Model Building Research Articles

Open Data-Driven 3D Building Models for Micro-Population Mapping in a Data-Limited Setting

Urban planning and management increasingly depend on accurate building and population data. However, many regions lack sufficient resources to acquire and maintain these data, creating challenges in data availability. Our methodology integrates multiple data sources, including aerial imagery, Points of Interest (POIs), and digital elevation models, employing Light Gradient Boosting Machine (LightGBM) and Gradient Boosting Decision Tree (GBDT) to classify building uses and morphological filtration to estimate heights. This research contributes to bridging the gap between data needs and availability in resource-constrained urban environments, offering a scalable solution for global application in urban planning and population mapping.

Reviewing Open Data Semantic 3D City Models to Develop Novel 3D Reconstruction Methods

Abstract. Structured semantic 3D city models are pivotal in creating urban 3D digital twins. The wide adoption of such models has been primarily enabled by robust, model-based, and automatic 3D reconstruction methods. However, these methods impose requirements on the reconstruction, mainly restricting the solution space to several model types and relying on accurate 2D footprints. Recent research shows that deep-learning-based methods promise highly generic solution space and are footprint-free. Yet, the current training and test datasets are limited, hindering the methods’ development. In this work, we analyze the ubiquity of already existing, open 3D city model datasets and their potential to serve as a large-scale training and test set for 3D reconstruction, where 27 potential dataset collections have been identified. Our review shows that more than 215 million building models are readily available. We firmly believe that this review will facilitate further research on robust automatic 3D city model reconstruction and serve as a reference for benchmarking 3D city models.

Development and validation of a machine learning‐based model of ischemic stroke risk in the Chinese elderly hypertensive population

AbstractMachine learning (ML) has made some significant contributions to stroke prevention, but the stability and accuracy of existing models for clinical applications are uncertain. This study develops and validates an interpretable ML model using metabolic and coagulation biomarkers to predict ischemic stroke in elderly hypertensive patients in Northwest China. The prediction model used 453 electronic medical records for the model building (80% as a training set and 20% as a test set) and 132 for external validation. The final seven key features (D‐dimer, cystatin C, homocysteine, hemoglobin A1c, prothrombin time, low‐density lipoprotein C, and triglyceride glucose‐body mass index) were selected by the advanced approach, elastic net, and classical wrapping approaches. The final model, eXtreme gradient boosting, was identified as having superior performance than the other 9 classifers (random forest, Gaussian process, multilayer perceptron, logistic regression, support vector machine, K‐nearest neighbor, decision tree, Gaussian naive bayes, and ensemble model), with area under the receiver‐operating characteristic curves of 0.97 and 0.94 for the test and external validation sets, respectively. The final model demonstrates excellent stability, accuracy, and clinical usefulness through various metrics and decision curve analysis. Additionally, an online human–machine interface application has been developed for clinical practice to help early identification and intervention for ischemic stroke in elderly hypertensive patients.

A 3D X-tree optimized for Complex Indoor Building Models

Abstract. Studying complex indoor buildings is crucial for population evacuation research in our modern world, where indoor spaces are the primary living environment. However, the existing spatial indexes do not fully align with the unique characteristics of indoor space, leading to a significant bottleneck in the efficiency of searching many space areas. This article introduces a novel model that can efficiently retrieve and update compact three-dimensional indoor space by introducing space X-trees. Our experiments, which have shown that the introduction of X-trees, which can dynamically adjust the size of space nodes, has significantly improved the retrieval speed during the query operation process, underscore the reliability and potential of our new model.

