEC/FM industry has constantly lagged behind other industries in terms of change and development and is often regarded as conservative. Strongly associated with digitalization in construction are the concepts of Building Information Modeling (BIM) and Virtual Design and Construction (VDC), which partly include technology and models for integrated and model-based approaches to, for example, reduce fragmentation between project members who traditionally work independently of one another. In research, it is revealed that there are large gains with a successful implementation of BIM/VDC in projects and this is something that many companies in the industry are working with and seeking to develop. The purpose of the thesis is to investigate how the project team members have worked and how the working methods are perceived by them in a well-known construction project in Sweden, where they have taken a step further in digitalizing construction by building after a legally binding digital model instead of the traditional paper drawings. The subject is explored with a qualitative method, in the form of a case study, where a scientific literature study, interview study, and observations together form the basis of the study and these parts act as a basis for the discussion. The literature study covers previous research as well as concepts relevant for answering formulated research questions and concepts that emerged during the interview study that are important to understand for a qualitative discussion and, consequently, qualitative conclusions. In the interview study, 13 respondents were interviewed in so-called semi-structured interviews and all of them were involved in the case project. The findings indicate that the BIM-model can contribute to better communication, higher resource efficiency, better quality and, at the same time for a lower total cost of the project. Identified perceptions in designing a BIM-model and then building after the model instead of 2D drawings are predominantly positive. While advantages and opportunities are demonstrated by this way of working, new challenges and risks arise. This entails legal risks, technical risks and management risks. There are new types of errors that arise with a more detailed design. / Digitalisering har blivit något av ett modeord inom dagens samhälle och det med all rätt då det medför sina fördelar och möjligheter. Bygg- och fastighetsbranschen har ständigt släpat efter övriga industrier vad gäller förändring och utveckling och ses därefter ofta som konservativ. Starkt associerat med digitalisering inom bygg är koncepten Building Information Modeling (BIM) och Virtual Design and Construction (VDC) som dels innefattar teknologi och modeller för integrerade och modellbaserade arbetssätt. De används exempelvis för att minska fragmentering mellan projektmedlemmar som vanligen enbart fokuserar på sina egna teknikområden. Inom forskningsvärlden sägs det finnas stora vinningar med en lyckad implementering av BIM/VDC i projekt och det är något som många företag inom branschen arbetar med och söker utveckla. Samtidigt anger forskningen också att det finns stora utmaningar och att man ännu inte kommit så långt med digitaliseringen. Syftet med detta examensarbete är bland annat att undersöka hur projektmedlemmar har arbetat och hur arbetssätten har upplevts i ett av få välkända byggprojekt i Sverige där man tagit ett steg längre i att digitalisera byggandet genom att bygga efter en digital modell som juridisk bygghandling istället för de traditionella pappersritningarna. Ämnet utforskas kvalitativt, i form av en fallstudie, där en vetenskaplig litteraturstudie, intervjustudie samt observationer tillsammans utgör grunden för arbetet och som alla agerar underlag för analysdelen. Litteraturstudien täcker tidigare studier på området för att beskriva kunskapsläget samt koncept som är relevanta för att besvara formulerade frågeställningar samt begrepp som dykt upp under intervjustudien som är viktiga att förstå för en kvalitativ diskussion och följaktligen likaså kvalitativa slutsatser. I intervjustudien har 13 respondenter intervjuats i så kallade semi-strukturerade intervjuer och som alla varit inblandade i det undersökta projektet. Resultatet tyder på att BIM-modellen kan bidra till en bättre kommunikation, högre resurseffektivitet, bättre kvalitet och samtidigt till en lägre totalkostnad av projektet. Identifierade upplevelser med att projektera en BIM-modell och att därefter bygga efter modellen istället för 2Dritningar är till övervägande del positiva. Samtidigt som fördelar och möjligheter påvisas med detta arbetssätt så uppkommer nya utmaningar och risker. Det medför juridiska risker, tekniska risker och hanteringsrisker. Det är exempelvis nya typer av fel som uppkommer med en mer detaljerad projektering.

