Data-driven Decision Research Articles

Designing a reliability engineering framework to minimize downtime and enhance output in energy production

Energy production systems face significant challenges in minimizing downtime and enhancing operational efficiency, necessitating the development of robust reliability engineering frameworks. This paper proposes a comprehensive framework designed to improve system reliability by integrating advanced predictive maintenance techniques, failure analysis, and risk assessment. The study examines theoretical foundations and existing frameworks, identifying critical gaps in current approaches, particularly in their capacity to leverage emerging technologies such as machine learning, IoT, and advanced analytics. The proposed framework emphasizes data-driven decision-making, real-time monitoring, and proactive maintenance to reduce unplanned downtime and optimize output. The paper also details a structured implementation strategy, including step-by-step integration into existing operational practices, stakeholder engagement, and resource optimization to address potential barriers such as technological limitations and resistance to change. Practical implications of the framework include enhanced cost efficiency, improved sustainability, and alignment with global energy efficiency goals. Furthermore, the study identifies future research opportunities in areas such as decentralized monitoring systems, digital twins, and integrating renewable energy sources into reliability strategies. Actionable recommendations are provided to guide energy production companies and policymakers in adopting and supporting these frameworks, ensuring long-term improvements in reliability, cost-effectiveness, and environmental stewardship. Keywords: Reliability Engineering, Predictive Maintenance, Energy Production Systems, Operational Efficiency, Data-Driven Decision-Making, Sustainability.

Artificial Neural Network Framework for Hybrid Control and Monitoring in Turning Operations

In the era of Industry 4.0 and the transition toward Industry 5.0, advanced manufacturing is increasingly driven by data analytics, artificial intelligence, and cyber-physical systems. The integration of intelligent monitoring systems and self-learning algorithms is reshaping machining processes, enabling higher efficiency, precision, and sustainability. Recent advancements in smart factories emphasize the use of AI-powered process control, enabling machines to self-optimize, self-correct, and even self-retrain to maintain optimal performance. This paper proposes a hybrid control and monitoring framework designed to enhance turning operations by integrating artificial neural networks (ANNs) for predictive modeling and adaptive recalibration. The system leverages machine learning (ML) to improve machining efficiency, tool longevity, and energy consumption optimization. By implementing forward and inverse ANN models, the framework enables real-time estimation of cutting forces and energy consumption, facilitating data-driven decision-making in machining processes. Furthermore, an adaptive recalibration mechanism ensures continuous model updates, allowing the system to dynamically adjust based on evolving machining conditions such as tool wear, material properties, and environmental variations. This research contributes to these advancements by proposing an ANN-based hybrid approach, predictive modeling, and adaptive recalibration. The proposed framework ensures continuous monitoring, automated adjustments, and intelligent decision-making, making it a scalable and adaptable solution for modern machining operations.

The role of business analytics in enhancing revenue optimization and competitive advantage in e-commerce

In the rapidly evolving e-commerce landscape, businesses are increasingly relying on business analytics to optimize revenue and maintain a competitive advantage. This paper explores the role of business analytics in enhancing operational efficiency and driving strategic decision-making within the e-commerce sector. The study examines various analytical techniques, including predictive analytics, customer segmentation, and dynamic pricing strategies, which have been crucial in improving revenue optimization. It also delves into how these analytics tools contribute to gaining and sustaining competitive advantage through applications such as personalized marketing, real-time data insights, and efficient resource allocation. Furthermore, the paper reviews the current trends in e-commerce analytics, such as the integration of artificial intelligence (AI) and machine learning to predict consumer behavior and optimize pricing models. The findings highlight that data-driven decision-making enables businesses to enhance profitability, improve customer satisfaction, and respond swiftly to changing market dynamics. Additionally, the study discusses common challenges in adopting business analytics, such as data quality issues, resistance to change, and high implementation costs, and offers practical solutions to overcome these barriers. This paper concludes with actionable recommendations for e-commerce businesses to leverage the power of business analytics fully, ensuring they remain competitive in an increasingly data-driven environment. Future research is suggested to explore the ethical implications of data usage, the potential of AI in analytics, and the impact of real-time analytics on long-term business strategies. Keywords: Business Analytics, E-Commerce, Revenue Optimization, Competitive Advantage, Predictive Analytics, Artificial Intelligence.