Study Report on Optimum Utilisation Plan of Ngoma 22 Irrigation Scheme

The study was conducted in Ngoma 22 irrigation scheme located in Remera and Rurenge sectors of Ngoma District in Eastern Province of Rwanda between November 2018 and January 2019. It was recommended by RAB Board of Directors. The methodology used encompasses the study area, sampling procedures, data collection and analysis tools. The main data collection tool used was FGD which targeted site agronomists and household head farmers producing rice, fruits and vegetables as high value crops in the scheme area. Through pair wise ranking technique most profitable cash crops in the scheme were ranked as follows: (1) tree tomato, (2) maracuja, (3) egg plants, (4) water melon, (5) tomato, (6) irish potatoes, (7) french beans, (8) onions, (9) carrots, (10) cabbages, (11) beetroots. From this point of views, findings from the study show only two most profitable crops: maracuja with 696 Frw/kg followed by tree tomato with Frw 249 Frw/kg. The third ranked cash crop being Irish potato (with a loss of -40 Frw/kg), the fourth is egg plant with a loss of- 177 Frw. The existing paddy rice production has shown a loss of – 270 Frw inspite the water availability all the year-round. Maize and bush bean show respectively losses of -543 Frw/kg and -1743 Frw. Hence it is understandable why farmers in the scheme area have already abandoned maize and beans cultivation. Findings show also water fees (20,000 Frw/season) determined based on maintenance costs and farmers income. For capacity building model, more than one thousand beneficiary farmers in Ngoma 22 Irrigation Scheme need strong farmer organisation for ownership of the scheme and increased profitability. To that effect, they need continued capacity building as stated in the tripartite IMTA: administrative and financial management, record keeping, and conflicts resolution management, integrated water management, operation and maintenance of irrigation infrastructures, best agricultural practices for sustainable farmers’ organisations and commercial farming. Key recommendations include fixing minimum prices for agricultural products in relation to costs of production- or setting subsidy prices, increase yield for each crop by meeting all requirements: lime, fertilisers and improvement of irrigation technologies in order to reach automatically increased yield. Farmers should continue to leave out traditional crops and adopt profitable cash crops identified in this study (e.g maracuja and tree tomato). They should also increase the ownership spirit through IWUA and cooperatives performance. They should adopt collective marketing of their products and pay water fees per season.

A Specialized Pipeline for Efficient and Reliable 3D Semantic Model Reconstruction of Buildings from Indoor Point Clouds.

Recent advances in laser scanning systems have enabled the acquisition of 3D point cloud representations of scenes, revolutionizing the fields of Architecture, Engineering, and Construction (AEC). This paper presents a novel pipeline for the automatic generation of 3D semantic models of multi-level buildings from indoor point clouds. The architectural components are extracted hierarchically. After segmenting the point clouds into potential building floors, a wall detection process is performed on each floor segment. Then, room, ground, and ceiling extraction are conducted using the walls 2D constellation obtained from the projection of the walls onto the ground plan. The identification of the openings in the walls is performed using a deep learning-based classifier that separates doors and windows from non-consistent holes. Based on the geometric and semantic information from previously detected elements, the final model is generated in IFC format. The effectiveness and reliability of the proposed pipeline are demonstrated through extensive experiments and visual inspections. The results reveal high precision and recall values in the extraction of architectural elements, ensuring the fidelity of the generated models. In addition, the pipeline's efficiency and accuracy offer valuable contributions to future advancements in point cloud processing.

Deep attention for enhanced OCT image analysis in clinical retinal diagnosis

AbstractRetinal illnesses such as age-related macular degeneration (AMD) and diabetic maculopathy pose serious risks to vision in the developed world. The diagnosis and assessment of these disorders have undergone revolutionary change with the development of optical coherence tomography (OCT). This study proposes a novel method for improving clinical precision in retinal disease diagnosis by utilizing the strength of Attention-Based DenseNet, a deep learning architecture with attention processes. For model building and evaluation, a dataset of 84495 high-resolution OCT images divided into NORMAL, CNV, DME, and DRUSEN classes was used. Data augmentation techniques were employed to enhance the model's robustness. The Attention-Based DenseNet model achieved a validation accuracy of 0.9167 with a batch size of 32 and 50 training epochs. This discovery presents a promising route for more precise and speedy identification of retinal illnesses, ultimately enhancing patient care and outcomes in clinical settings by integrating cutting-edge technology with powerful neural network architectures.

Analisis Dinamik Model Predator-Prey dengan Efek Pemanenan Pada Populasi Ikan Nike (Awaous Melanocephalus) di Provinsi Gorontalo

Gorontalo Province is one of the areas directly adjacent to Tomini Bay, which is known for its rich natural resources, especially in the fisheries sector such as Nike fish (Awaous melanocephalus). Nike fish is the main food commodity for people in the Gorontalo region. However, intensive fishing activities can pose an extinction threat to this species. This study aims to understand the dynamics of the Nike fish population using a predator-prey model that integrates the effects of harvesting on the prey population. The research process includes formulation of assumptions, model building, model validation, and data analysis through mathematical analysis and numerical simulation. The analysis showed that the model has four equilibrium points, each with different stability conditions. Numerical simulations showed that an increase in harvesting rate and natural mortality of prey, as well as a decrease in birth rate, could threaten the viability of the Nike fish population. Thus, appropriate management efforts are needed to maintain ecosystem balance and the sustainability of the Nike fish population. The calculation results show that the Nike fish production threshold in Gorontalo Province is 8.470.858 kg per year. However, the amount of Nike fish production in Gorontalo Province in 2023 only reached 56,75% of the threshold. This indicates that the production of Nike fish in Gorontalo can still be increased as long as the harvest limit is observed to prevent extinction.

