Planning maintenance in manufacturing systems is challenging, especially when dealing with unpredictable machine breakdowns. This paper presents a Deep Reinforcement Learning (DRL) framework to automate and optimise maintenance decisions. Our approach uniquely models a realistic industrial environment where machine failures are stochastic and can occur in succession, a critical factor often simplified in traditional methods. The DRL agents learn to make decisions using local machine data combined with key system-wide performance metrics, enabling a modular yet globally-aware strategy. We benchmarked our DRL policy against a state-of-the-art Metaheuristic Genetic Algorithm (MGA). The results demonstrate that our DRL approach achieves two key advantages. First, it matches the production throughput of the benchmark, particularly under moderate operational stress. Second, it significantly reduces the overall maintenance workload and enhances system robustness against the variability of machine failures. This work highlights the potential of DRL to create intelligent, decentralized maintenance strategies that improve both efficiency and resilience in complex, high-variability industrial environments.

This study addresses the scheduling problem in a single-machine system with variable processing speed. The performance of the Earliest Due Date (EDD) rule, operating at a fixed maximum speed, is compared with a Reinforcement Learning (RL)-based approach capable of dynamically adapting both processing speed and job selection. The analysis considers both temporal performance (tardiness) and sustainability (energy consumption). Simulation results, supported by statistical analysis, show that the RL approach achieves a significant reduction in total energy consumption compared to the EDD rule. Regarding temporal performance, analysis of variance reveals that the chosen policy significantly affects the number of tardy jobs, with an effect that depends on the tightness of job deadlines, whereas the overall average tardiness across all jobs does not differ significantly between the two policies. These findings highlight the potential of adaptive RL-based policies for more sustainable resource management, demonstrating the ability to achieve substantial energy savings while maintaining competitive temporal performance and dynamically managing the trade-off according to deadline pressure.

Maintenance scheduling is critical function across numerous industries, where the dynamic complexity of operations often challenge the systems efficiency. This study explores the application of Deep Reinforcement Learning (DRL) to refine scheduling decisions by integrating a simulation tool that replicates an industrial production line. This integration eases the modelling and simulation in real-time of machine operations, job flows, and maintenance activities, capturing the dynamics and complexities of a typical production environment. Our developed tool assesses various maintenance strategies and their direct impacts on productivity, leveraging DRL to enhance decision-making capabilities. We introduce an innovative job-sequencing rule that complements the DRL framework, systematically analysing its effectiveness against traditional heuristic methods. The comparative analysis confirms that our DRL-based approach, coupled with the job sequencing rule, significantly optimises maintenance timing and resource allocation in a flow shop setting. By synthesizing simulation with intelligent algorithms, our method not only optimize maintenance tasks but also boosts overall production efficiency.

In today’s environment, characterized by high complexity and volatility of demand, responsiveness, quality, and timeliness in the transmission of information between all parties involved in Supply Chain operations, are critical aspects to manage. In this context, the most successful companies have developed an integrated view of the Supply Chain to improve its efficiency. The realization of these objectives is achieved through adopting Supply Chain management methods and tools appropriate to their operations, with a view to continuous improvement through data analysis and forecasting. The difficulty lies in intercepting and organising data from disparate sources, multiple data sets provide incomplete information that inaccurately represents the performance and service levels received by suppliers and offered to customers, caused by the competitive nature of different companies in wanting to keep their information confidential. For this reason, this work proposes an ontological Supply Chain model with a governance element that enables the information exchange, preventing misreporting behaviour by different companies and optimising the parameters of the entire Supply Chain. In addition to the definition of all major incoming and outgoing information flows that characterise the relationships and performance of the Supply Chain actors as individual elements and as a whole.

To stay competitive, modern market scenarios are forcing a radical shift in the manufacturing concept, focusing companies’ attention on customer satisfaction through increased product customization and quick response strategies. Significant progress has been made in the field of Industry 4.0 technologies, but there is still an open gap in the literature regarding methodologies for efficiently managing a manufacturing system’s available productive resources. Spearman et al. proposed the CONtrolled Work-In-Progress (CONWIP) production logic, which allows controlling Work-In-Progress (WIP) in a production system while monitoring throughput. However, in order to face with the increased variability that enters into the production system, an affordable performances estimation tool is still required. Taking advantages of the recent innovation in the field of machine learning, this paper contributes to the development of a tool for estimating the performance of a production line using a deep learning neural network. The results demonstrated that the proposed estimation tool outperforms the current best-known mathematical model when estimating the throughput of a CONWIP Flow-Shop production line with a given processing time distribution and WIP value.