Computer-Driven Assessment of Weighted Attributes for E-Learning Optimization

Computer-driven assessment has revolutionized the way educational and professional assessments are conducted. Using artificial intelligence for data analytics, computer-based assessment improves efficiency, accuracy, and optimization of learning across disciplines. Optimizing e-learning requires a structured approach to analyzing learners’ progress and adjusting instruction accordingly. Although learning effectiveness is influenced by numerous parameters, competency-based assessment provides a structured and measurable way to evaluate learners’ achievements. This study explores the application of artificial intelligence algorithms to optimize e-learners’ studying within a generalized e-course framework. A competency-based assessment model was developed using weighted parameters derived from Bloom’s taxonomy. The key contribution of this work is an innovative method for calculating competency scores using weighted attributes and a dynamic assessment parameter, making the optimization process applicable to both learners and instructors. The results indicate that using the weighted attribute method with a dynamic assessment parameter can improve the structuring of e-courses, increase learner engagement, and provide instructors with a clearer understanding of learners’ progress. The proposed approach supports data-driven decision making in e-learning, ensuring a personalized learning experience, and improving overall learning outcomes.

From Elusive to Opaque Boundaries of the F&A Function: An Assessment of Digitalization Impact

ABSTRACT Previous research on digital transformation of Finance & Accounting is largely conceptual, lacking insights into financial specialists’ roles in adopting and implementing digital technologies. By employing Resource Orchestration Theory, this study examines digitalization’s impact on F&A functions through four case studies. Findings reveal diverse adoption strategies, evolving F&A processes, and data-driven decision-making. The results show that due to the adoption of digital technologies, the elusive boundaries of F&A functions have become opaque. This paper contributes to F&A functions literature and provides new knowledge on how to seize the opportunity arising from the more elusive boundaries of F&A within diverse organizations.

Integrating machine learning and data envelopment analysis for reliable reservoir water quality index assessment considering uncertainty

ABSTRACT Water quality assessment is crucial for environmental health and quality of life. This study introduces a novel water quality index (WQI) model for reservoirs, using Wadi Dayqah Dam in Oman as a case study. The model advances water quality assessment using a data-driven approach, reducing reliance on subjective expert opinions. A large dataset of water samples was analysed using machine learning (ML) to select water quality variables (WQVs). Using a bootstrapping and subsampling approach, the proposed WQI was then calculated through sub-indexing, weighting, and aggregating sub-indices. WQV weights were estimated using gradient boosting and rank order centroid techniques, while aggregation involved scoring and data envelopment analysis (DEA). The model effectively captures uncertainty, prioritizes WQVs, and provides solutions to issues such as eclipsing, ranking, and dealing with bad variable values. The results were validated through uncertainty and sensitivity analyses, highlighting the model’s potential for enhancing data-driven decision making in reservoir management.

Improved cancer detection through feature selection using the binary Al Biruni Earth radius algorithm