GCLmf: A Novel Molecular Graph Contrastive Learning Framework Based on Hard Negatives and Application in Toxicity Prediction.

In silico methods for prediction of chemical toxicity can decrease the cost and increase the efficiency in the early stage of drug discovery. However, due to low accessibility of sufficient and reliable toxicity data, constructing robust and accurate prediction models is challenging. Contrastive learning, a type of self-supervised learning, leverages large unlabeled data to obtain more expressive molecular representations, which can boost the prediction performance on downstream tasks. While molecular graph contrastive learning has gathered growing attentions, current models neglect the quality of negative data set. Here, we proposed a self-supervised pretraining deep learning framework named GCLmf. We first utilized molecular fragments that meet specific conditions as hard negative samples to boost the quality of the negative set and thus increase the difficulty of the proxy tasks during pre-training to learn informative representations. GCLmf has shown excellent predictive power on various molecular property benchmarks and demonstrates high performance in 33 toxicity tasks in comparison with multiple baselines. In addition, we further investigated the necessity of introducing hard negatives in model building and the impact of the proportion of hard negatives on the model.

Geometry, Topology and p-Adic Numbers in Geospatial Data Modelling, Management and Processing: Review and Future Approaches

Abstract. Geometry and topology are both central concepts in geospatial data modelling and management. While geometry is the central concept for computational geometry applications, e.g. to intersect surfaces and solids in 3D space, topology is helpful for many application classes starting from city modelling to subsurface modelling and indoor navigation. Furthermore, p-adic numbers are useful for describing hierarchical processes on topological models. In this paper we first shortly review geometry- and topology-based approaches used for geospatial applications. We then describe our approach on turning geometry “upside down” focusing on topology during the whole process of distributed geospatial computing, data modelling, data management, and simulation. Furthermore, the way to use p-adic numbers for the description of hierarchical processes on topological models is shown for the example of simulation and the idea of p-adic analysis in distributed simulations is presented in the context of topological city and building models. The approach opens new insights such as topological relationships, components, and efficiently studying of the approximate behaviour of processes in the built environment using distributed computational systems. Finally, exemplary applications are presented to underline the importance of this new approach.

Wind tunnel experiment of ventilation rate and airflow characteristics in natural ventilation induced by wind and buoyancy effects

In this paper, the investigation focused on natural ventilation induced by wind and buoyancy effects through a wind tunnel experiment. A scaled-down model of a single-zone building featuring two opposing openings at varying heights was subjected to uniform heating from a source on the interior floor. Wind pressure coefficient differences were measured in a sealed model to assess the wind’s driving force. Indoor and outdoor temperatures were recorded to ascertain the buoyancy-driven driving force. The ventilation rate was assessed using the tracer gas constant emission method, while airflow characteristics were quantified through particle image velocimetry (PIV). The comparison between predicted and measured ventilation rates revealed generally good accuracy in cases where wind assists buoyancy. However, errors were evident in cases of opposing wind and buoyancy due to neglecting the effect of wind turbulence. The airflow in cases of opposing wind and buoyancy effects demonstrates the most dynamic changes, as illustrated through flow visualizations and transient temperature fluctuations at the openings.

Research on Energy-Saving Transformation of Rural Residential Building Envelope Structures and Heating Modes in Northeast China

Rural areas in Northeast China present a large demand for heating energy in winter, but there are problems in such areas with poor thermal performance of building envelopes and poor indoor thermal comfort. In addition, coal-fired boilers are still widely used. China’s “Dual-Carbon Goals” and “Clean Heating” policy call for the creation of a green and comfortable living environment for rural residential buildings. This paper considers the impact of the improvement of the thermal performance of envelope structures on the initial investment of the transformation program and the rated power of the ASHP and proposes an energy-saving transformation method to replace traditional coal-fired boilers with the ASHP on the basis of the improvement of the thermal performance of envelope structures. By establishing a typical rural residential building model in Northeast China, this energy-saving method is simulated based on TRNSYS. The results show that the payback period of investment of the transformation method of “envelope structure + heating system” is not superior to that of the transformation method of only improving the thermal performance of the envelope structure, but it has advantages in the comprehensive life-cycle benefits and it has great advantages in improving the satisfaction of rural residents in the use of heating systems.

