Artificial intelligence (AI) chips are the core hardware supporting the development of intelligent technologies and their technological advancements hold significant importance for national scientific innovation, industrial development, and economic growth. This study summarizes the global development trends of AI chips from three perspectives: cloud AI chips, edge AI chips, and neuromorphic chips. It analyzes the application demands of AI chips in China and reviews the current status and development trends of related industries and technologies from the aspects of chip design, manufacturing, and packaging and testing. Currently, AI chips manufactured in China have shortcomings in terms of performance, technologies, and supply chain, necessitating independent innovation and industrial collaboration. The development of these chips also encounters challenges such as high costs and long cycles, requiring stable financing channels and the accumulation of development experiences. Moreover, the AI chip sector in China suffers from a talent shortage, demanding improved training quality and better talent-retaining strategies. Accordingly, this study proposes development paths for China’s AI chip industry, including overcoming technical bottlenecks, accelerating industrialization, expanding internationalization, and implementing market support. Key measures include advancing technological innovation and major project development, fostering new chip architectures and open-source industry ecosystems, establishing technical standards, and promoting the integration of industry, education, and research. These efforts aim to drive the sustainable and high-quality development of China’s AI chip industry.

Stepwell is one of the unique medieval groundwater resource structures distributed across most of all the states in India. The main purpose of constructing stepped wells was to get access to the groundwater, but was also used for the storage of food, and as a summer shelter in medieval times. In tropical regions such as India, temperature is a major problem that causes thermal distress. Traditional architectural concepts were incorporated in the medieval Indian structures to achieve thermal comfort. Traditional architecture is optimization of building design to climate, adaptive crafts, and building techniques learnt over time and experience. The objective of this study is to interpret and evaluate the architectural and engineering aspects of stepwells pertaining to passive cooling techniques used in stepwells and the shelters associated with them. This paper reviews technical aspects pertaining to passive cooling in the design of stepwells, which has identified fundamental principles that can adopted for the building of various structures presently.

Tessellations have been widely used in architectural designs to create visually striking surfaces by repeating a small number of elements. While there has been extensive research on 2D tiling over the past 2000 years, this study focuses on the systematic development of a technique for the modular construction of 2D tiling with 3D texture surfaces using identical tiles, which is of great importance in architectural designs for cost-effective constructions through the mass production of repeating components. This study applies parametric geometrical modelling using Grasshopper scripting to generate a wide range of 3D reliefs on planar surfaces by repeating a single tile or a few different tiles. Based on the findings of this study, it is possible to arrange tiles with an identical 3D texture surface in multiple configurations, resulting in a range of 3D reliefs on tessellated surfaces that exhibit smooth transitions across adjacent tiles. A significant application of this technique is in producing stunning facades and other 3D surfaces using identical modules, offering affordable modular construction through repetition. The study also demonstrates the versatility of the technique by creating various attractive non-periodic 3D surfaces using triangular, square, hexagonal, or even non-regular tiles.

Interactive architecture designs enable spatial forms to respond to human activities by integrating technology, programming, and spatial experience. Most interactive programs respond to established actions in a uniform manner and do not take into account the individual emotional state of the user. However, the emotional states of children and adolescents with emotional/behavioral disorders play a crucial role in their interactions, and uniform spatial responses cannot be adapted to different people. This study developed an interactive architectural system for children with emotional/behavioral disorders based on individual emotion recognition and corresponding spatial transformation. The method proposed in this study enables individuals in certain emotional categories, such as extroverted, introverted, aggressive, and defensive states, to receive matching spatial responses. Using this individuality-oriented interactive system, we built a practical pavilion in a primary school in Tianjin. This study demonstrates how combining individual emotions and spatial variations opens up new design possibilities for interactive architecture.

In the past decade, there has been an increasing recognition of the role of computational design optimization in early-stage performance-based architectural design exploration. However, it remains challenging for designers to apply such optimization-based design explorations in practice. To address this issue, this paper introduces a design tool, called EvoMass, and an associated design method that facilitates design exploration for building massing typologies in performance-based design tasks. EvoMass is capable of offering architects design options reflecting performance-related building massing typologies for the design task, without necessitating advanced computational design skills. More importantly, it can provide architects with insights into the underlying performance implications, thereby enhancing early-stage performance-based design exploration. EvoMass and its associated design method overcome the limitation in the conventional typology-first-optimization-second design procedure adopted by most existing tools, and it promotes a typology-oriented design exploration method of using computational optimization in performance-based architectural design. To demonstrate the efficacy of EvoMass, case studies derived from architectural design studio tasks, incorporating daylighting, solar exposure, and subjective design intents, and the result of a user survey are presented, which highlights how EvoMass and the performance-based design optimization and exploration can enable architects to achieve a more performance-aware design.

Urban public spaces are pivotal to the welfare and prosperity of modern cities. Recognizing their importance, this research addresses the critical gap in understanding and enhancing the qualities of these spaces through advanced analytics, focusing on Tehran’s main traditional market, the Bazaar. A novel methodological framework combining Social Network Analysis (SNA), and Strengths, Weaknesses, Opportunities, and Threats (SWOT) analysis, supported by location-based social media reviews, was employed. This innovative approach assessed the Bazaar’s comfort, vitality, and safety, analyzing real-time public interactions and perceptions through social media data. The findings highlighted the Bazaar’s central role in Tehran’s urban landscape and identified the need for strategic design interventions. These interventions aimed to improve walkability, comfort, safety, and diversity, and have been successfully implemented, significantly enhancing the Bazaar’s quality and usability. This study not only advances urban studies and planning by providing a model for urban public space analysis but also underscores the value of social media data in urban analytics. The successful revitalization of Tehran’s Bazaar sets a precedent for enriching urban experiences and boosting city vitality through similar interventions in other urban spaces.

Numerous metropolises worldwide have adopted transit-oriented development (TOD) as a sustainable urban development tactic. And the node-place model is the most commonly used model for TOD typology. However, the spatial vitality, one of the most fundamental aspects of TOD, has not been systematically taken into account in the node-place model. Therefore, this article introduces vitality as a third dimension into node-place model and develops the node-place-vitality (NPV) model, aiming to comprehensively re-classify and re-evaluate TOD performance. The independent weight coefficient method is used to compute place values, node values, vitality values, and overall TOD degree. Moreover, the categorization of TOD typologies makes use of the K-means++ clustering approach. The model is employed in a case study in Wuhan, China. The results show that the TOD degree exhibits a spatial discrepancy, characterized as “high-medium-low” from the center to the periphery, not only in the whole city but also in the three towns. The geographical characteristics of TOD degree within station catchment area are revealed by the heat map. Five TOD types are identified and they present obvious spatial differentiation. The approach proposed in this paper furnishes urban planners and decision-makers with a scientific instrument for assisting TOD planning and strategies designing.