This study develops a structured framework for the quantitative assessment of supplier-related risk in organizational supply networks. The proposed methodology is based on the Action Priority (AP) concept from Failure Mode and Effects Analysis (FMEA), which evaluates risk using three factors: Severity (S), Occurrence (O), and Detectability (D). Based on expert assessments and AP decision matrices, individual suppliers are classified into three risk categories: Low (L), Medium (M), and High (H). To enable a more rigorous analytical representation of these qualitative assessments, the risk categories are modeled using triangular fuzzy numbers (TFNs). The fuzzy values associated with individual suppliers are aggregated using the fuzzy arithmetic mean operator and subsequently defuzzified through the centroid method. After normalization, a single quantitative indicator—the Overall Supplier Risk Index—is obtained, providing insight into the company’s overall dependence on its supplier base. The proposed framework is demonstrated through a case study of a furniture manufacturing company in the wood-processing industry involving 39 strategically important suppliers. The results indicate that the analyzed company belongs to the second risk priority level, corresponding to a low overall supply risk exposure. The developed model enables the transformation of qualitative expert evaluations into a single analytical indicator, thereby supporting managerial decision-making in supplier risk monitoring and supply strategy development.
The escalating volume of municipal solid waste has intensified the need for intelligent waste management systems capable of improving operational efficiency, classification accuracy, and sustainability. In recent years, the integration of Internet of Things technologies, deep learning algorithms, and sensor-based monitoring has significantly transformed conventional waste collection and sorting practices. In this study, an intelligent waste management framework was proposed and comparatively evaluated against twelve contemporary smart waste management systems reported in the literature. The proposed architecture integrates a Raspberry Pi 3 embedded platform, You Only Look Once version 8 (YOLOv8) deep learning models for real-time waste classification, and ultrasonic bin-fill sensors for monitoring container capacity, enabling automated lid operation, and supporting optimized waste collection scheduling. A comprehensive comparative analysis was conducted across multiple performance dimensions, including classification accuracy, system responsiveness, scalability, deployment cost, and operational efficiency. Experimental evaluation demonstrates that the deep learning–driven framework achieved high real-time classification accuracy while maintaining low computational overhead on resource-constrained edge devices. In addition, the incorporation of bin-fill sensing and automated actuation enhanced system responsiveness and supported data-driven collection planning, thereby reducing unnecessary collection trips and operational costs. The findings highlight the significant potential of combining advanced deep learning algorithms with sensor-based Internet of Things infrastructures to develop sustainable, intelligent, and cost-effective waste management ecosystems. These insights provide a foundation for future research aimed at enhancing intelligent waste infrastructure and supporting environmentally sustainable urban development.
Industrialization and population growth pose significant environmental issues, particularly in water quality, making many sources unsuitable for domestic use. Natural organic compounds and metal nanoparticles (NPs) are used as wastewater adsorbents. The current research investigated the adsorption kinetics, isotherms, and reusability of the manganese oxide NPs synthesized from star anise (SA) (Illicium verum) extract (MnO@SE) to aid in the creation of environmentally friendly water purification solutions. MnO@SE was prepared with SA extract and manganese acetate (II) tetrahydrate solution. The green-synthesized biosorbent was characterized employing methods including Fourier transform infrared spectroscopy, X-ray diffraction (XRD), energy-dispersive X-ray spectroscopy (EDX), and scanning electron microscopy (SEM). These evaluations offered good information on surface shape and surface-available functional groups. The influences of pH, adsorbent dosage, ion concentration, and contact time on metal ion adsorption were all examined. The results revealed that model solutions with a pH of 2.0, a biosorbent dosage of 0.8 g/L, an initial concentration of 25 mg/L, and a contact time of 50 minutes produced the best removal efficiency (96.34% for Cr(VI) and 87.01% for Pb(II)). The adsorption processes of both metal ions occurred in a multilayer fashion on the heterogeneous surface of the biosorbent through diffusion kinetics, according to the isotherm and kinetic findings. The adsorption process is endothermic and spontaneous, according to thermodynamic analysis. The study revealed that the green-synthesized MnO@SE effectively removed 96.34% Cr(VI) and 87.01% Pb(II) under optimal conditions, promoting eco-friendly water purification through multilayer, endothermic, spontaneous, and diffusion-driven adsorption.
Green growth has emerged as a crucial central policy objective for reconciling economic performance with environmental sustainability. This study investigates the determinants of green growth in OECD countries over the period 1990−2022. Specifically, it evaluates the roles of environmental policies (EPS), research and development (R&D), and institutional quality, particularly the Rule of Law (IQRL), in shaping economic green growth. To address cross-country interdependence and structural heterogeneity, we employ the Common Correlated Effects Estimator with a Cross-Sectionally Augmented Autoregressive Distributed Lag (CS-ARDL) framework. The approach appropriates the dataset exhibiting unobserved common factors, cross-sectional dependence, and mixed order of integration. The model effectively manages cross-sectional dependence, accommodates heterogeneous dynamics, and addresses mixed stationarity. It therefore provides reliable estimates of both short-run and long-run effects and is well suited for modern applied panel data analysis. The empirical results show that an inclusive strategy promoting technological innovation, strong environmental governance, and following the Rule of Law is essential for advancing green growth in advanced economies. Trade openness, GDP per capita, and population growth are found to be significant positive drivers. In contrast, renewable energy consumption (REC) exerts a negative effect, suggesting the presence of short-term adjustment costs associated with the energy transition. Overall, these findings highlight the importance of coordinated environmental governance, sustained R&D support, and strong institutional frameworks in advancing green growth. Policymakers should prioritize targeted R&D investments, strengthen environmental policy design, and enhance the institutional frameworks that support the ecological transformation.
