Maintenance Planning Research Articles

Revolutionising Educational Management with AI and Wireless Networks: A Framework for Smart Resource Allocation and Decision-Making

Educational institutions face growing challenges. Rising enrolment, limited budgets, and sustainability goals demand more efficient resource management and administrative decision-making. To address challenges like these, this work proposes a conceptual framework for smart campus management which integrates Artificial Intelligence (AI) and advanced wireless networks based on 5G. The framework’s design outlines layers for campus data collection (via sensors and connected devices), high-speed communication, and AI-driven analytics for decision support. By leveraging data-driven insights enabled by reliable wireless connectivity, institutions can make more informed decisions, use resources more effectively, and automate routine tasks. Envisioned AI capabilities include forecasting (for predictive maintenance and demand planning), anomaly detection (for fault or irregularity identification), and optimisation (for resource scheduling). Rather than reporting empirical results, the framework is illustrated through hypothetical scenarios (e.g., anticipating equipment maintenance, dynamically scheduling classrooms, or reallocating resources) to present potential benefits and tools for researchers. The discussion also highlights how the framework incorporates data privacy, security, and accessibility considerations to ensure inclusive adoption. Eventually, this conceptual proposal provides a roadmap for administrators and planners, guiding the adoption of AI and wireless innovations in educational management to enable more responsive, efficient governance and, ultimately, improve outcomes for students and staff.

Reliability analysis of hospital infusion pumps: a case study

BackgroundInfusion pumps (IPs) are medical devices used for the continuous and precise delivery of medications or nutrients. Their use has expanded and is now widespread in emergency rooms, ICUs, pediatrics, and other hospital departments. Failures in IPs can lead to adverse events, compromising patient health. In addition to the risks to patients, IPs are the medical devices most frequently associated with reports of adverse events in Brazil, highlighting the need to monitor their operational conditions to minimize failures during use.ResultsThus, the objective of this research is to analyze the reliability of infusion pumps (IPs) in a Brazilian hospital using an internal database from Clinical Engineering software. Probability distributions for repair time and time between failures were modeled, and parameters such as reliability and availability were calculated, with a focus on investigating hospital departments with recurring failures.ConclusionIn evaluating the operating equipment, a lack of detail in failure notes and service order openings was observed, which can hinder maintenance planning. The longest repair times were recorded in the ICU (Neurology), which houses the majority of IPs. Graphical analysis and testing demonstrated that the Weibull distribution effectively models both time between failures and repair time. The IP A model showed better results in terms of availability and reliability, thereby improving the security of the IPs.

Spatially aware Markov chain‐based deterioration prediction of bridge components using a Graph Transformer

AbstractThis study proposes a Markov chain‐based deterioration prediction framework that incorporates spatial relationships between structural components. Despite spatial clustering and propagation of damage, conventional research has left spatial dependencies underexplored. This study constructs graph representations that reflect component adjacency and employs a Graph Transformer to capture both local and distant dependencies. Synthetic datasets confirm the advantage of introducing spatial positioning in settings with probabilistic transitions and various component topologies. The model is also tested on a semi‐automatically generated Tokyo girder bridge dataset. It improves precision sixfold over the percentage prediction method, surpasses a graph neural network, and outperforms a Transformer model without spatial information by five points on the real dataset and eight on a synthetic dataset. Attention weight analysis reveals that the model captures spatial dependencies and aligns with empirical deterioration mechanisms, offering interpretability. The proposed approach enables detailed element‐level deterioration predictions, enhancing maintenance planning and safety.

