Hyperparameter search was found not making good use of compute resources as surrogate-based optimizers consume extensive memory and demand long set-up time. Meanwhile, projects running with fixed budgets require lean tuning tools. The current study presents Bounding Box Tuner (BBT) and conducts tests of its capability to attain maximum validation accuracy while reducing tuning time and memory use. The project team compared BBT with Random Search, Gaussian Processes for Bayesian Optimization, Tree-Structured Parzen Estimator (TPE), Evolutionary Search and Local Search to decide on the optimum option. Modified National Institute of Standards and Technology (MNIST) classification with a multilayer perceptron (0.11 M weights) and Tiny Vision Transformer (TinyViT) (9.5 M weights) were adopted. Each optimizer was assigned to run 50 trials. During the trial, early pruning stopped a run if validation loss rose for four epochs. All tests applied one NVIDIA GTX 1650 Ti GPU; the key metrics for measurement included best validation accuracy, total search time, and time per trial. As regards the perceptron task, BBT reached 97.88\% validation accuracy in 1994 s whereas TPE obtained 97.98\% in 2976 s. Concerning TinyViT, BBT achieved 94.92\% in 2364 s, and GP-Bayesian gained 94.66\% in 2191 s. It was discovered that BBT kept accuracy within 0.1 percentage points of the best competitor and reduced tuning time by one-third. The algorithm renders the surrogate model unnecessary, enforces constraints by design and exposes solely three user parameters. Supported by the evidence of these benefits, BBT was considered to be a practical option for rapid and resource-aware hyperparameter optimization in deep-learning pipelines.
In light of the European Union’s 2050 decarbonization objectives, a fundamental transformation of urban energy systems is required—characterized by decentralization, decarbonization, and digitalization. Within this context, the Renewable Energy Community (REC) model has been identified as a pivotal mechanism for enabling the integration and equitable sharing of locally generated renewable energy, while simultaneously delivering environmental, social, and economic co-benefits. A systemic and place-based approach has therefore been proposed, in which the interactions among buildings, neighborhoods, and communities are holistically considered in the design and governance of urban energy systems. The operationalization of RECs has been shown to rely heavily on the deployment of digital technologies, including Information and Communication Technology (ICT) platforms, smart metering infrastructure, automated control of energy flows, and demand response mechanisms. These technologies serve not only to optimize energy efficiency and flexibility but also to enhance user engagement and energy awareness. A national standard recently published in Italy has formalized this integrated methodology, supporting the coordinated development of smart and low-carbon cities. Concurrently, innovative tools are being developed to facilitate decision-making and strategic planning for RECs at multiple spatial scales. Among them, the Italian geo-portal for RECs and the Public Energy Living Lab (PELL) have been introduced to support the acquisition, organization, and interpretation of territorial and urban energy data. These tools have also enabled the definition and monitoring of context-specific Key Performance Indicators (KPIs), critical for assessing the performance and scalability of REC initiatives. The framework presented herein contributes to the broader objectives of Smart Cities by enabling data-driven, participatory, and resilient energy transitions in urban contexts. Particular emphasis has been placed on harmonizing spatial data infrastructures with energy governance processes, thereby laying the groundwork for replicable and adaptable REC models across diverse territorial configurations.
The accelerating demand for sustainable energy solutions in urban environments has prompted the application of building-integrated photovoltaic (BIPV) systems in electric vehicles (EVs). This study assessed the impact of BIPV-EV systems in Surabaya, Indonesia, forecasting its energy production, environmental advantages, and economic viability between 2026 and 2036. Simulations conducted using HOMER Pro and photovoltaic system (PVsyst) suggested that the rooftop photovoltaic (RPV) capacity will increase from 4.6 GW in 2026 to 6.0 GW by 2036, while facade photovoltaic (FaPV) capacity is projected to grow from 1.6 GW to 2.0 GW. The combined generation of RPV and FaPV is anticipated to reach 9.71 GWh annually by 2036, ultimately reducing grid dependency to 36.6%. Additionally, carbon emissions from the BIPV-EV systems are expected to decrease from 616 tons per year in a grid-based scenario to 520 tons annually, hence reducing carbon intensity to 0.05 kg CO₂/kWh. Although the initial investment is projected at USD 3.2 billion and USD 4.8 billion in 2026 and 2036, respectively, the implementation of BIPV-EV systems is advantageous owing to significant savings on energy costs in the long run and decreasing reliance on fossil fuels. These findings underscored the potential of BIPV in advancing urban sustainability and accomplishing the objectives of energy transition in Indonesia.
