The ability of deep learning has been tested to learn graphical features for building-plan generation. However, whether the deeper space allocation strategies can be obtained and thus reduce energy consumption has still not been investigated. In the present study, we aimed to train a neural network by employing a characterized sample set to generate a residential building floor plan (RBFP) for achieving energy reduction effects. The network is based on Pix2Pix, including two sub-models: functional segmentation layout (FSL) generation and building floor plan (BFP) generation. To better characterize the energy efficiency, 98 screened floor plans of Solar Decathlon (SD) entries were labeled as the sample set. The data augmentation method was adopted to improve the performance of the FSL sub-model after the preliminary testing. Three existing residential buildings were used as cases to observe whether the network-generated RBFP gained the effect of decreasing energy consumption with decent space allocation. The results showed that, under the same simulation settings and building exterior profile (BEP) conditions, the function arrangement of the generated scheme was more reasonable compared to the original scheme in each case. The annual total energy consumption was reduced by 13.38%, 12.74%, and 7.47%, respectively. In conclusion, trained by the sample set that characterizes energy efficiency, the RBFP generation network has a positive effect in both optimizing the space allocation and reducing energy consumption. The implemented data augmentation method can significantly improve the network’s training results with a small sample size.

The elderly community day care model is an emergent solution to the aging problem influenced by the Eastern perspective of family in China. Due to the structural problem of spatial disorder in most of China’s urban built-up areas, the planning and construction of elderly day care centers (EDCCs) is facing great challenges. This study aims to comprehensively compare the spatial distribution and accessibility measurement methods for elderly residents and EDCCs in typical Chinese urban built-up areas based on the accessibility theory and spatial analysis methods from the community living circle perspective. The results show that different spatial distribution analysis methods have their own emphases and limitations, requiring comprehensive application in practice. The potential model method is most suitable for the accessibility measurement in this scenario. The threshold setting of service distance for the urban built-up areas public service facilities in the current Chinese standard needs to be further optimized. The existing EDCCs suffer from serious quantity deficiencies and misplaced supplies in the region. These findings can reveal the EDCCs distribution characteristics of typical Chinese urban built-up areas and provide new insights for urban planners and policy makers who are assessing the equity and efficiency of public service facilities.

In recent years, as machine learning has been widely studied in the field of architecture, scholars have demonstrated that computers can be used to learn the graphical features of building façade generation. However, existing deep learning in façade generation has yet to generate only a single façade, without comprehensive generation of five façades including the roof. Moreover, most of the existing literature has utilized the Pix2Pix algorithm for façade generation experiments, failing to attempt to replace the original generator in Pix2Pix with a different generator for experiments. This study addresses the above issues by collecting and filtering entries from the international Solar Decathlon (SD competition) to obtain a data set. Subsequently, a low-rise residential building façade generation model based on the Pix2Pix neural network was constructed for training and testing. At the same time, the original U-net generator in Pix2Pix was replaced with three different generators, U-net++, HRNet and AttU-net, for training and test results were obtained. The results were evaluated from both subjective and objective aspects and it was found that the AttU-net generative network showed the best comprehensive generation performance for such façades. HRNet is acceptable if there is a need for fast training and generation

Street design quality and socio-economic factors jointly influence housing prices, but their intertwined effects and spatial variations remain under-quantified. Housing prices not only reflect residents’ neighborhood experiences but also stem from the spillover value of public streets perceived and used by different users. This study takes Tianjin as a case and views the street environment as an immediate experience proxy for through-travelers, combining street view images and crowdsourced perception data to extract both subjective and objective indicators of the street environment, and integrating neighborhood and location characteristics. We use Geographical-XGBoost to evaluate the relative contributions of multiple factors to housing prices and their spatial variations. The results show that incorporating both subjective and objective street information into the Hedonic Pricing Model (HPM) improves its explanatory power, while local modeling with G-XGBoost further reveals significant heterogeneity in the strength and direction of effects across different locations. The results indicate that incorporating both subjective and objective street information into the HPM enhances explanatory power, while local modeling with G-XGBoost reveals significant heterogeneity in the strength and direction of effects across different locations. Street greening, educational resources, and transportation accessibility are consistently associated with higher housing prices, but their strength varies by location. Core urban areas exhibit a “counterproductive effect” in terms of complexity and recognizability, while peripheral areas show a “barely acceptable effect,” which may increase cognitive load and uncertainty for through-travelers. In summary, street environments and socio-economic conditions jointly influence housing prices via a “corridor-side–community-side” dual-pathway: the former (enclosure, safety, recognizability) corresponds to immediate improvements for through-travelers, while the latter (education and public services) corresponds to long-term improvements for residents. Therefore, core urban areas should control design complexity and optimize human-scale safety cues, while peripheral areas should focus on enhancing public services and transportation, and meeting basic quality thresholds with green spaces and open areas. Urban renewal within a 15 min walking radius of residential areas is expected to collaboratively improve daily travel experiences and neighborhood quality for both residents and through-travelers, supporting differentiated housing policy development and enhancing overall quality of life.