Modeling Ecological Risk in Bottom Sediments Using Predictive Data Analytics: Implications for Energy Systems

Sediment accumulation in dam reservoirs significantly impacts hydropower efficiency and infrastructure sustainability. Bottom sediments often contain heavy metals such as Cr, Ni, Cu, Zn, Cd, and Pb, which can pose ecological risks and affect water quality. Moreover, excessive sedimentation reduces reservoir capacity, increases turbine wear, and raises operational costs, ultimately hindering energy production. This study examined the ecological risk of heavy metals in bottom sediments and explored predictive approaches to support sediment management. Using 27 sediment samples from Zemborzyce Lake, the concentrations of selected heavy metals were measured at two depths (5 cm and 30 cm). Ecological risk index (ERI) values for the deep layer were predicted based on surface data using artificial neural networks (ANNs) and multiple linear regression (MLR). Both models showed a high predictive accuracy, demonstrating the potential of data-driven methods in sediment quality assessment. The early identification of high-risk areas allows for targeted dredging and optimized maintenance planning, minimizing disruption to dam operations. Integrating predictive analytics into hydropower management enhances system resilience, environmental protection, and long-term energy efficiency.

Human Digital Healthcare Engineering for Enhancing the Health and Well-Being of Seafarers and Offshore Workers: A Comprehensive Review

With over 50,000 merchant vessels and nearly two million seafarers operating globally, more than 12,000 maritime incidents in the past decade underscore the urgent need for proactive safety measures to ensure the structural integrity of aging ships and safeguard the well-being of seafarers, who face harsh ocean environments in remote locations. The Digital Healthcare Engineering (DHE) framework offers a proactive solution to these challenges, comprising five interconnected modules: (1) real-time monitoring and measurement of health parameters, (2) transmission of collected data to land-based analytics centers, (3) data analytics and simulations leveraging digital twins, (4) AI-driven diagnostics and recommendations for remedial actions, and (5) predictive health analysis for optimal maintenance planning. This paper reviews the core technologies required to implement the DHE framework in real-world settings, with a specific focus on the well-being of seafarers and offshore workers, referred to as Human DHE (HDHE). Key technical challenges are identified, and practical solutions to address these challenges are proposed for each individual module. This paper also outlines future research directions to advance the development of an HDHE system, aiming to enhance the safety, health, and overall well-being of seafarers operating in demanding maritime environments.

Availability predictions of solar power plants using multiple regression and neural networks: an analytical study

This analysis aims to develop an efficient mathematical model for prediction of the system availability of a solar photovoltaic power plant under the concept of redundancy and exponentially distributed random variables. For this objective, a stochastic model of the photovoltaic power plant is created with the help of the Markov birth-death technique. It is assumed that all the repairs are perfect and random variables statistically independent. The predictive techniques, namely regression analysis and artificial neural networks are used to predict the availability of the PV power plant in different experimental setups with the help of SPSS software. The impact of failure and repair rates on the availability of the PV power plant investigated. Experimental data used to calculate the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) of both predictive techniques. It is identified that the MAE and RMSE of the regression model are less in comparison to the ANN model. So, the regression model outperforms ANN in the performance prediction of PV power plants. The outcomes of this study may help design PV solar plants and plan maintenance strategies for solar PV power plants.

Deep Learning based enhanced aerial object detection

In congested urban environments, accurate detection and counting of humans and vehicles provide valuable insights for optimizing traffic flow, identifying congestion hotspots, and designing efficient transportation systems. By leveraging computer vision algorithms, such as deep learning based object detection models, real-time monitoring of pedestrian and vehicular traffic can be achieved with high accuracy and granularity. The ability to precisely quantify pedestrian and vehicle movements enables urban planners and policymakers to make data-driven decisions regarding infrastructure development, road maintenance, and public transit planning. In this work, we enhanced the existing deep learning based network architecture for object detection using UAV images. The enhanced network architecture can detect and give a count of the number of objects for any particular area in the image.

