This study examines the application of Artificial Intelligence (AI) functions in recruitment processes, specifically focusing on six AI functions, such as Expert System (ES), Machine Learning (ML), Robotic Process Automation (RPA), Natural Language Processing (NLP), Machine Vision (MV), and finally Speech Recognition (SR). The main objective was to determine how each AI function contributes to identified recruitment outcomes by automating repetitive tasks and supporting decision-making in talent acquisition. The research addresses a gap in the existing literature by synthesising current knowledge of the roles of individual AI functions in recruitment, through employing secondary data in the form of relevant publications. A qualitative archival research design was adopted, involving the systematic analysis of secondary sources, including academic journals, articles, web pages, industry reports, and technical publications from 1987 to 2025. As a point of departure, data were examined through a predefined codebook informed by the AI functions model proposed by Dejoux and Leon (2018). The analysis revealed that each AI function serves a distinct purpose in recruitment automation. The conceptual framework that emerges from secondary qualitative data analysis results, illustrating AI functions for recruitment, is measured by 17 independent variables grouped under 6 higher-level constructs (Expert system in recruitment, Machine learning in recruitment, Robotic in recruitment, NLP in recruitment, Machine vision in recruitment, and Speech recognition in recruitment). A moderating construct called adoption of AI function-specific system for recruitment, measured by 4 variables (Technology integration, Implementation quality, Human oversight, Ethical governance), as well as an outcome construct called recruitment outcome, measured by 18 dependent variables grouped under 4 higher-level constructs (Process Efficiency outcome, Decision Making quality outcome, Candidate experience outcome, and Reduction of Ethical and Bias errors). In the future, the conceptual framework could be operationalised in the form of a quantitative questionnaire and tested for validity and reliability based on a statistically significant sample of HR professionals involved in AI-powered recruitment processes across various industries.

Classic car restoration workshops operate as complex, resource-intensive environments in which activities are spatially distributed, energy-demanding, and tightly coupled to the physical organization of tools, parts, and vehicles. Despite this complexity, workshop management is often supported by fragmented digital tools—or none at all—which limits real-time visibility into resource consumption, space utilization, and operational flows. This lack of integrated, data-driven oversight constrains decision-making, reduces efficiency, and hampers the ability to analyze past operations or optimize workshop layout and processes.

To address these limitations, this paper presents the current stage of an ongoing Digital Twin research program developed for a real-world classic car workshop. The work extends a prior prototype into a more operational, data-driven management tool. Building on earlier research that introduced a BPMN-based process model aligned with restoration best practices defined in the FIVA Charter of Turin, the proposed system is structured around an interactive 3D representation of the workshop environment.

At this stage, the Digital Twin is enhanced with three main capabilities: (i) energy monitoring through IoT-based sensing, (ii) retrospective spatial analysis using heatmaps, and (iii) 3D warehouse visualization supporting vehicle-based part localization and layout management. These enhancements strengthen the integration between physical workshop operations and their digital counterpart, improving operational visibility, traceability, and historical analysis.

The implemented solution combines Unity3D-based visualization with BPMN-driven process services and dedicated data pipelines for energy and warehouse data. Controlled technical validation demonstrates that the proposed modules operate consistently and achieve adequate runtime performance, supporting the evolution of the Digital Twin into a robust, process-aware system for workshop monitoring, organization, and decision support.

Airports are increasingly adopting Industry 4.0 technologies to enhance operational efficiency, improve passenger comfort, and expand nonaeronautical revenue streams. Smart airport initiatives aim to integrate Internet of Things (IoT), mobile platforms and indoor positioning systems to optimize terminal operations and passenger flow. Despite these advances, terminal congestion and limited connection times present a strategic opportunity to redesign food service models at airport to enhance nonaeronautical revenue generation. This paper proposes a conceptual Food Delivery to Gate system enabled by indoor navigation system. To evaluate operational feasibility, the proposed framework is supported by a discrete-event simulation (DES) that analyzes delivery routing within the terminal to assess order feasibility under varying demand conditions. Simulation results demonstrate the operational feasibility of the proposed system under varying demand conditions, indicating its potential to meet boarding time constraints while enhancing passenger convenience and generating additional revenue.

This work presents a study on the use of a deep reinforcement learning model to optimize viewpoint planning in complex underwater environments, particularly in inspection scenarios relevant to the offshore oil industry. The DRQN model is trained from an environment representation and a defined objective, selecting the most appropriate viewpoint to maximize the visual quality of the target object. The proposed approach makes the viewpoint planning task more efficient and accurate. The results indicate that DRQN can be a valuable tool for improving inspection and monitoring in challenging environments. Although the focus of this work is underwater inspection, the proposed strategy can also be extended to other domains, such as robotics, autonomous vehicles, and computer vision systems in general.