Publications
Sentiment and semantic analysis: Urban quality inference using machine learning algorithms
Year: 2024
Author(s): Emily Ho, Michelle Schneider, Sanjay Somanath, Yinan Yu, Liane Thuvander
Publication Type: Journal article
Sustainable urban transformation requires comprehensive knowledge about the built environment, including people's perceptions, use of sites, and wishes. Qualitative interviews are conducted to understand better people's opinions about a specific topic or location. This study explores the automatization of the interview coding process by investigating how state-of-the-art natural language processing techniques classify sentiment and semantic orientation from interviews transcribed in Swedish. For the sentiment analysis, the Swedish bidirectional encoder representations from transformers (BERT) model KB-BERT was used to perform a multi-class classification task on a text sentence level into three different classes: positive, negative, and neutral. Named entity recognition (NER) and string search were used for the semantic analysis to perform multi-label classification to match domain-related topics to the sentence. The models were trained and evaluated on partially annotated datasets. The results demonstrate that the implemented deep learning techniques are a possible and promising solution to achieve the stated goal.
Temporal effective stress response of soil elements below the base of an excavation in sensitive clay
Year: 2024
Author(s): Johannes Tornborg, Mats Karlsson, Jelke Dijkstra
Publication Type: Journal article
Investigations into the temporally evolving stress state below the base of excavations and underground structures are very scarce, in contrast to studies of horizontal earth pressures during the construction stage. Therefore in this work, the measured temporal response in terms of vertical and horizontal effective stresses and displacements below a tunnel slab at the base of an excavation located in a deep sensitive clay deposit is reported. In addition to the measured unloading response over time, the completeness of the site description and complementary measurements enables future benchmarking of numerical models at boundary and element level. Instrument clusters of earth pressure cells and piezometers were installed at three locations in one cross-section. The monitoring data allows the interpretation of effective stress paths and stress ratios, K=σ'h/σ'v, at soil element level covering the construction and the serviceability stages. The in situ stress ratios enable a unique comparison to prior laboratory studies of K during unloading. The data presented herein on the evolution of K corroborate, although approximately, previous studies at laboratory scale. Furthermore, at system level, the monitoring data reveal the intricate interplay between deformations resulting from excavation and pile driving.
FEniCS simulation of artesian conditions in clay slope
Year: 2024
Author(s): Kseniia Muratova, Ayman Abed, Minna Karstunen
Publication Type: Other conference contribution
This paper presents a numerical simulation of the response of a slope with ar-
tesian conditions under the impact of a long heavy rainfall event. Sensitivity
analysis of surface and bedrock inclination are presented. The robust FEniCS
finite element tool is used to solve the governing coupled Hydro-Mechanical
(HM) balance equations. An extended version of the Mohr-Coulomb model is
implemented in Fortran and used as soil model. The paper discusses how the
stability is influenced by changes in pore water pressure and the occurrence of
artesian conditions, exploring slopes with different bedrock inclinations. It also
presents potential failure surfaces as part of the discussion. The results provide
better understanding of critical geological conditions, which helps to identify
regions most at risk and take the required safety measures.
Scintillating and decorrelating signals for different propagation paths in a random medium
Year: 2024
Author(s): Jens Forssén
Publication Type: Journal article
Randomness in the medium of propagation affects the signal received at a distance from a source resulting in amplitude and phase fluctuations as well as reduced coherence between different paths of propagation. A method is described and validated for incorporating such fluctuations and decorrelations in sound signals in a single approach. Here, simplified conditions are used of open space, with or without a single reflecting surface, assuming fluctuations as for plane waves in homogeneous and isotropic temperature turbulence. Validating results are shown for Gaussian and von Karman turbulence models and exemplifying sound clips are provided as supplementary material.
Buildings as material mines - Towards digitalization of resource cadasters for circular economy
Year: 2024
Author(s): Maud Lanau, Leonardo Rosado, Danielle Densley Tingley, Holger Wallbaum
Publication Type: Book chapter
Existing buildings are valuable resources of secondary material, whose carefully planned recovery and reuse have the potential to decrease the construction industry’s resource use, waste generation, and associated embodied carbon emissions. To enable circular strategies, stakeholders across the value chain need varying types of information, including type, quantity, quality, and location of material and building components. The bottom-up approach used in material stock analysis (MSA) provides useful estimates of material stocks in buildings by applying average material quantities to typical buildings. But such results are too coarse to inform circular strategies – building-specific and component-level information is also needed. In that respect, data capture and artificial intelligence open many possibilities. This chapter investigates the level of information needed by stakeholders, and the potential of integrating machine learning, big data, and remote sensing into traditional MSA. The chapter also discusses how these approaches may be leveraged to automatically update buildings’ material and component information with the support of a digital twin platform, thereby facilitating the sharing of information with key stakeholders.