With the advancement of medical technology, a large amount of complex data on cancers is produced for diagnosing and treating cancers. However, not all this data is useful, as many features are redundant or irrelevant, which can reduce the accuracy of machine learning models. Metaheuristic algorithms have been employed to select features to address this issue. Although the efficacy of these algorithms has been demonstrated, challenges related to scalability and efficiency persist when handling large medical datasets. In this study, a binary version of the Advanced Al-Biruni Earth Radius (bABER) algorithm is proposed for the intelligent removal of unnecessary data and identifying the most essential features for cancer detection. Unlike traditional methods that rely on a single approach, bABER is evaluated using seven medical datasets and compared with eight widely used binary metaheuristic algorithms, including bSC, bPSO, bWAO, bGWO, bMVO, bSBO, bFA, and bGA. Statistical tests such as ANOVA and the Wilcoxon signed-rank test are conducted to ensure a thorough performance assessment. The results indicate that the bABER algorithm significantly outperforms other methods, making it a valuable tool for improving cancer diagnosis. By refining feature selection, this approach enhances existing machine learning models, leading to more accurate and reliable medical predictions. This study contributes to improved data-driven decision-making in healthcare, bringing the field closer to faster and more precise cancer detection.

Leveraging Artificial Intelligence, Data Analysis, and Computer Science in Primary Care: Enhancing Electronic Health Records for Improved Patient Outcomes

The integration of artificial intelligence (AI) with primary care elements, including data analytics and computer science, facilitates transformative electronic health record (EHR) management, ultimately enhancing patient outcomes. Modern healthcare systems increasingly rely on electronic records for sustainable data storage, efficient retrieval, and the secure distribution of health information. However, the full potential of EHRs remains unrealized due to three primary challenges: fragmented information systems, incomplete clinical documentation, and inefficient healthcare workflows. The implementation of AI-driven solutions and advanced data analytics can mitigate these issues by enhancing data integration and providing healthcare professionals with actionable insights, thereby supporting more informed clinical decision-making. AI algorithms are capable of processing vast healthcare datasets to identify patterns that can predict patient outcomes, assisting healthcare providers in making precise, data-driven clinical decisions. Predictive analytics tools facilitate the early identification of patients at risk of developing chronic conditions, enabling timely intervention and significantly altering disease trajectories. Additionally, natural language processing (NLP) technologies convert unstructured healthcare data—such as physician notes and patient assessments—into structured, analyzable formats, improving the overall utility of healthcare information systems. These advancements streamline data management while allowing healthcare professionals to allocate more time to direct patient care rather than administrative tasks. The interoperability of EHR systems improves when healthcare providers apply computer science principles to develop integrated infrastructures that enable seamless data exchange across disparate platforms. The implementation of unified patient information management strategies fosters continuity of care while reducing redundancies and minimizing system errors. Emerging technological advancements present numerous benefits for healthcare operations, facilitating large-scale data analysis for scientific research and evidence-based policy development. The strategic integration of AI, data analytics, and computer science into primary care EHR systems will drive superior patient outcomes, optimize clinical workflows, and enhance adaptability to evolving healthcare demands.

The IFMIF-DONES Control Architecture: The State-of-the-Art Design of Central and Local Control Systems and Communication Networks.

Abstract The International Fusion Materials Irradiation Facility-DEMO-Oriented Neutron Source (IFMIF-DONES) is an advanced neutron source driven by an accelerator, designed to generate high-energy neutrons for testing materials intended for DEMO, the upcoming fusion reactor. Due to the plant’s complexity, a reliable central control system is essential to manage and supervise operations safely. This paper reviews recent progress in the design of the control systems for IFMIF-DONES, with a focus on the transition into the full definition design phase. The aim is to provide here a clear and comprehensive description of the current state of the art in control systems design, outlining the latest advancements and challenges. The IFMIF-DONES control systems is composed of two levels: the Central Instrumentation and Control Systems (CICS) and the Local Instrumentation and Control Systems (LICS), connected togeher by a complex set of communication networks and buses. CICS consists of three core systems: CODAC, MPS, and SCS, each tasked with specific functions. CODAC handles overall coordination, orchestration, and data management; MPS is responsible for machine protection; SCS ensures safety for personnel and the environment. The CICS architecture follows a hierarchical structure that supports a modular and scalable design, integrating redundancy and fault tolerance. Utilizing a distributed approach, the architecture incorporates fast devices and specialized networks for real-time communication between control units. This paper details the current design status of each CICS system and outlines both ongoing and future integration plans within a unified control structure. One key challenge in this integration is synchronizing data acquisition and managing interlocks. Artificial intelligence (AI) tools can significantly enhance CICS subsystems, enabling data-driven decision-making, predictive maintenance, adaptive control, and intelligent optimization.