In an era defined by digital transformation and systemic volatility, conventional approaches to strategic risk management have been increasingly challenged by the complexity and unpredictability of modern operational environments. To address these limitations, a novel artificial intelligence (AI)–driven framework has been developed to enhance organizational resilience and optimize strategic decision-making. Constructed through a systematic review conducted in accordance with PRISMA 2020 guidelines, this study synthesizes current academic literature and industry publications to identify critical enablers, practical gaps, and methodological advancements in AI-enabled risk governance. The proposed framework integrates real-time analytics, predictive modelling, and adaptive governance mechanisms, aligning them with enterprise-wide strategic objectives to support decision-making under volatile, uncertain, complex, and ambiguous (VUCA) conditions. Anchored in dynamic capabilities theory and decision support systems (DSS) literature, the framework is designed to facilitate proactive risk anticipation, reduce cognitive and algorithmic biases in decision-making, and foster strategic alignment in rapidly evolving contexts. Its adaptability to small and medium-sized enterprises (SMEs), as well as its cross-sectoral relevance, underscores its scalability and practical utility. Nonetheless, the effectiveness of the framework is contingent upon the availability of high-quality data, the level of digital maturity within organizations, and the implementation of responsible AI principles. By bridging the gap between theoretical innovation and real-world applicability, this study contributes a robust foundation for future empirical validation and sector-specific customization. The framework is expected to inform governance and technology leaders aiming to institutionalize AI-based resilience capabilities, thereby supporting sustainable strategic outcomes in both developed and emerging markets.
The roles of stakeholders in the development of sustainable organic rice farming in Yogyakarta, Indonesia, were investigated, along with the patterns of collaboration among them. The study was conducted in the Special Region of Yogyakarta through purposive sampling to identify key informants. A qualitative methodology was employed, utilizing data collection through structured observations and in-depth interviews. The data were analyzed using an interactive model combined with thematic analysis, involving iterative stages of data gathering, reduction, presentation, and conclusion drawing. Stakeholder roles identified included those in education and socialization, technical mentoring, organic farming training, marketing facilitation, organic fertilizer production, organic rice production, policy formulation, scientific research contributions, and the establishment of association institutions. Collaborative linkages were identified among various actors, including academic institutions, business units, farmer groups, government bodies, and media organizations. Collaborative linkages were observed between academic institutions and business units, farmer groups, government agencies, and media organizations; business units and academic institutions; farmer groups and academic, business, governmental, and media stakeholders; government agencies and academic, business, farmer, and media sectors; and media organizations with academic, business, governmental, and farming communities. These findings underscore the complexity and significance of multi-stakeholder cooperation in advancing sustainable organic agriculture. Strengthening these collaborations is considered essential for the long-term success and resilience of organic farming initiatives in the region.