With the extensive application of data analysis in various parts of the landscape design process, Building Information Modeling (BIM), City Information Modeling (CIM), and Landscape Information Modeling (LIM) all aim to achieve key data sharing and collaboration in the whole cycle and promote the development of smart cities. Landscape element indicators are complex, diverse, and difficult to quantify, which is one of the reasons for the slow development of LIM. However, with the development requirements of LIM, quantifying landscape elements and transforming landscapes into digital landscape information has become a hot spot in the landscape design industry. Landscape parametric design aims to transform the design elements into quantifiable parameters, obtain the design scheme by changing the value of the parameters, and form the design results based on digital information. It is one of the foundations of LIM. Based on the Rhino + Grasshopper (R+G) platform, this study takes Shuixizhuang Park as an example and establishes the parametric design approach for the park entrance. The approach involves several steps: (1) Confirming the boundary and key points of the park to prepare the basic data for parametric design. (2) Utilizing the Physarealm Algorithm Method (PAM) to simulate crowd paths, the Site Slope Algorithm Model (SSAM) to analyze terrain slope, the Road Intersection Algorithm Model (RIAM) to determine the scope of the park entrance area based on the national and local design codes, and the Crowd Density Algorithm Model (CDAM) and Crowd Convenience Algorithm Model (CCAM) to analyze the density and convenience of the crowd to preliminarily confirm the park entrance. (3) Meeting the basic requirements of the crowd and vehicle gathering and spread by using the Square Area Review Algorithm Model (SARAM) and Parking Lot Review Algorithm Model (PLRAM) to recheck the site area of the park entrance square and park lot to optimize the park entrance. This approach constructs several site analysis models based on the R+G platform and introduces PAM to analyze crowd activity paths, proposing a landscape parametric design method that integrates crowd activity and landscape design requirements. Compared with the classical design, the landscape parametric design derived from the comprehensive data analysis reduces human interference, is more scientific and practical, and better meets the requirements of people entering the park. The approach also provides ideas for other landscape parametric site selections. By adjusting the values of element parameters, the approach can also be applied to the site selection and design of other landscapes.

Understanding and recognizing urban morphology evolution is a crucial issue in urban planning, with extensive research dedicated to detecting the extent of urban expansion. However, as urban development patterns shift from incremental expansion to stock optimization, related studies on meso- and microscale urban morphology evolution face limitations such as insufficient spatiotemporal data granularity, poor generalizability, and inability to extract internal evolution patterns. This study employs deep learning and meso-/microscopic urban form indicators to develop a generic framework for extracting and describing the evolution of meso-/microscale urban morphology. The framework includes three steps: constructing specific urban morphology datasets, semantic segmentation to extract urban form, and mapping urban form evolution using the Tile-based Urban Change (TUC) classification system. We applied this framework to conduct a combined quantitative and qualitative analysis of the internal urban morphology evolution of Binhai New Area from 2009 to 2022, with detailed visualizations of morphology evolution at each time point. The study identified that different locations in the area exhibited seven distinct evolution patterns: edge areal expansion, preservation of developmental potential, industrial land development pattern, rapid comprehensive demolition and construction pattern, linear development pattern, mixed evolution, and stable evolution. The results indicate that in the stock development phase, high-density urban areas exhibit multidimensional development characteristics by region, period, and function. Our work demonstrates the potential of using deep learning and grid classification indicators to study meso-/microscale urban morphology evolution, providing a scalable, cost-effective, quantitative, and portable approach for historical urban morphology understanding.