Artificial intelligence in educational games and consent under general data protection regulation

As Artificial Intelligence becomes increasingly integrated into educational games, conforming with the General Data Protection Regulation (GDPR)—a legal framework governing data protection and privacy in the European Union—remains an important yet complex challenge, particularly when minors are involved. Users are required to provide consent multiple times, often unexpectedly, at different game levels. This process is further complicated by the varying durations for which consent remains valid. As a result, users—especially minors—may become confused about the consent they have given. Additional concerns arise when the educational game is AI-equipped. If AI is not involved, no new data is generated. However, if AI is present, new data is continuously produced, necessitating ongoing consent. For example, a user may consent to personalisation, which could lead the game to categorise them in unintended ways, such as labelling them a ‘poor student’. This paper explores GDPR challenges in AI-empowered educational games, focusing on user consent, AI-inferred data, and compliance gaps. Intelligent educational games rely on adaptive decision-making algorithms to personalise learning experiences, making them a subset of Intelligent Decision Technologies. Our research is based on a fuzzy-based educational game developed as a testbed for studying GDPR compliance in AI-driven decision-making. The findings provide insights into ethical AI governance, dynamic consent management, and the intersection of regulatory compliance with adaptive, data-driven decision systems in intelligent educational technologies. Based on our research, not all personal data exist from the beginning and upon original consent granting, as personal data are also generated throughout the process.

Enhancing Teacher Development through Professional Learning Communities: A Mixed-Methods Study in Chinese Universities

This study investigated the implementation and impact of Professional Learning Communities (PLCs) on English language teacher development in selected Chinese universities. Employing a mixed-method approach with a convergent parallel design, the research examined the extent of PLC implementation across various dimensions and its influence on teachers' professional growth. Quantitative data from 136 EFL teachers and qualitative insights from 14 participants (7 PLC leaders and 7 members) were collected and analyzed. Results revealed high levels of teacher involvement in key PLC components, including shared vision and goals, collaborative learning, reflective practice, data-driven decision-making, leadership, and structured programs. The study found that PLCs significantly enhanced teachers' content knowledge, instructional practices, personal growth, and collegial relationships. Qualitative findings corroborated these results, highlighting improved collaboration, instructional strategies, and support for English learners. The integration of quantitative and qualitative data provided a comprehensive understanding of PLC effectiveness in the Chinese higher education context. However, challenges were identified in areas such as advocating for PLC work beyond immediate communities and translating data insights into concrete learning objectives. The study contributes to the growing body of research on PLCs in diverse cultural contexts and offers valuable insights for implementing and sustaining effective professional development models in Chinese universities. Recommendations include enhancing institutional support, developing teacher leadership skills, and fostering a culture of open feedback and data literacy. These findings have implications for educational policy and practice in China and globally, particularly in adapting PLCs to different cultural and institutional settings.

Exploring the Role of Generative AI in Procurement, Contract Lifecycle Management, Supplier Risk Assessment, and Supply Chain Planning

Generative AI is transforming the landscape of procurement, contract lifecycle management, supplier risk assessment, and supply chain planning by enabling automation, data-driven decision-making, and predictive insights. This paper explores the integration of generative AI in these domains, highlighting its potential to streamline procurement workflows, enhance contract generation and analysis, mitigate supplier risks through predictive modeling, and optimize supply chain strategies. The study also discusses key challenges such as data privacy, model biases, and regulatory considerations. By examining real-world applications and emerging trends, this paper provides a comprehensive understanding of how generative AI is reshaping supply chain and procurement ecosystems. DOI:https://doi.org/10.52783/pst.1648