Development of an Ultrasonic Surface Roughness Meter for Road Maintenance: A Prototype for IRI Measurement

The importance of the road network in Indonesia as a vital infrastructure that connects various regions has made road maintenance a top priority in development planning. However, various challenges such as ineffective handling methods, limited experts, and minimal equipment have caused road management to not be optimal. Therefore, innovations are needed in road condition measurement, one of which is through the development of an ultrasonic sensor-based surface roughness measuring instrument as a prototype of International Roughness Index (IRI) measurement to support more accurate road maintenance evaluation and planning. The purpose of this research is to measure road roughness through IRI and pavement modulus values to improve road condition assessment.This study employs the International Roughness Index (IRI) to assess the functional condition of roads and the Pavement Modulus to evaluate the structural strength of the pavement. The IRI is measured through road surface roughness surveys using a roughness meter, with the results used to classify the severity of road damage. The IRI calculation is based on a quarter-car simulation model that utilizes vehicle dynamic parameters in response to road surface profiles, following the mathematical approach developed by Sayers, Gillespie, and Paterson (1986). The research results show that the prototype Ultrasonic Surface Roughness Meter was able to measure IRI values ranging from 4 to 8 at three different locations. These measurements fall within the "Good–Fair" classification, indicating relatively mild surface roughness. Based on these findings, the Directorate General of Highways recommends light rehabilitation and periodic maintenance, and the prototype device has the potential to serve as an effective, low-cost alternative for road condition monitoring, especially in areas with limited access to conventional IRI measurement tools.

Effect of Interface Adhesion Characteristics between Components of Embedded Tracks on Track Stability in Mountain Railways

In this study, numerical analyses were conducted to evaluate the effects of interface adhesion characteristics between embedded rail system (ERS) components on the stability of mountain railway tracks. The analysis results indicated that the interface adhesion characteristics between the elastic poured compound (EPC) and concrete panel significantly influence track stability. In particular, when the length of the non-bonded interface exceeds 2 m, the buckling stability markedly decreases, and the 1st buckling mode transitions from the lateral to the vertical direction. Based on these research results, the interface adhesion between components should be periodically inspected and a systematic maintenance plan should be established to prevent the degradation of embedded track stability.

Predicting Filter Medium Performances in Chamber Filter Presses with Digital Twins Using Neural Network Technologies

Efficient solid–liquid separation is crucial in industries like mining, but traditional chamber filter presses depend heavily on manual monitoring, leading to inefficiencies, downtime, and resource wastage. This paper introduces a machine learning-powered digital twin framework to improve the operational flexibility and predictive control of a traditional chamber filter press. A key challenge addressed is the degradation of the filter medium due to repeated cycles and clogging, which reduces filtration efficiency. To solve this, a neural network-based predictive model was developed to forecast operational parameters, such as pressure and flow rates, under various conditions. This predictive capability allows for optimized filtration cycles, reduced downtime, and improved process efficiency. Additionally, the model predicts the filter medium’s lifespan, aiding in maintenance planning and resource sustainability. The digital twin framework enables seamless data exchange between filter press sensors and the predictive model, ensuring continuous updates to the training data and enhancing accuracy over time. Two neural network architectures, feedforward and recurrent, were evaluated. The recurrent neural network outperformed the feedforward model, demonstrating superior generalization. It achieved a relative L2-norm error of 5% for pressure and 9.3% for flow rate prediction on partially known data. For completely unknown data, the relative errors were 18.4% and 15.4%, respectively. Qualitative analysis showed strong alignment between predicted and measured data, with deviations within a confidence band of 8.2% for pressure and 4.8% for flow rate predictions. This work contributes an accurate predictive model, a new approach to predicting filter medium cycle impacts, and a real-time interface for model updates, ensuring adaptability to changing operational conditions.

Total Station-Reflective Target Pier Deviation Measurement Error Control

In bridge engineering, monitoring pier offsets is crucial for ensuring both structural safety and construction quality. The total station measurement method using a reflector is widely employed, offering significant advantages in specific scenarios. During measurements, errors are influenced by various factors. Initially, misalignment causes the lateral relative error to increase before decreasing, while longitudinal relative errors fluctuate due to instrument characteristics and operational factors. Lateral movements have a more pronounced impact on these errors. Investigating the positioning layout of pier offsets holds substantial importance as it enables precise displacement monitoring, prevents accidents, aids in maintenance planning, provides valuable references for design and construction, and enhances the pier’s resistance to deflection. Controlling and correcting subsequent errors is essential to ensure the overall safety of the bridge structure.