A hybrid workflow connecting a network and an agent-based model for predictive pedestrian movement modelling
Year: 2024
Author(s): Anita Ullrich, Franziska Hunger, Ioanna Stavroulaki, Adam Bilock, Klas Jareteg, Yury Tarakanov, Alexander Gösta, Meta Berghauser Pont, Fredrik Edelvik
Publication Type: Journal article
Pedestrian movement has always been a main concern for urban planning and design, but has become more important within the sustainable development agenda, as walking is crucial to reduce urban emissions and foster livable cities. Therefore, urban planners need to be able to take pedestrian movement into consideration as part of the workflow of planning and designing cities. This study outlines a comprehensive workflow tailored for urban planners. It proposes a hybrid model that integrates an agent-based model, which simulates the micro-scale movement of pedestrians in outdoor urban environments, with a network model, which predicts the aggregated pedestrian flows on a macro-scale. The hybrid model has been applied to a pedestrian precinct in the city centre of Gothenburg, Sweden and has been compared to real-world measurements. The reasonable agreement between the simulation results and the real-world data supports the reliability of the proposed workflow, underscoring the model’s capability of predicting pedestrian movement statistically on a large scale and individually on a local scale. Furthermore, the model enables the analysis of flow distributions and movement restrictions and facilitates the analysis of different design scenarios as well as specific pedestrian behavior. This functionality is valuable for urban design and planning practice, contributing to the optimization of pedestrian flow dynamics.
Collaborative site layout planning using multi-touch table and immersive VR
Year: 2023
Author(s): Mikael Johansson, Mattias Roupé, Mikael Viklund Tallgren
Publication Type: Paper in proceeding
Building Information Modeling (BIM) is changing the way architects and engineers produce and deliver design results, and object-oriented 3D models are now starting to replace traditional 2D drawings during the construction phase. This allows for a number of applications to increase efficiency, such as quantity take-off, cost-estimation, and planning, but it also supports better communication and increased understanding at the construction site by means of detailed 3D models together with various visualization techniques. However, even in projects with a fully BIM-based design, there is one remaining part that is still done primarily using 2D drawings and sketches – the construction site layout plan. In addition to not take advantage of the benefits offered by 3D, it also makes it difficult to integrate site layout planning within the openBIM ecosystem. In this paper we present the design and evaluation of a user-friendly, IFC-compatible software system that supports collaborative, multi-user creation of construction site layout plans using both multi-touch table and immersive VR. By allowing temporary structures, machines, and other components to be easily added and updated it is possible to continuously produce and communicate 3D site layout plans that are aligned with the schedule and supports integration with other BIM-tools.
DTCC Builder: A mesh generator for automatic, efficient, and robust mesh generation for large-scale city modeling and simulation
Year: 2023
Author(s): Vasilis Naserentin, Anders Logg, Dag Wästberg
Publication Type: Journal article
Digital Twin Cities Centre (DTCC) Builder is a mesh generator for automatic, efficient, and robust mesh generation for large-scale city modeling and simulation. Using standard and widely available raw data sources in the form of point clouds and cadastral data, DTCC Builder generates high-quality 3D surface and volume meshes, suitable for both visualization and simulation. In particular, DTCC Builder is capable of generating large-scale, conforming tetrahedral volume meshes of cities suitable for finite element (FEM) simulation.
Visualisering och auralisering av buller i stadsmiljö
Year: 2020
Author(s): Beata Stahre Wästberg, Jens Forssén, Liane Thuvander, Monica Billger, Anders Logg, Fabio Latino
Publication Type: Magazine article
Inom dagens stadsplanering används oftast 2D-kartor för att visa beräknade värden för framtida bullernivåer. Dessa kartor kan ibland vara svåra att tolka. För att öka förståelsen och underlätta kommunikationen mellan olika aktörer inom planeringsprocessen, har vi inom forskningsprojektet DemoVirPEN på Chalmers och Göteborgs universitet utvecklat en prototyp av ett verktyg för visualisering och auralisering (ljudsimulering) av buller från vägtrafik i en stadsmodell.
Data-Informed Urban Design: An Overview of the Use of Data and Digital Tools in Urban Planning and Design
Year: 2020
Author(s): Alexander Gösta, André Agi, Jacob Flårback, Jesper Karlsson, Ellen Simonsson
Publication Type: Journal article
This article aims to map how different digital tools can be useful for architects and how they might affect their work processes. Researchers and professionals were interviewed to investigate what they found valuable to measure, which methods they used within their analyses, as well
as the opportunities and risks they see for the future of the field with regards to digital tools. As part of the survey, a workshop was held with architects and project managers examining the possibilities of connecting existing methods and tools to the sustainability certification system,
City Lab Action Guide, and through that, to achieve a more ambitious set of sustainability goals for the projects. Findings from the study indicate that there are risks associated with giving data an increasingly important role in the design work. A working model never provides the full truth
but is inherently limited by its constraints. It is important to acknowledge that all angles and aspects of a problem can never be represented in a model. Another possible risk identified lies in the quality of, and access to, data. In a scenario where data plays an increasingly important
role, it is not only the quality of the datasets that is of utmost importance, but it is equally important that the urban planners who request the analyses ask the questions first, and then collect the necessary data, instead of vice versa.