A Predictive Models for Advertisement Campaign Budget Allocation

This study explores the role of predictive models in optimizing advertisement campaign budget allocation. As digital marketing grows more complex, predictive models offer data-driven insights that help advertisers allocate budgets more efficiently. These models use machine learning to analyze past performance, predict trends, and optimize resource distribution across channels, improving campaign outcomes and return on investment (ROI). Techniques such as real-time bidding (RTB), customer segmentation, and multi-touch attribution have enhanced budget allocation. However, challenges like data quality, model interpretability, and integration complexity limit widespread use. Predictive models are integrated into platforms like Google Ads and Facebook Ads Manager, optimizing cost-per-click (CPC) and conversion rates. Balancing automation with human oversight remains crucial. Research should focus on real-time data integration and ethical concerns around data privacy to ensure responsible use. Refining these models will empower advertisers to make better data-driven decisions, improving budget allocation and campaign success.

A Predictive Models for Advertisement Campaign Budget Allocation

This study explores the role of predictive models in optimizing advertisement campaign budget allocation. As digital marketing grows more complex, predictive models offer data-driven insights that help advertisers allocate budgets more efficiently. These models use machine learning to analyze past performance, predict trends, and optimize resource distribution across channels, improving campaign outcomes and return on investment (ROI). Techniques such as real-time bidding (RTB), customer segmentation, and multi-touch attribution have enhanced budget allocation. However, challenges like data quality, model interpretability, and integration complexity limit widespread use. Predictive models are integrated into platforms like Google Ads and Facebook Ads Manager, optimizing cost-per-click (CPC) and conversion rates. Balancing automation with human oversight remains crucial. Research should focus on real-time data integration and ethical concerns around data privacy to ensure responsible use. Refining these models will empower advertisers to make better data-driven decisions, improving budget allocation and campaign success.

Automated MLOps Pipeline Implementation for Intelligent Procurement Replenishment: A Predictive Analytics Approach

This article presents an advanced implementation of automated model retraining pipelines for procurement replenishment recommendations, leveraging Machine Learning Operations (MLOps) to enhance supply chain efficiency and decision-making accuracy. The integration of artificial intelligence and machine learning into procurement operations represents a paradigm shift from traditional rule-based systems to intelligent, adaptive frameworks capable of responding to market dynamics in real-time. Through multi-source demand forecasting, dynamic feature engineering, and multi-model ensemble approaches, the system achieves superior prediction accuracy compared to conventional methods. Real-time inventory monitoring, comprehensive supplier evaluation, and lead time variability tracking provide the foundation for data-driven procurement decisions. The containerized MLOps infrastructure ensures scalable, reliable deployment while automated data quality validation and model retraining maintain system accuracy over time. Smart reorder point calculations, dynamic safety stock optimization, and cost-benefit analysis automation deliver operational intelligence that optimizes inventory levels and procurement costs. The risk-adjusted recommendation engine incorporates vendor performance metrics and external risk factors to generate resilient procurement strategies, with real-time dashboards translating complex analytics into actionable recommendations for procurement teams.

Assessing the reliability of AI-driven predictive models in food safety risk management

Sustaining food safety in a world that has undergone globalization with complex supply chains makes it very challenging. Conventional risk management methodologies are sometimes inadequate in addressing emerging risks, making the use of AI and predictive modeling invaluable tools in enhancing food safety evaluation and decision making. AI-based models have the prospects of identifying contamination risks at an early stage, enhancing the optimization of hazardous control measures, and increasing the effectiveness of compliance monitoring. Nevertheless, the reliability of their predictions is still questionable due to factors such as data integrity, model interpretability, and regulatory compliance that influence their applicability in practice. Despite AI's potential, challenges such as inconsistent data sources, varying regulatory standards, and adaptability across diverse food production environments limit its efficacy. Addressing these issues is crucial for AI-driven models to be fully integrated into food safety management systems. This paper provides a critical review of the current AI-based predictive models for food safety risk management synthesizing insights from existing literature, industry studies and regulatory reports. Through an evaluation of the strengths, drawbacks and limitations of these models, this research emphasizes the importance of developing standardized validation frameworks and improved strategies for integrating data across models. Overall, this research enriches the current discourse regarding the use of AI in food safety and offers key recommendations for improving model reliability and performance in protecting consumers and safeguarding public health Keywords: Artificial Intelligence (AI), Predictive Modeling, Food Safety Risk Management, Data-Driven Decision Making.