Intelligent UAV swarm scheduling algorithm for urban inspection task

This research develops an intelligent UAV swarm scheduling algorithm to optimize urban infrastructure inspection processes by minimizing inspection time while ensuring comprehensive coverage. We formulate the challenge as a mixed integer non-linear programming problem and propose a decomposition approach addressing three critical components: structure-specific path planning, market-based task allocation, and conflict-free scheduling. Our methodology integrates these components through an iterative process within a hybrid centralized-decentralized architecture tailored for urban environments. Simulation results demonstrate that our algorithm reduces inspection time by 35% compared to single-UAV approaches while maintaining 98% coverage completeness. The approach exhibits 40% improved energy efficiency in limited-battery scenarios and polynomial-time computational complexity that scales efficiently with increasing swarm size. The algorithm typically converges within 3-5 iterations to near-optimal solutions. The proposed framework successfully balances inspection quality and resource efficiency while adapting to urban-specific challenges, including GPS degradation, obstacle avoidance, and structural complexity. Structure-specific inspection patterns significantly enhance efficiency across different infrastructure elements. This research advances UAV-based infrastructure monitoring capabilities, offering potential benefits for maintenance planning, public safety, and urban resilience. The computational efficiency makes the solution suitable for deployment on resource-constrained platforms typical in UAV applications.

Construction of distributed photovoltaic operation and maintenance knowledge base based on time series knowledge map

By constructing a distributed photovoltaic operation and maintenance knowledge base based on a time series knowledge graph, the problems of lack of specific knowledge base support and low efficiency in formulating operation and maintenance plans have been solved. The entity recognition method using multiple neural networks and attention mechanisms successfully recognized entities in text, and generated an operations knowledge graph through an entity relationship extraction model based on the JSA model. At the same time, by using incremental updates and redundancy processing methods, time series information is combined with the operation and maintenance knowledge graph to form an efficient and non redundant operation and maintenance knowledge base. The experimental results show that the similarity of the time series knowledge graph is low, always below 0.02, and the redundancy of operation and maintenance knowledge is small. The maximum power generation time of the operation and maintenance scheme is significantly shortened, thereby improving the power generation efficiency of the distributed photovoltaic operation and maintenance scheme.

LiDAR-Based Road Cracking Detection: Machine Learning Comparison, Intensity Normalization, and Open-Source WebGIS for Infrastructure Maintenance

This study introduces an innovative and scalable approach for automated road surface assessment by integrating Mobile Mapping System (MMS)-based LiDAR data analysis with an open-source WebGIS platform. In a U.S.-based case study, over 20 datasets were collected along Interstate I-65 in West Lafayette, Indiana, using the Purdue Wheel-based Mobile Mapping System—Ultra High Accuracy (PWMMS-UHA), following Indiana Department of Transportation (INDOT) guidelines. Preprocessing included noise removal, resolution reduction to 2 cm, and ground/non-ground separation using the Cloth Simulation Filter (CSF), resulting in Bare Earth (BE), Digital Terrain Model (DTM), and Above Ground (AG) point clouds. The optimized BE layer, enriched with intensity and color information, enabled crack detection through Density-Based Spatial Clustering of Applications with Noise (DBSCAN) and Random Forest (RF) classification, with and without intensity normalization. DBSCAN parameter tuning was guided by silhouette scores, while model performance was evaluated using precision, recall, F1-score, and the Jaccard Index, benchmarked against reference data. Results demonstrate that RF consistently outperformed DBSCAN, particularly under intensity normalization, achieving Jaccard Index values of 94% for longitudinal and 88% for transverse cracks. A key contribution of this work is the integration of geospatial analytics into an interactive, open-source WebGIS environment—developed using Blender, QGIS, and Lizmap—to support predictive maintenance planning. Moreover, intervention thresholds were defined based on crack surface area, aligned with the Pavement Condition Index (PCI) and FHWA standards, offering a data-driven framework for infrastructure monitoring. This study emphasizes the practical advantages of comparing clustering and machine learning techniques on 3D LiDAR point clouds, both with and without intensity normalization, and proposes a replicable, computationally efficient alternative to deep learning methods, which often require extensive training datasets and high computational resources.

