Bacterial cellulose (BC), with its self-organizing fiber networks, provides a compelling model for sustainable and bio-inspired material design. This study introduces a code-driven simulation of BC growth using the Diffusion-Limited Aggregation (DLA) algorithm, implemented in JavaScript to replicate the stochastic processes of branching, density formation, and radial expansion. The model integrates nutrient diffusion and Brownian motion principles, producing networks that mirror experimental features such as nodal density and branching angles, while also revealing emergent behaviors, including loop formation, not easily observed in physical samples. Laboratory cultivation of BC was subsequently conducted to compare and validate the computational results, highlighting both the fidelity and the exploratory capacity of simulation. Quantitative and qualitative analyses demonstrate that DLA can approximate biological fiber organization while offering new insights for design applications. This approach bridges microbial processes and computational design, suggesting applications in regenerative architecture where ecological intelligence and adaptive material systems are prioritized.

The planning of safe assembly areas in disaster and emergency management processes is of paramount importance for ensuring public safety and minimizing risks. Particularly in regions undergoing rapid urbanization, the accurate identification of such areas is critical to mitigating chaos during disasters and expediting the post-disaster recovery process. This study evaluates the adequacy of existing emergency assembly areas within the boundaries of Karaman Neighborhood in Nilüfer District, Bursa, in terms of population density and accessibility. Furthermore, it proposes new assembly areas where deemed necessary. The study employs a combination of Voronoi diagrams and space syntax methods. In the initial phase, a correlation between population density data and urban voids within Karaman Neighborhood was established. Using Voronoi diagrams, the influence areas of these voids were delineated based on appropriate accessibility distances. This analysis enabled the assessment of how effectively the urban voids accommodate the current population density and their spatial proximity to residential zones. In the subsequent phase, the accessibility levels of the proposed and existing assembly areas were analyzed through space syntax methods. The findings indicate that the existing assembly areas in Karaman Neighborhood are insufficient in certain regions in terms of both population density and accessibility. The results derived from the Voronoi diagrams underscore the necessity of additional assembly areas, particularly in densely populated zones. Spatial analysis further reveals significant disparities in the accessibility levels of the existing urban voids, highlighting the need for their enhancement to meet the requirements of disaster preparedness effectively. The use of these two methods together in determining disaster assembly areas constitutes the novelty of the study.

Air pollution is a significant concern in Izmir, the third-largest city in Türkiye, with adverse impacts on public health and urban quality of life. Leveraging Internet of Things (IoT) monitoring, this study forecasts PM₁₀ and SO₂ using machine learning, deep learning, and time-series models over 1996-2024. Our main contribution is a controlled, city-scale evaluation that isolates the role of extreme-value handling by creating three preprocessing variants for each pollutant (keep, cap, remove) and benchmarking four model families (SARIMA, SVR, LSTM, xLSTM) under a leakagefree, time-ordered train/validation/test protocol. To our knowledge, this is the first Izmir study that quantifies how peak treatment interacts with model choice to affect forecast accuracy, and that reports results in a way that directly supports operational planning. Results show that for PM₁₀, SARIMA often attains the lowest errors when peaks are capped or removed, while xLSTM provides competitive accuracy across all strategies. For SO₂, xLSTM consistently outperforms traditional methods under capped and kept conditions, yielding lower errors and higher explanatory power. By combining IoT-based monitoring with a rigorous modeling method and explicit sensitivity to peak processing, the study offers actionable guidance for air-quality management, informing traffic control, industrial regulation, and urban design decisions in Izmir.

As the effects of the Industrial Revolution have significantly contributed to global climate change and the studies reveal that by 2050, the world’s energy demand will double while fossil fuel reserves are projected to deplete. To mitigate the risks associated with exceeding the 1.5°C global warming threshold, international organizations, including the United Nations, as well as governments and NGOs, are implementing strategies aimed at achieving carbon neutrality by mid-century. One such approach involves transitioning from the “take-make-waste” model of the linear economy to the circular economy (CE), which operates on principles of “make-use-reuse-recycle.” The AECO (Architecture-Construction-EngineeringOperation) industry, a key driver of global economic activity, plays a central role in this transformation by rethinking and restructuring the AECO industry to align with CE principles. This study examines the intersection of circular economy principles and the built environment through the lens of Industry 4.0. A systematic literature review combining descriptive and bibliometric analysis, covering publications from 2016 to 2024 from Scopus and Web of Science (WoS), was conducted using keywords related to CE, the built environment, and digital technologies. Structuring the review is processed through PRISMA and selecting the literature is strengthened by Quality Assessment (QA). The analysis, performed with VOSviewer, highlights key trends, thematic connections, and potential areas for future research. Continuous and significant increase in number of publications obtained since 2016. United Kingdom, Netherlands and Italy were the leading countries in this research area. “Sustainability“ is the pioneering journal in this field with highest number of publications and citations. Digital technologies such as building information modeling (BIM), blockchain, and the Internet of Things (IoT) have emerged as powerful tools to support CE implementation. By enabling
resource optimization, reducing waste, and enhancing lifecycle assessments, these technologies facilitate the adoption of sustainable practices within the AECO sector. These studies underscore the importance of advancing frameworks like life cycle assessment (LCA) and cradle-to-cradle (C2C) to support circular transitions in the built environment. Furthermore, the results demonstrate the increasing momentum and global interest in leveraging digitalization to achieve sustainability goals in the sector. This study offers valuable insights for academics, industry professionals, and policymakers by mapping the current state of CE research in the AECO industry and identifying emerging trends. It emphasizes the need for compr hensive frameworks and strategic roadmaps to enable the circular transition of the AECO sector. Future research should focus on deepening interdisciplinary collaboration and integrating CE with related sustainability concepts to enhance understanding and accelerate implementation. By doing so, the AECO industry can make significant contributions toward global carbon  neutrality and sustainable development goals.