Maintenance of trains has a major impact on their availability, safety and cost-effectiveness. To date, planning-intensive processes in train maintenance workshops have been carried out with traditional static planning systems. While these systems have provided a foundation for organizing maintenance work, they are unable to generate optimized maintenance programs. As a result, unplanned additional tasks frequently arise during the actual execution of the maintenance operations, thereby disrupting workflows in maintenance workshops and reducing both economic and operational efficiency. It has been demonstrated that such disruptions often lead to increased maintenance costs, longer lead times, and reduced plant efficiency. Therefore, there is a growing need to shift from static to more dynamic, data-driven approaches for maintenance planning and control in the railway sector. Artificial Intelligence (AI) offers significant potential in this regard, as it facilitates the development of predictive capabilities, dynamic scheduling, and intelligent resource allocation. To this aim, this work presents a conceptual framework designed to enhance maintenance order planning and control in train maintenance workshops. The proposed framework serves as a foundational model for utilizing the AI technologies to reduce inefficiencies, improve scheduling, and ultimately support timely and cost-effective maintenance planning and control strategies for trains. The framework remains conceptual at this stage, which is the main limitation of the work; however, as future work, its practical effectiveness will be explored through implementation in trains maintenance workshop and empirical validation.
Hayat et al. (Thu,) studied this question.