The design and construction of high-performance buildings have emerged as a preferred solution for reducing energy consumption and greenhouse gas emissions. However, sometimes there is a considerable gap between the design performance and the actual performance of the buildings. Post Occupancy Evaluations (POE) provide tools to quantify the performance relative to the occupant's health, well-being, and comfort. POE is getting widely accepted to obtain feedback for various parameters such as water, energy, indoor environmental quality, and occupant comfort. Key Performance Indicators (KPIs) can be derived based on the obtained feedback to determine the performance gaps. POE has evolved to be a robust scientific methodology; however, traditional methods of conducting POE have been proven time-consuming, inconsistent, and inefficient. This research aims to conceptualize the next generation of post occupancy evaluations that leverages cutting-edge technologies such as Building Information Modeling (BIM), Internet of Things based sensors (IoT), Geographic Information Systems (GIS), and digital twins. The key contributions of this research are presented in a series of manuscripts. In the first paper, the gaps in the existing POE were determined by conducting a thorough literature review. The observed gaps were classified in data collection, analysis, and visualization categories. Broader POE definition, spot measurements of parameters, and 2D plans and charts for visualization made the existing POE procedure time-consuming. Using digital twins that combine the geometric and parametric data from BIM models and built-environment data from GIS and sensor measurements were recommended as potential solutions to address the observed gaps. The second paper explored the application of BIM-IoT-GIS integration to conduct POE. Use case scenarios were developed to derive system requirements to host the BIM-IoT-GIS-integrated POE. Four sequential tests were conducted to integrate a BIM model from Revit and sensors' data from Excel with ArcGIS pro that contained the surrounding environment data. Based on lessons learned from the tests, an optimized workflow was recommended that can be used across a variety of projects. The third paper used the BIM-IoT-GIS-integration concept to create a holistic proof of concept for digital-twin-enabled POE. The proof of concept was validated by conducting a digital-twin-based POE on the STTC building on the Red River College campus in Winnipeg. The indoor thermal comfort was visualized within the STTC digital twin developed in ArcGIS Pro. The preliminary energy consumption analysis concluded that the STTC buildings' average energy savings were approximately 70,000 KWH/year. The potential users for digital-twin-enabled POE were presented with a comparison of iv existing POE and digital-twin-based POE over a survey and a focus group discussion. Based on opinion-based feedback, the conclusion can be made that digital twins improve the overall efficiency of POE. The fourth paper recommended the digital-twin-enabled POE procedure for UVic's engineering expansion project. It established the semantics for POE, followed by a digital twin execution plan that can be used for developing a digital twin during each phase (from planning to operations) of the project. Furthermore, the benefits of the digital-twin-enabled POE procedure were demonstrated by comparison with the existing POE procedure relative to the project phases. This study concluded that conducting the POE on the UVic ECS expansion project will enable the researchers to determine the effectiveness of sustainable features by comparing the performance of existing and proposed facilities. In conclusion, BIM-IoT-GIS-integrated digital twins address the limitations of data collection, analysis, and visualization. These digital twins will enable multi-objective analysis and spatial-temporal visualization and provide deeper insights into the way these high-performance buildings function. / Graduate / 2023-05-24