Applications for Semantic 3D Streetspace Models and Their Requirements—A Review and Look at the Road Ahead

In addition to geometric accuracy, topological information, appearance and georeferenced data, semantic capabilities are key strengths of digital 3D city models. This provides the foundation for a growing number of use cases, far beyond visualization. While these use cases mostly focused on models of buildings or the terrain so far, the increasing availability of data on roads and other transportation infrastructure opened up a range of emerging use cases in the field of semantic 3D streetspace models. While there are already a number of implemented examples, there is also a potential for new use cases not yet established in the field of 3D city modeling, which benefit from detailed representations of roads and their environment. To ensure clarity in our discussions, we introduce an unambiguous distinction between the terms `application domain’, `use case’, `functionality’ and `software application’. Based on these definitions, use cases are categorized according to their primary application domain and discussed with respect to relevant literature and required functionalities. Furthermore, requirements of functionalities towards semantic 3D streetspace models are determined and evaluated in detail with regard to geometric, semantic, topological, temporal and visual aspects. This article aims to give an overview on use cases in the context of semantic 3D streetspace models and to present requirements of respective functionalities, in order to provide insight for researchers, municipalities, companies, data providers, mapping agencies and other stakeholders interested in creating and using a digital twin of the streetspace.

Comparison of energy prediction models for residential buildings: A case study in Himachal Pradesh, India

Abstract The demand for consumption of energy has been increasing globally with the tremendous increase in population. Different studies have proved that inadequate energy management and planning may lead to energy crisis which is a result of inadequacies in energy prediction. Accurate prediction of energy demand is important as underestimation may lead to shortage in supply and overestimation may lead to overinvestment in energy generation. Various available literature has been reviewed for determining the various factors responsible for affecting the energy consumption of residential buildings. Based on the factors determined, survey questionnaire has been formulated and survey was conducted in 400 residential buildings in one of northern states of India, i.e., Himachal Pradesh. It was observed by reviewing various studies that different models developed for energy consumption by different researchers were based on either of the three approaches, namely, engineering-based, AI-based, and hybrid approaches. Three tools namely, case-based reasoning, artificial neural network, and multilinear regression, based on these approaches were individually used for developing the model in this study, and their prediction results were compared. It was observed that the accuracy in the overall predicted results was highest in the proposed ANN model, followed by CBR model, and MLR model, with an overall accuracy of 99.93%, 96.3%, and 91.7%, respectively. The error obtained in the predicted results using ANN, CBR and MLR ranges from -4.0 to +3.0%, -15.0 to +26.0%, -30.0% to 20.0%, respectively. The overall RMSE of ANN, CBR, and MLR model was 1.44%, 11.7%, and 19.5%, respectively. It is concluded that ANN model is best suitable for predicting the short and long-term energy consumption with very high accuracy, as compared to the CBR and MLR. The results discussed in this study can be advantageously used for enhancing the consumption of operational energy in existing as well as proposed buildings.

Assessing the Fire Properties of Various Surface Treatments on Timber Components in Ancient Chinese Buildings: A Case Study from the Xianqing Temple in Changzhi, Shanxi, China

Traditional and modern coatings play a key role in enhancing the fire resistance of ancient Chinese buildings. However, further comparative analysis is needed on the fire properties of the two coatings and their effects on different timber structural components. This study focuses on the main hall of the Shanxi Changzhi Xianqing Temple, a typical traditional column and beam construction built between the Song and Jin periods. Firstly, the combustion characteristics of various timber structural component samples with different surface treatments (traditional “Yi-ma-wu-hui” and modern flame retardants) were analyzed using cone calorimeter. Secondly, the fire development process of the Xianqing Temple building model was analyzed by a fire dynamics simulator (FDS), and the effect mechanism of different surface treatments on the burning process was further studied. The results show that the fire resistance of timber structural components is significantly improved after modern and traditional surface treatments. The traditional method is more effective in delaying the peak heat release rate and reducing the surface temperature during combustion, while the modern surface treatment significantly prolongs the ignition time of the timber structural components. The FDS results confirm that modern and traditional surface treatments significantly improve the fire resistance of the building, delaying the flashover time by about 300 s, with no collapse occurring within 800 s. In addition, the fire resistance of buildings after traditional surface treatment is better compared to traditional methods. The above research results can provide direct data support for the selection and optimization of fireproof coatings and treatment methods for ancient buildings.

Transformative social work education through a joint MSW programme: from participatory action research to policy and practice innovation in China

ABSTRACT In China, social work was reestablished in 1988. Despite the number of social work programmes increased very fast throughout the 1990s, there was a huge need for social work-trained educators. The Hong Kong Polytechnic University and Peking University launched the MSW (China) Programme in 2000 to create a critical mass of social work educators who would assume future leadership in social work development in mainland China. Participatory action research was employed as a methodology to explore how social work education could transform students into catalysts for social work education and social development. Purposive sampling was adopted, and seven cohorts totaling 238 graduates who graduated from the programme were the participants in this research. Data collection and analysis were carried out throughout the action research process. We found the triple capacity building model to be a good pedagogy to foster transformation among participants, and to steer the knowledge co-creation, practice innovation, policy reform, and social change through social work education. In this paper, we will provide an overview of the background, vision, and mission, as well as the teaching pedagogy of the programme and a case study to demonstrate how social work education has an impact on policy and practice innovation in China.