Assessing the Impact of Infrastructure Degradation on Educational Quality in Haveli Kahuta, AJK: A Quantitative Study

School infrastructure serves as a cornerstone for educational quality, exerting a profound influence on both student learning outcomes and teacher effectiveness. This study investigates the impact of deteriorating school infrastructure on the quality of education in six strategically selected schools in Haveli Kahuta, Azad Jammu and Kashmir (AJK). Specifically, it explores critical aspects such as classroom conditions, the availability of basic facilities, and the overall physical learning environment. Employing a fully quantitative research design, data was gathered through structured, closed-ended questionnaires administered to 100 respondents comprising both students and teachers. The data, analyzed using SPSS, revealed consistent and troubling trends. Key infrastructural deficiencies—such as overcrowded and poorly ventilated classrooms, absence of electricity, broken furniture, and a lack of essential learning resources were found to significantly hinder the teaching-learning process. Students attending schools with substandard infrastructure reported low motivation, reduced concentration, and diminished academic performance. Teachers, on the other hand, faced obstacles in delivering quality instruction, citing inadequate teaching aids and uncomfortable working conditions as major constraints. The study underscores the critical need for systemic reforms and targeted policy measures. Recommendations include immediate renovation and upgrading of school facilities, provision of adequate teaching and learning materials, improvement in classroom ergonomics, and the development of sustainable infrastructure maintenance plans. By addressing these infrastructural challenges, stakeholders can foster an educational environment that promotes student engagement, supports effective pedagogy, and enhances overall academic achievement.

Smart Pothole Detection and Geospatial Visualization using Deep Learning and ArcGIS

This paper presents a complete system that combines artificial intelligence and geospatial analysis for pothole detection and demonstration on roads. Overcoming the inefficiencies of manual inspection of delay, time-consuming, and susceptible to human error, the system combines deep learning algorithms and ArcGIS to effectively detect potholes. Deep networks like VGG16, ResNet50, and Mask R-CNN are used in the model to enable accurate pothole detection on road surfaces. Moreover, integration with ArcGIS enables spatial mapping for better maintenance planning, while web interface with GPS support enables citizen engagement through real-time pothole reporting. The system performed quite well with an accuracy of 93% and F1 score of 95.86%. The Mask R-CNN model achieved 73% precision much higher than baseline models like YOLOv3. Results vindicate the system's potential as a viable, scalable, and community-based device for road monitoring and infrastructure maintenance.

Optimizing concrete pump maintenance in the construction sector using enhanced MCDM techniques

India’s construction sector has made significant contributions to national growth. Heavy machinery is vital to construction. A concrete pump is an essential piece of equipment used for producing, mixing, and pouring concrete. From an execution and maintenance perspective, this equipment is vital. In order to prevent malfunctions when pouring concrete, a better maintenance plan is required for the concrete pump. The example company currently struggles with determining which maintenance tasks should come first, which drives up maintenance costs, reduces machine availability and profitability, and reduces production. The current effort aims to enhance maintenance strategies by utilising several multi-criteria decision-making (MCDM) techniques, such as modified TOPSIS, VIKOR, and PROMETHEE, to identify the most critical reason for failure and enhance concrete pump maintenance decision-making. Three MCDM strategies are then employed to construct the maintenance criticality ranking after the weight of each criterion is calculated using the fuzzy SAW method. Using three MCDM approaches, the best maintenance strategy is determined out of nine likely causes of failure. According to modified TOPSIS, VIKOR, and PROMETHEE rankings, material wear out is the most critical cause of failure, followed by rubber quality, and high pumping strokes are the least critical factor.