Inom byggbranschen är förseningar i byggprojekts leveranser vanligt förekommande och utgör ett utmanande problem. För att övervinna detta problem krävs en utvecklad och noggranna tidsplaner som följs av alla berörda parter. I dagsläget använder sig byggindustrin huvudsakligen av aktivitetsbaserade tidsplaner, som representeras av Critical Path Method (CPM) och dess Gantt-schema. Trots att en annan tidsplaneringsmetod, baserad på lokalitet, visar lovande potential för att effektivt hantera förseningar i leveranser, har den inte fått samma uppmärksamhet. Line of Balance (LOB) är en metod inom location-based scheduling (LBS), där platsen behandlas som en dimension i produktionsprocessen och den har visat sig vara effektivare alternativ än Gantt-schema. I stället för att endast fokusera på aktiviteter fördelar LOB-metoden aktiviteter på lokaliteter där de ska genomföras, vilket underlättar för alla parter att följa en tydlig plan och övervaka arbeten på ett enklare sätt. BIM-verktyg har visat sig vara effektiva för att förbättra och effektivisera LOB-tidsplaner genom att tillhandahålla specialiserade verktyg som underlättar skapandet av sådana tidsplaner. CPM-metoden utgör grunden för många programvaror som används för att skapa LOB-tidsplaner såsom programmet Vico Office. Dessutom kan användning av BIM-verktyg möjliggör tillämpning av Lean-tänkande som bidrar till att förbättra arbetsflöden och öka kvalitet. Detta arbete undersöker hur tidsplaner påverkas vid övergången från aktivitetsbaserade- till lokalitetsbaserade tidsplaner, med hjälp av BIM-verktyget Vico Office. Vidare undersöks hur tillämpningen av Lean-principer såsom ”Visual Management” och ”Reduced Batch Size” påverkar lokalitetsbaserade tidsplaner. Gantt-schemat som analyserades i detta arbete är specifikt för byggföretaget NCC och omfattar utförandet av Hus A, Hus B och en förskola. Dock var endast detaljerad information om aktiviteterna som ingår i Hus A tillgängliga. För att konvertera Gantt-schemat till en lokalitetsbaserad tidsplan användes aktiviteternas varaktighet och lokaliteter från Gantt-schemat. För aktiviteter som saknade lokaliteter användes logiskt tänkande och erfarenhet för att fördela dem i lämpliga lokaliteter. Dessutom användes byggritningar som tillhandahölls av Karlstads kommunen för att tillämpa Lean-principen ”Reduced Batch Size”. LOB-diagrammet som genereras av Vico:s Schedule Planner är i enlighet med Lean-principen ”Visual Management”. Det har visat en förbättrad förmåga att upptäcka tomma områden, kollisioner och diskontinuerliga aktiviteter jämfört med Gantt-scheman. Dessutom användes Lean-principen ”Reduced Batch Size” för att minska väntetiderna mellan arbetsgrupper. Trots att byggprojektets ledtid överskreds med två veckor, resulterade konverteringen i en lokalitetsbaserad tidsplan som främjade ett kontinuerligt arbetsflöde, en effektivare arbetsrytm och en förbättrad uppföljning av byggprocessen. / In the construction industry, delays in project deliveries are a common occurrence. In order to overcome this problem, a developed and accurate schedule is required which is followed by all parties concerned. Currently, traditional activity-based Gantt charts are primarily used. However, another scheduling method based on location shows promising potential for effectively managing delivery delays but has not received the same attention. Line of Balance (LOB) is a method within location-based scheduling (LBS), where the location is treated as a dimension in the production process, and it has proven to be a more efficient alternative than the Gantt chart. BIM tools, such as Vico Office, have proven effective in improving and streamlining LOB schedules by providing specialized tools that facilitate the creation of such schedules. Additionally, the use of BIM tools can facilitate the application of Lean thinking, which helps improve workflows. This study investigates how schedules are affected by the transition from activity-based to location-based schedules using the BIM tool Vico Office. Furthermore, it examines how the application of Lean principles such as “Visual Management” and “Reduced Batch Size” affect location-based schedules. The Gantt chart analysed in this study belongs to the construction company NCC and contains detailed information on the activities in House A. The LOB diagram generated by Vico's Schedule Planner aligns with the Lean principle of “Visual Management”. It has demonstrated an improved ability to detect empty areas, collisions, and discontinuous activities compared to Gantt charts. Additionally, the Lean principle of “Reduced Batch Size” was utilized to minimize waiting times between work groups. Despite the project exceeding the allotted time by two weeks, the conversion resulted in a location-based schedule that promoted continuous workflow, enhanced efficiency, and improved monitoring of the construction process.