The role of natural resource rents (NRR) in driving environmental degradation has attracted increasing scholarly attention, particularly in resource-dependent economies. In the case of Saudi Arabia, where oil and gas extraction constitutes a substantial proportion of GDP, the relationship between resource rents and environmental quality warrants rigorous investigation. This study examines the effects of oil, natural gas, mineral, and forest rents on carbon dioxide (CO₂) emissions within the framework of the Environmental Kuznets Curve (EKC), over the period 1970–2023. Employing optimal lag selection criteria, augmented Dickey–Fuller and Phillips–Perron unit root tests were applied to ensure the stationarity of variables, followed by Johansen cointegration analysis to establish the existence of long-run relationships among them. The EKC hypothesis is empirically validated, with a turning point identified at 65,914 Saudi Riyals (SR) in the long term and 65,912 SR in the short term, indicating a non-linear relationship between economic growth and CO₂ emissions. Oil Rents (OR) were found to exert statistically significant positive effects on CO₂ emissions in both the short and long run, suggesting that oil dependence remains a critical driver of environmental degradation. Conversely, natural gas, mineral, and forest rents exhibited statistically insignificant impacts in the long run, although short-run analyses revealed a positive but marginally significant influence of natural gas and forest rents. These findings underscore the asymmetric environmental implications of different types of resource rents. Policy implications point toward the urgent need to diversify the economic base away from oil dependency and enhance regulatory frameworks to mitigate the ecological costs of resource exploitation. By integrating the EKC hypothesis with disaggregated rent variables, this study contributes to the nuanced understanding of resource–environment dynamics in hydrocarbon-reliant economies.
Green supplier selection (GSS) as a critical strategic element is placed in the limelight in contemporary supply chain management (SCM), owing to the growing emphasis on environmental responsibility and sustainability. This study presents a fuzzy multi-criteria decision-making (FMCDM) framework, employing Fuzzy Logarithmic Percentage Change-Driven Objective Weighting (FLOPCOW) method to determine the relative importance of sustainability criteria under uncertainty. A panel of five academic and industry experts was selected to identify 21 criteria, which were categorized into three main dimensions including environmental performance (C1), resource efficiency (C2), and corporate sustainability policies (C3). Triangular fuzzy numbers (TFNs) were adopted to model linguistic ambiguities in expert judgments whereas fuzzy normalization was applied to ascertain the weights of criteria. Key findings indicated that corporate sustainability policies (C3) were prioritized as the most influential dimension, followed by environmental performance (C1) and resource efficiency (C2). This suggested the centrality of institutional governance in advancing long-term sustainability objectives. Sub-criteria analysis further revealed ecological training programs, air emissions control, and sustainability reporting as the most critical indicators in the interplay of operational practices and transparent governance. FLOPCOW has effectively processed expert opinions with the use of fuzzy normalization, hence advocating a clear and repeatable approach for the evaluation of green suppliers. Furthermore, it highlighted the importance of policy-based criteria in supplier assessment and organizations could then align their purchasing decisions with sustainability goals by considering more on governance-related factors like compliance and stakeholder engagement.
The problem of job scheduling in parallel machine environments, where both processing times and setup times are characterized by stochastic variability, has been investigated with a focus on enhancing the efficiency of resource allocation in complex production systems. Job scheduling, as a critical component of operations research and systems engineering, plays a vital role in the optimization of large-scale, flexible manufacturing and service environments. In this study, a stochastic scheduling model has been formulated to minimize the maximum completion time (denoted as $Ct_{\textit{max}}$), under the simultaneous influence of probabilistic job durations and setup times associated with tool preparation. The problem has been addressed using two prominent metaheuristic algorithms: Genetic Algorithm (GA) and Simulated Annealing (SA). These methods were selected due to their demonstrated capacity to navigate large, non-deterministic search spaces efficiently and their adaptability to multi-constraint scheduling problems. A comparative analysis has been conducted by applying both algorithms under identical initial conditions, with algorithmic performance evaluated in terms of solution quality, computational efficiency, and robustness to input variability. The model incorporates key practical considerations, including randomized setup times which are often neglected in conventional deterministic scheduling models, thereby improving its relevance to real-world industrial settings. The formulation of the problem allows for additional constraints and objectives to be flexibly integrated in future research, including resource conflicts, machine eligibility constraints, and energy-aware scheduling. Empirical results suggest that while both algorithms are effective in deriving near-optimal schedules, notable differences exist in convergence behavior and sensitivity to parameter tuning. The findings offer critical insights into the comparative strengths of GA and SA in managing the stochastic nature of parallel machine scheduling problems. By advancing a robust metaheuristic framework that accounts for real-world uncertainties, this study contributes to the ongoing development of intelligent scheduling systems in systems engineering, manufacturing logistics, and automated production planning.
For Authors
For Reviewers
For Editors
Open Access
