Landslide crack identification is crucial for risk management and mitigation, yet challenges like limited data and the lack of model interpretability hinder progress. This study proposes a deep learning-based solution for classifying landslide cracks using the ResNet50 architecture integrated with a squeeze-and-excitation (SE) attention mechanism, the transfer learning (TL) strategy, and a model interpretability approach. We introduced the SE module to enhance the model’s ability to capture subtle crack features, and, in the TL framework, we first pretrained the model on a large concrete crack dataset to learn general crack characteristics, then fine-tuned it on a landslide crack dataset. To further improve interpretability, we applied gradient-weighted class activation mapping (Grad-CAM) to visualize the areas of the image most influential to the model’s decisions. Our results demonstrated that ResNet50+SE outperforms both the standard convolutional neural network (CNN) and residual network with 50 layers (ResNet50), particularly in recall and F1 score, highlighting its superior ability to detect challenging cracks and improve overall landslide crack identification accuracy. Additionally, TL boosts performance, even with the limited availability of landslide crack data. Grad-CAM heatmaps provide valuable insights into the model’s focus and decision-making, enhancing transparency. This study also tackled data imbalance challenges. Overall, the proposed approach offers an effective, interpretable solution for landslide crack identification, enhancing accuracy and transparency for landslide early warning and risk assessment.
This article examined the pivotal role of environmental factors within the environmental, social, and governance (ESG) framework in shaping the development of Vietnam’s cement industry. It highlights how sustainability drivers influence environmental management practices and overall sectoral performance. Rising cement demand, evolving regulations, and the growing emphasis on sustainable construction have propelled the prioritization of environmental sustainability, facilitated by the adoption of robust environmental management systems (EMS). This environmental focus has resulted in tangible improvements in community health, employee well-being, stakeholder relations, regulatory compliance, and investor confidence, ultimately driving innovation, operational efficiency, and cost reduction. This industry-wide shift aligns with consumer preferences for eco-friendly construction materials and fosters a culture of sustainability across Vietnam’s cement sector. Notably, the industry has made significant progress in embracing cleaner production technologies, waste utilization, and green financing to drive its environmental sustainability agenda, underscoring the increasing prominence of ESG considerations. While the importance of ESG factors is widely recognized, a research gap persists in quantifying the specific contributions of the ESG pillars to overall ESG performance in the cement industry.
Geological hazards caused by unstable rocks, including rock collapse and fall, pose significant threats to both engineering productivity and the safety of residents, resulting in substantial economic losses. This study proposed a comprehensive CNN (Convolutional Neural Networks) classification recognition algorithm based on continuous wavelet analysis of microseismic monitoring data, which establishes a CNN recognition-based method for identifying risk levels of unstable rocks to predict their real-time collapse. To collect training data, fixed-base tests were conducted, and validation was performed using vibration data from freeze-thaw tests. Results of the study showed that the accuracy of the CNN classification recognition model trained with fixed-base test data reached 97.6%, and the per-second classification accuracy of vibration segments from freeze-thaw tests was above 86%. Furthermore, the study discussed the correlation between the calculated risk-level eigenvalues and safety coefficients. The CNN risk-level eigenvalues were found to be highly negatively correlated with the safety coefficients, with correlation coefficients as high as 0.8703. Finally, the study verified the superiority of the precursor identification of the risk-level evaluation method by comparing it with the safety coefficients.
To investigate the effect of Polyvinyl Alcohol (PVA) fiber incorporation on the erosion resistance of soil under different environmental conditions, this study selected fiber length and fiber content as variables and systematically compared the performance difference between fiber-reinforced soil and plain soil through dry and wet cycles, freeze-thaw cycles, sulfate erosion, and scouring tests. The fiber soil was found to have significant erosion resistance compared to the plain soil. The results showed that, for dry-wet cycling, the addition of PVA fibers effectively enhances the soil sample's resistance. Particularly, when the fiber content ranges from 0.5% to 0.8%, the soil sample's resistance to dry-wet cycles is significantly improved. Moreover, the influence of fiber content on enhancing the dry-wet cycling performance of the sample is more pronounced than that of fiber length. Shrinkage tests showed that the soil undergoes three distinct stages during the drying process: constant rate, deceleration, and residual phases. Under the action of freeze-thaw cycles, specimens containing PVA fibers demonstrate good deformation resistance and high unconfined compressive strength and do not appear to be damaged in the process of freezing and thawing, indicating good freeze-thaw resistance. Meanwhile, the effect of fiber length on the strength-loss rate is minor, suggesting that the freeze-thaw resistance of the specimens is relatively insensitive to fiber length. For the sulfate erosion resistance test, with an increase in fiber content, the strength of the specimen improves, and its strength loss rate is significantly reduced, whereas the influence of fiber length on the specimen remains limited. In the scouring test, soil samples show good scour resistance after fiber content, and the scour growth rate decreases with time.