<p>  </p> <p>This research is motivated by the potential of autonomous mobile robots (AMRs) in enhancing safety, productivity, and efficiency in the construction industry. The dynamic and complex nature of construction sites presents significant challenges to AMRs, particularly in localization and mapping – a process where AMRs determine their own position in the environment while creating a map of the surrounding area. These capabilities are crucial for autonomous navigation and task execution but are inadequately addressed by existing solutions, which primarily rely on visual Simultaneous Localization and Mapping (SLAM) methods. These methods are often ineffective in construction sites due to their underlying assumption of a static environment, leading to unreliable outcomes. Therefore, there is a pressing need to enhance the applicability of AMRs in construction by addressing the limitations of current localization and mapping methods in addressing the dynamic nature of construction sites, thereby empowering AMRs to function more effectively and fully realize their potential in the construction industry.</p> <p>The overarching goal of this research is to fulfill this critical need by developing a novel visual SLAM framework that is capable of not only detecting and segmenting diverse dynamic objects in construction environments but also effectively interpreting the semantic structure of the environment. Furthermore, it can efficiently integrate these functionalities into a unified system to provide an improved SLAM solution for dynamic, complex, and unstructured environments. The rationale is that such a SLAM system could effectively address the dynamic nature of construction sites, thereby significantly improving the efficiency and accuracy of robot localization and mapping in the construction working environment. </p> <p>Towards this goal, three specific objectives have been formulated. The first objective is to develop a novel methodology for comprehensive dynamic object segmentation that can support visual SLAM within highly variable construction environments. This novel method integrates class-agnostic objectness masks and motion cues into video object segmentation, thereby significantly improving the identification and segmentation of dynamic objects within construction sites. These dynamic objects present a significant challenge to the reliable operation of AMRs and, by accurately identifying and segmenting them, the accuracy and reliability of SLAM-based localization is expected to greatly improve. The key to this innovative approach involves a four-stage method for dynamic object segmentation, including objectness mask generation, motion saliency estimation, fusion of objectness masks and motion saliency, and bi-directional propagation of the fused mask. Experimental results show that the proposed method achieves a highest of 6.4% improvement for dynamic object segmentation than state-of-the-art methods, as well as lowest localization errors when integrated into visual SLAM system over public dataset. </p> <p>The second objective focuses on developing a flexible, cost-effective method for semantic segmentation of construction images of structural elements. This method harnesses the power of image-level labels and Building Information Modeling (BIM) object data to replace the traditional and often labor-intensive pixel-level annotations. The hypothesis for this objective is that by fusing image-level labels with BIM-derived object information, a segmentation that is competitive with pixel-level annotations while drastically reducing the associated cost and labor intensity can be achieved. The research method involves initializing object location, extracting object information, and incorporating location priors. Extensive experiments indicate the proposed method with simple image-level labels achieves competitive results with the full pixel-level supervisions, but completely remove the need for laborious and expensive pixel-level annotations when adapting networks to unseen environments. </p> <p>The third objective aims to create an efficient integration of dynamic object segmentation and semantic interpretation within a unified visual SLAM framework. It is proposed that a more efficient dynamic object segmentation with adaptively selected frames combined with the leveraging of a semantic floorplan from an as-built BIM would speed up the removal of dynamic objects and enhance localization while reducing the frequency of scene segmentation. The technical approach to achieving this objective is through two major modifications to the classic visual SLAM system: adaptive dynamic object segmentation, and semantic-based feature reliability update. Upon the accomplishment of this objective, an efficient framework is developed that seamlessly integrates dynamic object segmentation and semantic interpretation into a visual SLAM framework. Experiments demonstrate the proposed framework achieves competitive performance over the testing scenarios, with processing time almost halved than the counterpart dynamic SLAM algorithms.</p> <p>In conclusion, this research contributes significantly to the adoption of AMRs in construction by tailoring a visual SLAM framework specifically for dynamic construction sites. Through the integration of dynamic object segmentation and semantic interpretation, it enhances localization accuracy, mapping efficiency, and overall SLAM performance. With broader implications of visual SLAM algorithms such as site inspection in dangerous zones, progress monitoring, and material transportation, the study promises to advance AMR capabilities, marking a significant step towards a new era in construction automation.</p>

