Volume 8.
Issue 08.

The integration of Artificial Intelligence (AI) and Machine Learning (ML) is transforming DevOps, enabling it to become a predictive, intelligent, and adaptive process across the software delivery lifecycle. This study aims to create an AI/ML based DevOps automation framework for deployment risk prediction, deployment optimization, rollback management, and continuous monitoring in a single CI/CD workflow. The framework’s data components are the build logs, deployment history, infrastructure monitoring, configuration repositories, incident records, and user feedback, which are then combined with DevOps data into a machine learning pipeline. The data is then preprocessed, combined, and processed by feature engineering to be split into training, validation, and testing sets. A Random Forest model is employed for deployment risk prediction, while an AI-driven decision engine uses predicted risk and operational conditions to support deployment optimization, resource allocation, automated rollback, and continuous monitoring. Experimental evaluation demonstrates a 50% reduction in average build time, a 60% reduction in deployment failure rate, a 50% reduction in rollback frequency, a 34% improvement in deployment risk prediction accuracy, and a 68% reduction in false positive rate. The findings demonstrate improved software delivery efficiency, reliability, operational stability, and decision-making.

Production planning in enterprise resource planning environments requires continuous coordination among demand, inventory, material availability, production capacity, and operational constraints. Conventional planning approaches implemented through SAP environments are effective for structured transaction processing but can become less responsive when production systems exhibit nonlinear demand patterns, stochastic disturbances, capacity limitations, and complex optimization landscapes. This research develops a conceptual machine learning-driven framework for optimizing SAP production planning with emphasis on inventory efficiency and production throughput. The proposed approach integrates SAP planning data with machine learning-based demand and production-state estimation, followed by iterative optimization of production quantities, material allocation, and scheduling decisions. The theoretical foundation is derived from research on gradient-based optimization, stochastic gradient methods, nonsmooth optimization, convergence analysis, and neural-network learning landscapes. In particular, the convergence properties discussed by Absil et al. (2005), Attouch and Bolte (2009), Bolte et al. (2007), Davis et al. (2020), Dereich and Kassing (2021, 2022, 2023), Eberle et al. (2023), and Jentzen and Riekert (2022) provide a mathematical basis for constructing stable iterative optimization procedures. The resulting framework seeks to reduce excess inventory while preserving service-oriented production capacity and improving throughput. The analysis indicates that the principal value of machine learning within SAP production planning is not simply prediction accuracy, but the integration of predictive intelligence with convergent and constraint-aware decision optimization. Limitations include dependence on data quality, model stability, SAP integration complexity, and the possibility of suboptimal local solutions.

The article presents a comprehensive analysis of the digitalization of construction control and the implementation of Building Information Modeling (BIM) technologies in the district heating sector as a key direction in the formation of sustainable, energy-efficient, and controllable engineering infrastructure. The study integrates engineering, architectural-management, and operational approaches, covering issues of thermal network modeling, organization of construction supervision, and optimization of the life cycle of heat supply facilities. The comparative assessment of traditional and digital methods showed that the transition to BIM platforms increases the accuracy of construction supervision, reduces decision-making time, and enhances the reliability of heat supply systems. Particular attention is given to the architecture of the digital model, which unites design, monitoring, and operation processes within a single information environment. It is established that the structural integration of BIM models with elements of virtual and augmented reality forms the foundation for predictive management and prevention of technological risks. The practical significance of the results lies in their applicability to the design, modernization, and management of heating networks at the municipal and corporate infrastructure levels. The article may be useful to professionals in construction, energy, digital design, and lifecycle management of engineering systems. The study demonstrates that BIM technologies in district heating represent not merely a visualization tool but a new form of organizational and technological thinking, where accuracy, reliability, and energy efficiency become systemic attributes of digital construction.

The increasing complexity of global supply chain operations has created an urgent need for intelligent decision-support systems capable of processing heterogeneous operational data and delivering accurate, real-time insights. Conventional supply chain management approaches often rely on single-source structured data and traditional analytical techniques, limiting their ability to capture the dynamic relationships among procurement, inventory, transportation, warehousing, supplier performance, and customer demand. To address these challenges, this study proposes a comprehensive Multimodal Artificial Intelligence (AI) and Data Analytics framework for intelligent supply chain decision-making. The proposed framework integrates numerical, categorical, temporal, and engineered business features extracted from an open-source supply chain dataset obtained from the Kaggle repository. A systematic methodology involving data preprocessing, feature extraction, feature engineering, model development, and model evaluation was implemented to develop robust predictive models. Eight machine learning and deep learning algorithms, including Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, XGBoost, LightGBM, CatBoost, and a proposed Multimodal Deep Neural Network (MDNN), were trained and evaluated using identical experimental settings. The experimental results demonstrate that the proposed MDNN significantly outperformed all benchmark models, achieving an accuracy of 98.43%, precision of 98.5%, recall of 98.3%, F1-score of 98.4%, AUC-ROC of 0.995, and Matthews Correlation Coefficient (MCC) of 0.968, indicating exceptional predictive capability and generalization performance. Furthermore, Explainable Artificial Intelligence (XAI) techniques, including SHAP and LIME, were incorporated to enhance model interpretability and support transparent managerial decision-making. The proposed framework provides a scalable and industry-ready solution that can be integrated with Enterprise Resource Planning (ERP), Internet of Things (IoT), cloud computing, and real-time analytics platforms to improve supply chain resilience, operational efficiency, inventory optimization, logistics planning, and risk management. The findings demonstrate that multimodal AI substantially enhances intelligent supply chain decision-making compared with conventional machine learning approaches and provides a practical pathway toward the realization of Industry 4.0 and Supply Chain 5.0.

