Alireza Ahmadi

Escalating track deterioration poses a significant challenge for infrastructure managers seeking to balance increasing capacity demands with long-term sustainability objectives. Although traffic loading is the primary driver of deterioration, existing deterioration models consider a wide range of additional factors whose representation varies across track forms. A comprehensive understanding of these factors is therefore necessary to support informed selection of railway track solutions. This study establishes a conceptual framework of deterioration factors for three railway track solutions: ballasted track, ballastless track, and combined track solutions (CTS). A systematic review of 30 deterioration models was conducted to identify deterioration factors and the mechanisms governing track degradation. Beyond synthesizing factors reported in existing deterioration models, the review identified five additional deterioration factors consistently associated with deterioration mechanisms in the broader railway engineering literature but not explicitly represented in the reviewed models. The review then evaluated the identified factors, grouped them by primary deterioration causes, and adapted them to reflect CTS-specific boundary conditions. The review revealed that existing deterioration models insufficiently address factors related to (i) properties of track subsystems and components, (ii) design characteristics of track forms, and (iii) design, location, and quantity of track transitions. Furthermore, deterioration-related parameters such as the elastic modulus of the track support and the amplitude and wavelength of differential settlement require specific consideration when assessing track form suitability within CTS. The resulting framework provides a structured conceptual representation of deterioration factors and their interrelationships, supporting future development of deterioration prediction models, life cycle cost assessments, and sustainability-oriented decision support methodologies for railway infrastructure.

The presence of silica in iron ore concentrate can have significant negative impacts on the efficiency and quality of steel production. As such, providing engineers with early and reliable information about the purity of iron ore concentrate is crucial for smooth mining operations. This paper reports on the development of a long short-term memory (LSTM) network and an ensemble data interpolation technique to enhance quality prediction in the froth flotation process of an iron ore mine. Our results demonstrate the ability of our model to accurately predict the silica content of iron ore concentrate on a minute-by-minute basis, as well as the ability to forecast hours in advance.

The study presents a decision support model for optimizing aircraft maintenance plans, targeting repairable units with hidden functions. The model addresses the complexities of integrating multiple factors, such as aging, the availability of hidden functions during operation, the varying effectiveness of restoration and repair tasks, and the operational and economic consequences of failure.It considers a combination of inspections, restorations, and a discard task at the unit’s predefined renewal age, for the “non-safety effect” failure category. It incorporates the Kijima type II model to account for the effect of imperfect restoration and the concept of mean fractional dead time to identify the optimal intervals for inspections, the number of inspections needed within each restoration cycle, and the frequency of restorations in a renewal cycle. Real-world adapted numerical examples are used to demonstrate the model’s application. Sensitivity analysis further enhances the model’s utility by determining the influence of reliability and cost parameters on maintenance optimization outcomes. The results show that, in the absence of aging, restoration effectiveness has a profound effect on the total cost and optimal set of solutions. With as-bad-as-old restoration effectiveness, the optimal alternative involves maximizing the number of inspections without resorting to restoration. Conversely, with as-good-as-new restoration effectiveness, the optimal alternative should harmonize a moderate number of inspections with a moderate number of restorations. Additionally, the influence of the costs associated with restoration and accidents suggests that frequent inspections at smaller intervals are essential to mitigate the risk of failure. Maintenance managers can utilize the proposed model as a practical tool to optimize maintenance by comparing and selecting strategies based on effectiveness and cost parameters, thereby enhancing the reliability of repairable systems with hidden failures.

The creation of optimal maintenance plans for repairable units with hidden functions poses a significant challenge during the development of aircraft-maintenance programs. This challenge stems from the complexity introduced by the need to simultaneously consider various factors, including the availability of hidden functions during operation, the varying effectiveness of restoration and repair effectiveness, degradation due to ageing and usage, inspection quality, the impacts of safety measures, and the operational and economic consequences of failure. This paper introduces a decision support model to identify the optimal maintenance plan for a repairable unit with hidden functions, one that considers a combination of inspections, restorations, and a discard task at the unit’s predefined renewal age. This model considers the “non-safety effect” failure category (e.g., operational disruptions or delays, higher maintenance costs, secondary damage to equipment). It incorporates the Kijima type II model to account for the effect of imperfect restoration and the concept of mean fractional dead time to identify the optimal intervals for failure-finding inspections (FFIs), number of inspections needed within each restoration cycle, and frequency of restorations in a renewal cycle. The proposed model also considers the costs associated with inspection, repair, restoration, and accidents stemming from multiple-failures. For a clearer understanding and application of the model, we employ real-world adapted numerical examples to elucidate the proposed optimization model and facilitate the exploration of maintenance strategies in the absence of comprehensive real-world aviation data. This approach serves as a crucial tool for initial model validation and lays the groundwork for future empirical research. In addition, a sensitivity analysis is also performed to evaluate the effects of changes in reliability and cost parameters and determine the parameters that most strongly affect the maintenance-optimization results. Numerical examples are particularly suited for such analyses, as they allow for controlled manipulation of parameters to observe their impacts on the model's outputs. The analysis results indicate that, in the absence of an in-unit ageing process, restoration effectiveness has a profound effect on the total cost and optimal set of solutions. In a maintenance plan with ABAO restoration effectiveness, the optimal alternative involves maximizing the number of FFIs without resorting to restoration. Conversely, in a maintenance plan with AGAN restoration effectiveness, the optimal alternative should harmonize a moderate number of inspections with a moderate number of restorations. Additionally, the influence of the costs associated with restoration and accidents suggests that frequent inspections at smaller intervals are essential to mitigate the risk of failure. Maintenance managers can leverage the proposed model to compare and select maintenance strategies that are tailored to different effectiveness and cost parameters. This model constitutes a practical tool for maintenance optimization and provides valuable insights for decision-makers looking to enhance the reliability of repairable systems with hidden failure points and compare the cost-effectiveness of maintenance plans.