Den ökande komplexiteten i byggprojekt ger upphov till behovet av ett effektivt sätt att designa, hantera och underhålla strukturer. Byggnadsinformationsmodellering (BIM) underlättar dessa processer genom att tillhandahålla en digital representation av fysiska strukturer. Tekla Structures (TS) har blivit en populär programvara för byggnadsinformationsmodellering inom konstruktionsdesign. I konstruktionsingenjörskap spelar kopplingar en viktig roll i att förena olika byggnadsobjekt. Trots det återstår utmaningen att effektivt och noggrant designa kopplingar i TS på grund av det breda spektrumet av tillgängliga kopplingstyper. Befintliga lösningar för rekommendation av anslutningar förlitar sig ofta på fördefinierade regler, vilket begränsar deras tillämplighet och kräver tidskrävande installation. Nylig forskning har undersökt maskininlärningsmetoder för rekommendation av anslutningar, men de lider av skalbarhetsproblem eller hög beräkningskostnad. Denna avhandling behandlar problemet med rekommendation av anslutningstyp i Tekla Structures som en klassificeringsuppgift, genom att dra nytta av de olika representationerna av BIM-objekt, inklusive 2D-bilder och attribut. Avhandlingen förbättrar befintliga metoder för enskilda datakällor genom att jämföra XGBoost med random forest för attribut, samtidigt som den förbättrar den tidigare CNN-modellen för bildklassificering. Dessutom undersöker detta projekt potentialen att kombinera bilder och attributdata för klassificering av anslutningstyper, genom att använda två multimodala strategier för datafusion: sen fusion och intermediär fusion. Resultaten visar att XGBoost med metadata från attributdatamängden ger bästa prestanda, med en maximal noggrannhet på 0.9283, och de experimentella multimodala datametoderna kan inte ytterligare optimera klassificeringsresultaten. Noggrannheten för attributbaserade metoder förbättras med upp till 0.6%. Förbättringen i CNN-modellen kan öka klassificeringsnoggrannheten med upp till 5%. Genom att jämföra olika datakällor och tillvägagångssätt syftar denna avhandling till att ge en praktisk rekommendation för anslutningsdesign och därigenom lägga grunden för bättre anslutningsdesignprocesser inom byggprojekt. / The increasing complexity of construction projects gives rise to the need for an efficient way of designing, managing, and maintaining structures. Building Information Modeling (BIM) facilitates these processes by providing a digital representation of physical structures. Tekla Structures (TS) has emerged as a popular building information modeling software for structural design. In structural engineering, connections play an important role in joining various building objects. However, the efficient and accurate design of connections in TS remains a challenge due to the wide range of available connection types. Existing solutions for connection recommendation often rely on predefined rules, limiting their applicability and requiring time-consuming setup. Recent research has explored machine learning approaches for connection recommendation, but they suffer from scalability issues or high computational costs. This thesis addresses the connection type recommendation problem in TS as a classification task, leveraging the diverse representations of the BIM objects, including 2D images and attributes. This thesis improves existing approaches for single modality data, comparing XGBoost with random forest for attributes, while enhancing the previous CNN model for image classification. Furthermore, this thesis investigates the potential of combining images and attribute data for connection type classification, using two multimodal data fusion strategies: late fusion and intermediate fusion. The results show that XGBoost with metadata of the attribute dataset yields the best performance, with a maximum accuracy of 0.9283, and the experimented multimodal data fusion methods are unable to further optimise the classification results. The accuracy of attribute-based methods is improved by up to 0.6%. The improvement in CNN model can enhance the classification accuracy by up to 5%. By comparing various data sources and approaches, this thesis aims to provide a practical connection recommendation design, thereby laying a foundation for better connection design processes in construction projects.