Enhancing Retail Performance through Customer Traffic Analysis and Personalized Marketing Strategies: A Systematic Approach

This article comprehensively examines how implementing advanced customer tracking systems and targeted marketing strategies can transform retail business performance. The article investigates the integration of cutting-edge counting systems, purchase tracking mechanisms, and personalized marketing approaches to enhance retail operations and customer engagement. The research demonstrates the effectiveness of a systematic approach to retail optimization through detailed analysis of traffic patterns, purchase behaviors, and customer responses to targeted marketing initiatives. The article encompasses sophisticated data collection protocols, advanced analytical frameworks, and carefully designed implementation strategies considering technical requirements and privacy concerns. The findings reveal significant improvements in key performance metrics, including traffic conversion rates, category-specific sales performance, and marketing response rates. The article also explores the business impact through revenue growth projections, customer loyalty measurements, and return on marketing investment analyses. This article provides valuable insights for retail practitioners seeking to enhance their operations through data-driven decision-making and offers a framework for implementing comprehensive customer tracking and marketing systems. The findings suggest that such integrated approaches are becoming increasingly crucial for maintaining competitiveness in the modern retail environment while emphasizing the importance of balancing technological advancement with customer privacy and ethical considerations.

Data-Driven Decision Making

The landscape of organizational intelligence is undergoing a profound transformation driven by artificial intelligence and advanced technological innovations. This article reveals the multifaceted evolution of data-driven decision-making, highlighting how emerging technologies are fundamentally reshaping organizational strategies across diverse domains. By synthesizing insights from cutting-edge scholarly investigations, the article demonstrates the pivotal role of artificial intelligence in transcending traditional computational constraints, enabling unprecedented levels of insight generation, predictive modeling, and strategic optimization. It illuminates a holistic approach to technological advancement that integrates sophisticated computational capabilities with ethical considerations, transparency, and human-centric design principles.

Ai-Powered Business Management: The Next Generation

The integration of the Artificial Intelligence (AI) into business management is ushering in a new era of efficiency, innovation, and data-driven decision-making (McKinsey & Company, 2023). From automating routine tasks to the predicting market trends and enhancing customer experiences, the AI is reshaping traditional business processes (Smith, 2024). This article explores the transformative role of AI in business management, covering its applications, benefits, challenges, and future potential. Real-world examples illustrate how the AI is revolutionizing industries, while also addressing the ethical considerations with strategic approaches required for successful implementation (Deloitte Insights, 2023).

Examining Principal Leadership in Cambodian High Schools: A Case Study Approach

This study examines the characteristics, challenges, and strategies of school leadership in Cambodian high schools, with a focus on leadership in resource-constrained settings. Using a qualitative case study approach, the research investigates how principals in six northwestern provinces implement leadership practices to improve educational outcomes. Data were collected through semi structured interviews with 21 principals and analyzed via thematic analysis. Key findings reveal that visionary approaches, collaborative cultures, data-driven decision-making, and resilience characterize effective leadership. Challenges such as limited resources, variable teacher quality, and insufficient professional development hinder progress. The study contributes to educational leadership theory by highlighting the critical role of tailored leadership development programs and resource management in transforming underperforming schools. Practical implications include recommendations for policymakers to expand leadership training, enhance community engagement, and prioritize funding for underperforming schools. This research offers valuable insights for improving leadership in post conflict and resource-limited educational contexts.