The convergence of Industrial Internet of Things (IIoT), artificial intelligence, automation, and cyber-physical systems (CPS) is transforming manufacturing into highly connected and intelligent production environments. However, increased connectivity also expands the attack surface and creates a requirement for security mechanisms capable of identifying anomalous behavior with low latency. Conventional centralized security architectures may introduce communication overhead, delayed responses, and dependency on continuous connectivity with cloud infrastructure. This paper proposes an Edge-AI-Based Anomaly Detection Framework for securing IIoT and cyber-physical manufacturing systems. The proposed framework places lightweight artificial intelligence inference capabilities close to industrial data sources, enabling continuous analysis of sensor, machine, process, and network observations. A layered architecture comprising data acquisition, preprocessing, edge intelligence, anomaly scoring, decision enforcement, and centralized coordination is developed conceptually. The methodology integrates contextual process behavior with machine-learning-based anomaly identification to distinguish operational deviations from potentially malicious activities. The literature synthesis demonstrates that intelligent agriculture, computational intelligence, and IoT-enabled management systems provide useful foundations for distributed sensing, intelligent decision-making, and real-time analytics, although these studies do not directly address the security requirements of industrial CPS. The proposed framework therefore extends these principles toward security-oriented edge intelligence. Findings indicate that edge-based detection can improve responsiveness, reduce unnecessary data transmission, and support localized decision-making, while introducing challenges related to resource constraints, model maintenance, false positives, and heterogeneous industrial environments. The framework provides a research-oriented foundation for deploying explainable, low-latency anomaly detection in next-generation smart manufacturing environments.

Agile software development emphasizes rapid iteration, continuous integration, frequent releases, and incremental delivery, making regression testing a central software quality challenge. Conventional regression testing approaches often depend on manually selected test suites, static prioritization rules, and repeated execution of tests that provide limited incremental fault-detection value. This paper develops a conceptual intelligent regression testing framework that applies artificial intelligence (AI) techniques to test selection, prioritization, execution, failure classification, and continuous learning within Agile development pipelines. The methodological foundation combines supervised learning, representation learning, historical test-result analysis, change-impact assessment, and feedback-driven optimization. Because the supplied literature primarily concerns AI-based detection and classification in biomedical signal-processing applications rather than software testing, the paper explicitly treats these studies as methodological evidence for transferable AI patterns rather than direct empirical evidence for regression testing. The framework consequently emphasizes feature extraction, automated classification, adaptive prediction, and real-time decision support. A conceptual evaluation indicates that AI-assisted regression testing can improve the alignment between code changes and test execution priorities, reduce redundant execution, and create feedback loops capable of adapting to changing Agile projects. However, model drift, insufficient historical data, explainability, false prioritization, and integration complexity remain significant constraints. The analysis positions intelligent regression testing as an adaptive decision-support layer rather than a complete replacement for conventional testing practices.

The rapid evolution of cloud computing has transformed digital infrastructures from passive data storage environments into intelligent computational ecosystems capable of autonomous decision-making, adaptive optimization, and cognitive interaction. However, contemporary cloud systems frequently emphasize computational efficiency while lacking deeper conceptual models for knowledge interpretation, reasoning, and value-oriented intelligence. This research proposes a Philosophical Cognitive Computing Model (PCCM) that integrates Platonic conceptual frameworks with modern cloud intelligence mechanisms to enhance knowledge optimization, adaptive reasoning, and intelligent system governance. The study develops a theoretical and functional framework by examining philosophical foundations of knowledge, cognitive modeling, and scientific reasoning and mapping them into cloud-based computational architectures. The proposed model introduces cognitive knowledge layers, philosophical reasoning mechanisms, adaptive intelligence modules, and optimization strategies for intelligent cloud environments. Through analytical synthesis of existing philosophical and cognitive perspectives, the research demonstrates that integrating philosophical reasoning principles can improve interpretability, decision consistency, and human-centered intelligence in cloud computing systems. The findings suggest that philosophy-driven cognitive architectures provide a promising direction for developing next-generation intelligent cloud platforms capable of combining computational power with conceptual understanding.