In this study, we used monitoring data on six representative pollutants, meteorological data, and socioeconomic data from January 2018 to December 2023 to clarify the spatiotemporal characteristics of air pollution in Changchun, Jilin Province, China, and to establish a short-term prediction model. First, correlation analysis (CA) was used to identify the key factors affecting pollution characteristics. Then, wavelet analysis (WA) was applied to capture the major periodic coupling characteristics of the time series, thereby analyzing the variation law of air quality. The results showed that the six air pollutants in Changchun exhibited significant seasonal differences. The peak concentration of O3 occurred in summer, while the peaks of other pollutants were concentrated in winter and spring. The average air quality index (AQI) exhibited a downward, overall fluctuating trend, with PM2.5 and PM10 as the major pollutants. Urbanization rate and green space coverage area were identified as key social driving factors, while temperature, relative humidity, and wind speed were identified as key meteorological driving factors. Furthermore, WA revealed a coupling phenomenon between the major periods of the AQI sequence and the PM2.5 and PM10 sequences. Additionally, the low-frequency wavelet signals showed an overall downward trend, confirming that composite pollution is a typical feature of air pollution in Changchun, while the regional air quality is gradually improving. A short-term AQI forecasting equation based on ordinary least squares (OLS) was constructed. Unlike previous AQI prediction models that used only meteorological factors as independent variables, this model included "AQI data of the same period in the previous year" as a key independent variable. Verification showed that the model's fit outperformed that of traditional meteorologically driven models, yielding more accurate short-term AQI predictions for Changchun, driven by the following two major innovations. First, combined wavelet and CA comprehensively determined the periodic coupling characteristics between AQI and major pollutants, providing an efficient tool for analyzing the causes of composite pollution. Second, incorporating historical AQI data into the OLS prediction model mitigated the limitation of traditional models that rely solely on meteorological factors.
Bisphenol A (BPA) is an industrial chemical commonly used as a monomer or additive in the production of polycarbonate plastics, epoxy resins, and other polymeric materials. BPA contamination is primarily associated with plastic bottled water containers. BPA has been associated with a range of adverse health effects. This study aims to investigate the impact of BPA exposure on public health in relation to the consumption of bottled drinking water. The research was conducted in Tambak Wedi Village, Surabaya, from July to October 2024. A total of 96 respondents participated in this cross-sectional study, which included laboratory analysis of bottled water samples and a human health risk assessment. Among ten brands of bottled water consumed by respondents, Brand A exhibited the highest BPA concentration (maximum 0.099 mg/L), while Brand G showed the lowest (maximum 0.082 mg/L). BPA exposure through bottled water consumption was not found to be significantly associated with carcinogenicity, reproductive system disorders, or endocrine disruption. However, a significant association was observed with immunosuppressive disorders (p-value=0.008; r-value=0,151), with very weak positive correlation strength. The noncarcinogenic and carcinogenic health risks of BPA across all bottled water brands were within acceptable safety limits. Collaborative efforts from governmental bodies, industry stakeholders, and consumers are essential to minimize the potential health risks associated with BPA exposure.
The term net-zero is frequently used in the literature; it refers to a situation in which an organization removes emissions using science-based methods and simply neutralizes residual emissions to reach balance in a given period [1]. Net-positive, however, means more than just being neutral. It implies that the complete footprint must be examined in order to demonstrate measurable changes in the environment and/or society when compared to a baseline [2]. Net-zero and net-positive are the two major principles influencing the industrial sustainability paradigm. Industrial sustainability is best understood as a co-evolutionary process shaped by successive Industrial Revolutions (IRs), each of which has fundamentally restructured socio-technical and economic systems. Digitalization, cyber-physical systems, AI (artificial intelligence), and the IOT (internet-of-things), for instance, have enabled new approaches to improve industrial systems and use resources more efficiently. These technologies have facilitated predictive maintenance by monitoring events in real time and making data-driven decisions [3]; this, consequently, has resulted in reduced waste and more environmental efficiency. However, IR 1.0–3.0's approach to sustainability has been mostly incremental, with a focus on technical advancements for financial sustainability. Conversely, IR 4.0 could better expand its coverage to other sustainability pillars: environment and social. The current shift to IR 5.0 has, however, resulted in a more inclusive and human-centric attitude than previously observed. IR 5.0 emphasizes resilience, human well-being, and ethical governance, moving beyond the narrow operational and financial gain logic of past IRs and embedding sustainability more deeply within socio-
Heavy metal contamination in industrial areas poses significant environmental challenges due to the lack of early detection and control mechanisms. Typically, remediation actions are taken only once contamination reaches hazardous levels. In this study, we evaluated the distribution of heavy metals in two rivers in Johor Bahru, Malaysia, namely the Selangkah and Kim Kim Rivers, with particular emphasis on concentrations exceeding threshold limits. Our primary objective was to assess the degree of heavy metal pollution and highlight the influence of both point and non-point pollution sources. Key water quality parameters were analyzed, including heavy metal concentrations, electrical conductivity (EC), (pH), salinity (SAL), and dissolved oxygen (DO). The overall mean concentrations of potassium (K), iron (Fe), and magnesium (Mg) were highest in both rivers, with Mg exceeding the permissible limit of 150 mg/L in the Kim Kim River (Maximum concentration recorded: 412.63 mg/L). In the Selangkah River, mean concentrations of copper (Cu) (0.157 mg/L), Fe (4.591 mg/L), and nickel (Ni) (0.493 mg/L) significantly exceeded the maximum allowable limits for surface water set by the Malaysian Ministry of Health and the National Water Quality Standards (NWQS). Spatial distribution identified sampling points SL1, SL4, SL5, and SL6 along the Selangkah River as having the highest concentrations of most heavy metals, correlating with the dense concentration of industrial activities in those areas. The Contamination Index (Cd) and Heavy Metal Pollution Index (HPI) calculations confirmed that the Selangkah River is classified as highly polluted, while the Kim Kim River presents a moderate-to-high pollution status. Given that these rivers support smallholder agriculture and contribute to groundwater recharge, protecting and conserving their fragile ecosystems is essential to ensure long-term environmental sustainability and public health safety.