Energy-efficient path planning for multi-Unmanned Aerial Vehicle (UAV) data-collection missions requires balancing trajectory efficiency, energy consumption, and workload distribution among UAVs. This study presents a controlled computational evaluation of three routing paradigms: random assignment, Greedy nearest-neighbor routing, and Greedy + K-means clustering. The evaluation is conducted using a mission-level energy model that incorporates propulsion energy and mission-phase components, including take-off, hovering, sensing, communication, and landing. Simulation experiments were performed using fleets of 1–10 UAVs serving 100 Points-of-Interest (PoIs) under two spatial deployment scenarios: a structured grid layout and a spatially heterogeneous random layout. Each configuration was executed over 20 independent episodes to ensure statistical robustness. The results demonstrate that routing structure significantly influences geometric mission efficiency. In the propulsion-dominated regime (U $\geq$ 5 under random PoI layouts), Greedy + K-means clustering reduces mission travel distance by approximately 11.6–24.5% compared with Greedy routing, corresponding to an energy reduction of approximately 4.6–10.5%. In contrast, under the phase-dominated regime, where fixed mission-phase energy dominates the total energy budget, performance differences between routing strategies remain below 5%. Statistical analysis further confirms large practical differences in geometric performance across algorithms ($\eta^2$ $>$ 0.86). These findings indicate that routing strategy selection should depend on mission scale and spatial characteristics rather than assuming universal optimality. Greedy routing performs effectively in small or spatially structured deployments, whereas Greedy + K-means clustering provides greater robustness and scalability in larger or spatially heterogeneous missions.
Deep neural network-based English handwriting recognition has revolutionised the security or verification system of today; nevertheless, there are certain risks that are inevitable. This paper examines the status of the present technologies in the handwriting recognition systems, and more particularly, by the various deep neural network architectures. It also evaluated common cybersecurity risks such as data poisoning, model inversion, and adversarial attacks, which can be devastating to such systems, as well as common privacy and ethical issues. The potential regulatory compliance and mitigation measures that can be taken to avert these risks and hurdles are also addressed in detail, with requisite emphasis being made on the future outlook of a more secure handwriting recognition system.
Urban public transport systems are required to respond to pronounced temporal variations in passenger demand driven by calendar effects, weather conditions, and evolving mobility patterns. Reliable short-term demand forecasts have therefore become an important role in supporting operational planning and service management in large-scale transit systems. This study examines the daily ridership dynamics of the Transjakarta bus rapid transit system and evaluates the forecasting performance of three modeling approaches: seasonal autoregressive integrated moving average with exogenous variables (SARIMAX), multilayer perceptron (MLP), and a dynamic moving-window model. The analysis is based on 851 daily observations from January 1, 2023 to April 30, 2025, with rainfall, working days, and national holidays included as exogenous variables. Each model is estimated using a training dataset and evaluated on a hold-out test set over a 30-day forecasting horizon. Forecast accuracy is assessed using the mean absolute percentage error (MAPE). The results indicate that the MLP model achieves the highest forecasting accuracy, with a MAPE of 8.547%, while SARIMAX and the dynamic model yield higher error levels of 33.345% and 37.754%, respectively. The findings suggest that non-linear modeling approaches are better suited to capturing the complex and irregular demand patterns observed in daily urban bus ridership data. The study provides empirical evidence that can support short-term planning and demand-aware operational decision-making in urban public transportation systems.
Heavy metals tracing and magnetic susceptibility are generally used as a proxy indicator of pollution in various depositional environments. This research focused on tracing the significant record of pollution of the Muntingia calabura tree to understand the sensitivity in recording pollution and mining production. The area of this research is a hydrocarbon mining site of Wonocolo Geopark in the Bojonegoro, East Java Indonesia. The samples were taken both polluted and non-polluted leaf and bark, from 20 sampling points. Polluted leaves then were characterized by the existence of elevated levels of Pb, Fe, Cu and Zn. The average magnetic susceptibility of leaves increases from 0.86 × 10$^{-8}$ m$^3$/kg in non-polluted samples to 13.55 × 10$^{-8}$ m$^3$/kg in polluted samples and in the same way, increasing of magnetic susceptibility was also seen in the barks, from an average of 0.21 × 10$^{-8}$ m$^3$/kg in non-polluted sample, to 2.55 × 10$^{-8}$ m$^3$/kg in polluted sample. The pattern of magnetic susceptibility on leaves and barks at each sampling point is also the same as the pattern of hydrocarbon production which is related to the level of pollution in the area. The increase of magnetic susceptibility in polluted leaves and barks is thought caused by input of magnetic minerals and heavy metals from the fly ash of diesel engines used for hydrocarbon mining process. The heavy metal concentration has the average of Pb (0.070 ppm), Fe (13.322 ppm), Cu (8.434 ppm), and Zn (11.668 ppm). This value has exceeded the threshold of heavy metals content and have a worst impact on health and the environment. Based on Pollution Load Index (PLI) calculations, the most of areas affected by heavy metal pollution in very high and extremly high levels with the highest pollutants input are Fe, Cu, Zn and Pb respectively.