Research on Summer Environmental Thermal Comfort Model of Buildings in Hot Summer and Cold Winter Area

Research on Summer Environmental Thermal Comfort Model of Buildings in Hot Summer and Cold Winter Area

Transcending disciplinary boundaries: Nature based solutions (NbSs) for Caribbean coastal resilience

Caribbean Small Island Developing States are experiencing multiple interconnected crises including susceptibility to natural hazards, increased vulnerability to climate change, biodiversity loss and reduced accessibility to development funding. Nature-based Solutions can provide an integrated approach to boost Caribbean coastal resilience as they are economic and practical solutions that capably balance competing interests. This study used a participatory method, Group Model Building, to engage stakeholders across varying disciplines and levels of decision-making to collaboratively identify the underlying causes of low Nature-based Solutions uptake regionally and potential intervention mechanisms. Six overarching factors contributed to low adoption, including data and knowledge limitations, a propensity towards hard engineering, low collaborative engagement, and limited implementation capacity. Some areas identified for intervention were comprehensive data collection, reducing uncertainty surrounding the implementation of Nature-based Solutions, effective collaboration, using case studies and best practices, and knowledge translation and dissemination. The outputs demonstrated a critical role for the engineer in mainstreaming Nature-based Solutions through adopting a systematic and holistic design approach. Concomitant to this is the need for engineers to operate successfully within transdisciplinary teams to enhance transformative communication skills and ensure that solutions not only minimize environmental impact but also support the conservation of biodiversity and ecosystem services.

Development of novel computational models based on artificial intelligence technique to predict liquids mixtures separation via vacuum membrane distillation

The fundamental objective of this paper is to use Machine Learning (ML) methods for building models on temperature (T) prediction using input features r and z for a membrane separation process. A hybrid model was developed based on computational fluid dynamics (CFD) to simulate the separation process and integrate the results into machine learning models. The CFD simulations were performed to estimate temperature distribution in a vacuum membrane distillation (VMD) process for separation of liquid mixtures. The evaluated ML models include Support Vector Machine (SVM), Elastic Net Regression (ENR), Extremely Randomized Trees (ERT), and Bayesian Ridge Regression (BRR). Performance was improved using Differential Evolution (DE) for hyper-parameter tuning, and model validation was performed using Monte Carlo Cross-Validation. The results clearly indicated the models’ effectiveness in temperature prediction, with SVM outperforming other models in terms of accuracy. The SVM model had a mean R2 value of 0.9969 and a standard deviation of 0.0001, indicating a strong and consistent fit to the membrane data. Furthermore, it exhibited the lowest mean squared error, mean absolute error, and mean absolute percentage error, signifying superior predictive accuracy and reliability. These outcomes highlight the importance of selecting a suitable model and optimizing hyperparameters to guarantee accurate predictions in ML tasks. It demonstrates that using SVM, optimized with DE improves accuracy and consistency for this specific predictive task in membrane separation context.

Assessing and mitigating building fire risk based on the scenario clusters in Bangladesh

The increasing occurrence of fire hazards in residential buildings in Bangladesh has led to severe property damage, physical injuries, and fatalities, underscoring a critical research gap in effective fire risk management. This study aims to address this gap by identifying key fire risks and proposing mitigation strategies using BIM technology. The research employs a comprehensive survey and scenario clustering methodology to identify these risks and assess the efficacy of various fire-resistant materials. Data collection involved questionnaires distributed to 200 respondents, including building owners, occupants, engineers, and firefighters. These clusters are based on the metrics of fatality numbers and property damage to quantify fire risk. Findings highlight the significant role of wall and ceiling lining materials (RII = 0.919) in contributing to fire risk. Further analysis was conducted using a residential building model developed in Autodesk Revit and simulated in Pyrosim software, assessing various lining materials. Quantitative results demonstrated that fire brick is the most effective material, exhibiting the lowest heat release rate (968 kW/m2), minimum adiabatic surface temperature (28°C), minimum heat transfer coefficient (1.75 W/m2/K), and lowest surface burning rate (7.5e−08 Kg/m2/s). These results suggest that fire brick is the optimal choice for enhancing fire resistance in residential buildings.