Prediction and risk assessment of stress corrosion failures of prestressed anchors in underground mines

ABSTRACT Stress corrosion cracking (SCC) in deep underground anchor systems presents a growing threat to mining infrastructure integrity, driven by extreme environmental corrosion coupled with mechanical stress. To anticipate the progression of stress corrosion in deep anchor bolts, this study establishes a machine learning framework integrating 12 critical corrosion-influencing features with optimised support vector machine (SVM) modelling. The corrosion data collected regarding bolt failures in underground conditions furnished valuable references for the relevant research. The rigorous data cleaning methodologies are employed to refine the raw data sourced from anchoring materials. The statistical and machine learning models are utilised to execute data imputation and normalisation, facilitating the establishment of an SVM model tailored specifically for anchor material corrosion prediction. To enhance the predictive capabilities of the SVM algorithm, the model was optimised through the integration of principal component analysis and gradient boosting tree algorithms. The high accuracy of the model in predicting bolt corrosion risk was verified, and the weight of influences of environmental factors on corrosion failure are analysed. Key mechanistic insights reveal that stress and drip water flow rate dominate corrosion progression through synergistic electrochemical-stress interactions. The validated framework provides mine operators with a decision-support tool for proactive maintenance planning and corrosion-resistant anchoring system design in extreme environments.

Knowledge Graph-Augmented ERNIE-CNN Method for Risk Assessment in Secondary Power System Operations

With the increasing complexity of modern power systems, traditional risk assessment methods relying on expert experience and historical data face challenges in accuracy and adaptability. This study proposes a knowledge graph-augmented ERNIE-CNN method to enhance risk assessment in secondary power system operations. First, we construct a domain-specific knowledge graph by integrating expert knowledge and operational standards, which enhances semantic understanding and logical reasoning capabilities. Second, an improved ERNIE-CNN model is developed, incorporating an attention mechanism to effectively fuse semantic features and spatial patterns from operational texts. The experimental results on a dataset of 3240 secondary operation records demonstrate the model’s superior performance, achieving precision, recall, and F1-scores of 0.878, 0.861, and 0.869, respectively, outperforming benchmarks like BERT. Furthermore, a visualization of the knowledge graph is implemented, providing interpretable decision support for risk management. The proposed method offers a robust framework for automating risk assessment in power systems, with potential applications in smart grid maintenance and safety-critical operational planning.

Integrated Approach to Marine Engine Maintenance Optimization: Weibull Analysis, Markov Chains, and DEA Model

This study addresses the growing need for predictive maintenance in the maritime industry by proposing an optimized strategy for ship engine maintenance. The aim is to reduce unplanned failures that cause significant financial losses and disrupt global logistics flows. The methodology integrates Weibull reliability analysis, Markov chains, and Data Envelopment Analysis (DEA). A dataset of 512 diesel engine components from container ships was analysed, where the Weibull distribution (β = 1.8; α = 18,500 h) accurately modelled failure patterns, and Markov chains captured transitions between operational states (normal, degraded, failure). DEA was used to evaluate the efficiency of different maintenance strategies. Results indicate that targeting interventions in the degraded state significantly reduces downtime and improves component reliability, particularly for high-pressure fuel pumps and turbochargers. Optimizing maintenance extended the Mean Time to Failure (MTTF) up to 22,000 h and reduced the proportion of failures in critical components from 64.3% to 40%. These findings support a transition from reactive to proactive maintenance models, contributing to enhanced fleet availability, safety, and cost-effectiveness. The approach provides a quantitative foundation for predictive maintenance planning, with potential application in fleet management systems and smart ship platforms.