The Fallujah Cement Plant constitutes a cornerstone of reconstruction efforts in Al-Anbar Province, yet it simultaneously represents one of the largest stationary sources of air pollution in the region. This study presents the first integrated assessment of ambient air quality impacts from a major Iraqi cement facility by combining field measurements with the American Meteorological Society/Environmental Protection Agency Regulatory Model (AERMOD). Concentrations of seven pollutants—sulfur dioxide (SO$_2$), nitrogen dioxide (NO$_2$), carbon monoxide (CO), carbon dioxide (CO$_2$), total suspended particulates (TSP), particulate matter $\leq$ 10 $\mu$m (PM$_{10}$), and fine particulate matter $\leq$ 2.5 $\mu$m (PM$_{2.5}$)—were monitored at three receptor sites surrounding the plant. Results revealed that measured concentrations consistently exceeded model predictions, particularly for CO$_2$ (+221%) and CO (+441%). Field data indicated exceedances of Iraqi national standards and World Health Organization (WHO) guidelines by up to 23-fold for SO$_2$ and 12-fold for PM$_{2.5}$. Spatial analysis confirmed that pollutant plumes predominantly extend southeastward under prevailing northwesterly winds, with the highest risks observed in nearby residential complexes located 2 km downwind. Overall, the findings demonstrate that the Fallujah Cement Plant poses significant public health risks, underscoring the urgent need for advanced emission-control technologies and the establishment of vegetative buffer zones to mitigate environmental and health impacts.
This study examines the relationship between seawater quality in the vicinity of the Labuan 2 Steam Power Plant (SPP) and the distribution of plankton and benthos communities. Seawater quality was assessed in accordance with the Minister of Environment Decree No. 115/2003, while biodiversity was evaluated using the Shannon–Wiener diversity index, the uniformity index, and the Simpson dominance index. Sampling was conducted at seven monitoring points during the dry season in July 2021. The results indicated that seawater quality at all sampling locations met the established quality standards. A total of 27 phytoplankton species were identified, with Skeletonema consistently observed as the dominant genus across all sites. The phytoplankton community exhibited high uniformity, moderate diversity, and moderate dominance. Zooplankton analysis identified 17 species, dominated by Temora (Copepoda), reflecting its role as a key primary consumer linking phytoplankton to higher trophic levels. Zooplankton communities showed high uniformity, low dominance, and moderate diversity. In addition, ten benthic species were recorded, with Arenicola sp. as the dominant taxon. The benthos community was characterized by moderate uniformity, low dominance, and relatively high diversity. Overall, the findings indicate that the waters surrounding the Labuan 2 SPP remain ecologically balanced, with plankton and benthos communities supporting stable marine food web structures.
The rapid diffusion of artificial intelligence (AI) into decision-making processes has raised critical questions about how AI reshapes human behavior, judgment, and responsibility. While existing studies often emphasize technical performance, less attention has been given to the behavioral dynamics that emerge when humans interact with AI-supported systems. This study addresses this gap by proposing an integrated Strengths, Weaknesses, Opportunities, and Threats–Analytic Hierarchy Process–Technique for Order Preference by Similarity to an Ideal Solution (SWOT–AHP–TOPSIS) framework to systematically evaluate the behavioral impact of AI-assisted decision making. First, key behavioral factors are identified using SWOT analysis, where strengths and weaknesses represent internal human behavioral traits, and opportunities and threats capture external and contextual influences related to human–AI interaction. These factors are then weighted using AHP based on expert judgments, with consistency checks ensuring methodological reliability. Finally, TOPSIS is applied to rank three AI-assisted decision scenarios—human-dominant, shared-control, and AI-dominant decision making—according to their overall behavioral performance. The results indicate that behavioral weaknesses, such as over-reliance on AI and reduced critical thinking, exert the strongest influence on decision quality. Among the evaluated scenarios, human-dominant decision making achieves the highest closeness coefficient, followed by shared-control and AI-dominant scenarios. Sensitivity analysis confirms the robustness of these rankings under reasonable variations in criterion weights. Methodologically, this study demonstrates that the SWOT–AHP–TOPSIS approach, traditionally used in strategic and operational research, can be effectively adapted to behavioral and socio-technical contexts. Substantively, the findings highlight the importance of preserving human cognitive agency in AI-assisted environments. The proposed framework offers a practical and theoretically grounded tool for researchers, designers, and policymakers to assess and guide the behavioral implications of AI-supported decision systems.
For Authors
For Reviewers
For Editors
Open Access
