Mario Caterino

Purpose: In the face of a competitive industrial landscape, optimising maintenance strategies is critical for manufacturing companies striving to improve operational efficiency, safety and cost-effectiveness. Maintenance strategy selection (MSS) is crucial in mitigating downtime and enhancing system performance, where inadequate strategy selection can increase costs and risk. This paper investigates the current methodologies for selecting maintenance strategies, focusing on economic, technical and safety criteria that challenge decision-makers. Design/methodology/approach: Through a systematic literature review (SLR), recent developments in maintenance strategies are explored, identifying key approaches such as preventive, predictive and condition-based maintenance alongside newer predictive models powered by machine learning and artificial intelligence (AI). Findings: This study highlights the widespread use of MCDM methods – especially AHP, TOPSIS and DEMATEL – in supporting MSS, often enhanced with fuzzy or hybrid approaches. Traditional strategies like corrective and time-based maintenance remain common, but there is a growing shift towards data-driven approaches such as condition-based and predictive maintenance. Technical and economic criteria are the most frequently used, with safety and feasibility also playing key roles. Recent literature shows increased focus on environmental social and governance related factors like energy consumption and sustainability. The analysis reveals a trend from static decision models towards integrated, AI-supported frameworks, reflecting evolving industrial priorities. Originality/value: This study provides a comprehensive overview of state of the art maintenance strategies, offering insights for industrial managers and researchers aiming to refine decision-making in maintenance management.

Predictive maintenance (PdM) plays a key role in improving the reliability and efficiency of industrial systems by anticipating abnormal operating conditions through data-driven approaches. In this context, many PdM solutions rely on temporal windowing strategies applied to multivariate time series; however, the choice of observation and prediction horizons is often fixed a priori and rarely analyzed in a systematic way. This paper investigates the impact of temporal window design by formulating alarm prediction as a window-based classification problem, explicitly distinguishing between the observation horizon and the prediction horizon. A comprehensive experimental study is conducted on three real-world industrial datasets characterized by different process dynamics. Multiple machine learning and deep learning models are evaluated across a wide range of combinations of observation and prediction horizons. The results show that predictive performance is strongly influenced by the joint configuration of observation and prediction horizons, rather than by their individual selection. While increasing the prediction horizon generally reduces accuracy, the magnitude of this effect depends on system dynamics. Similarly, longer observation horizons substantially improve performance in processes with slow dynamics, while providing limited benefits in faster-evolving systems. Overall, the study highlights the importance of explicitly modeling temporal window parameters in predictive maintenance design, providing both methodological insights and practical guidelines for selecting appropriate temporal configurations in industrial alarm prediction tasks.

The use of robots in industrial environments has become standard practice over the past 40 years. Advances in technology have enabled robots to actively interact with humans, thereby creating opportunities for enhanced collaboration. In remanufacturing, human–robot interaction (HRI) has contributed significantly to improving productivity and ergonomics, particularly during the disassembly phase, a critical stage due to the varying quality of returned products. However, the cognitive impact of robots on human operators performing disassembly tasks remains an underexplored area in existing literature. This paper presents an empirical study to address this gap. The study was conducted in a virtual reality (VR) remanufacturing environment. An experiment was designed involving 82 participants who completed a disassembly task in a shared workspace while a robot simultaneously performed other tasks. Following task execution, the participants completed a questionnaire to evaluate their subjective perceptions and experiences during the interaction. The findings indicate that the robot’s presence had minimal overall influence on the participants, with a few exceptions. Specific characteristics of the robot, such as noise level and proximity, were identified as moderately significant factors, particularly among female participants, who were the most impacted by these factors, reporting the highest levels of disturbance and distraction. These results provide valuable insights into the nuanced role of robotic features in shaping human perception during HRI. This study contributes to advancing knowledge in HRI by highlighting the need for further research on cognitive impacts and subjective experiences in shared human–robot remanufacturing workspaces. Graphical Abstract

This study investigated the impact of different reverse logistics strategies on the economic and environmental performance of a system within the rubber flooring sector. A simulation tool was developed to replicate the behavior of a real production system, focusing on the transition from linear to circular processes. By considering multiple factors influencing system performance, this research offers an overview of the sustainability of various RL strategies and provides realistic estimates for different scenarios. Three key factors were used to evaluate each strategy’s response: transportation distance, flooring thickness, and returned flooring quality. The findings suggest that an environmental advantage generally favors on-site inspections at the customer’s location to assess the returned product’s condition, regardless of distance. However, centralizing inspections at the manufacturer’s facility is more economically advantageous when distances are short, particularly when the company prioritizes recycling over other circular economy practices. Based on these results, practical implications and guidelines are proposed to help companies balance cost-effectiveness with sustainability, optimizing their operations within a circular economy framework.

Background: Emergency scenarios present unprecedented challenges for supply chains worldwide, particularly in the management and distribution of critical supplies, where timely delivery and maintaining integrity are crucial. Methods: This article explores an innovative approach to enhance the emergency management of supply chains using blockchain technology and simulation-based modelling. The proposed methodology aims to tackle issues such as transparency, efficiency, and security, which are vital for managing logistics during crises. A case study involving a vaccine rollout is used to demonstrate how blockchain can optimise supply chain operations, reduce bottlenecks, and ensure better traceability and accountability throughout the process. The case study is specifically developed based on the distribution of COVID-19 vaccines in Italy. Results: The integration of blockchain technology not only enhances data integrity and security but also facilitates real-time monitoring and decision-making. Conslusions: The findings suggest that the proposed blockchain-based model can significantly improve supply chain resilience in emergency situations compared to traditional methods, thereby offering valuable insights for policymakers and supply chain managers facing future crises.

This study investigates the impact of different reverse logistics (RL) strategies on the economic and environmental performance of a system within the rubber flooring sector. A simulation tool was developed to replicate the behavior of a real production system, focusing on the transition from linear to circular processes. By considering multiple factors influencing system performance, this research offers an overview of the sustainability of various RL strategies and provides realistic es-timates for different scenarios. Three key factors were identified to evaluate each strategy’s re-sponse: transportation distance, flooring thickness, and returned flooring quality. The findings suggest that the environmental advantage generally favors on-site inspections at the customer’s location to assess the returned product's condition, regardless of distance. However, centralizing inspections at the manufacturer’s facility is more economically advantageous when distances are short, particularly when the company prioritizes recycling over other CE practices. Based on these results, practical implications and guidelines are proposed to help companies balance cost-effectiveness with sustainability, optimizing their operations within a circular economy framework.