Application of Artificial Intelligence and Machine Learning for Predictive Maintenance in Academic Library Information Service Supply Chains

Cite this article as: Bello, M. I., Adamu, R., & Abubakar, S. (2026). Application of artificial intelligence and machine learning for predictive maintenance in academic library information service supply chains. Sokoto Journal of Linguistics and Communication Studies (SOJOLICS), 2(1), 1-10. https://doi.org/10.36349/sojolics.2026.v02i01.001

APPLICATION OF ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING FOR PREDICTIVE MAINTENANCE IN ACADEMIC LIBRARY INFORMATION SERVICE SUPPLY CHAINS

By

Musa Ibrahim Bello
Department of Computer Science
Kano State Polytechnic, Kano state, Nigeria
ORCID: https://orcid.org/0009-0002-2379-8169
danbattasaraki@gmail.com
08025852051

&

Rilwanu Adamu (PhD)
SPS Library
Northwest University, Kano
radamu@yumsuk.edu.ng
08035679267

&

Samira Abubakar
Sule Hamma Library Complex
Northwest University, Kano
sabubakar@yumsuk.edu.ng
08032160272

Abstract

Predictive maintenance has become an increasingly important application of machine learning across many sectors, including supply chain management and information-based service organizations such as academic libraries. Modern academic libraries rely heavily on interconnected digital infrastructures such as Integrated Library Systems (ILS), servers, RFID technologies, self-service circulation machines, and network facilities to support core services like access, circulation, and resource management. Together, these components form a library information service supply chain. When any part of this system fails, it can disrupt services, reduce operational efficiency, and negatively affect user satisfaction. This study examines the application of machine learning–driven predictive maintenance within the context of academic library information service supply chains. In many libraries, maintenance practices remain largely reactive or based on fixed schedules, which often leads to unexpected system breakdowns and unplanned downtime. Such failures not only interrupt daily library operations but can also trigger wider service disruptions across interconnected systems. This study therefore explores how predictive maintenance can help libraries anticipate potential failures and respond proactively. The study reviews various machine learning techniques used in predictive maintenance, including supervised learning, unsupervised learning, and deep learning approaches. It also examines data sources relevant to library environments, such as system usage logs, maintenance records, sensor-generated data, and environmental conditions that influence equipment performance. In addition, the study discusses key challenges associated with implementing predictive maintenance in academic libraries, including data quality and integration issues, cost constraints, real-time decision-making requirements, and limitations in technical expertise. Finally, the study highlights the potential role of emerging technologies such as the Internet of Things (IoT) and edge computing in improving system reliability, operational efficiency, and the overall resilience of academic library information service supply chains.

 

Keywords: Predictive Maintenance, Machine Learning, Artificial Intelligence, Academic Libraries Supply Chain, Internet of Things (IoT)

 

1. Introduction

The term supply chain refers to the interconnected network of entities involved in producing, distributing, and delivering goods, encompassing merchants, distributors, transporters, manufacturers, and ultimately the end consumers. Understanding actual consumer demand and consumption patterns is crucial, as customers represent the central node of any supply chain, driving businesses to produce and distribute goods accordingly.

Effective supply chain management requires collaboration and coordination among all entities within the network to align operations with real demand. Such cooperation not only ensures timely delivery of products but also reduces overall operational costs. In contrast, the absence of effective partnerships often results in a gap between the ideal and actual performance of supply chain networks. Ideal supply chains operate efficiently, maximizing resource utilization, whereas real-world supply chains frequently encounter inefficiencies due to factors such as misaligned business objectives, inadequate long-term relationship management, limited information sharing, the complexity of large-scale operations, staff competence, performance management challenges, and insufficient incentive systems.

Predictive maintenance, also referred to as condition-based maintenance, is a proactive approach that leverages data analysis and machine learning technologies to anticipate equipment failures and schedule timely repairs. This approach is particularly valuable in supply chain contexts, where unexpected equipment breakdowns can trigger cascading disruptions, affecting production schedules, delivery timelines, and overall operational efficiency.

 

Academic libraries increasingly operate as complex information service supply chains, relying on interconnected digital infrastructures such as Integrated Library Systems (ILS), OPACs, digital repositories, RFID technologies, servers, and network facilities to deliver timely and reliable services to users. These systems support core library operations including acquisition, cataloguing, circulation, and access to electronic resources.However, the growing dependence on ICT infrastructure has exposed academic libraries to frequent system failures, unexpected equipment breakdowns, and service disruptions, which negatively affect information access, user satisfaction, and operational efficiency. Traditional reactive and preventive maintenance approaches often fail to detect early warning signs of system malfunction, leading to downtime and cascading service failures across the library service chain.

 

Therefore, the emergence of Artificial Intelligence (AI) and Machine Learning (ML) brought apowerful technology capable of analyzing large volumes of operational data to predict potential system failures before they occur. Through predictive maintenance, libraries can anticipate equipment breakdowns, optimize maintenance schedules, reduce downtime, and enhance the resilience of their information service supply chains.Despite the growing global interest in smart libraries, empirical studies on the application of AI and ML for predictive maintenance in academic libraries particularly within developing nations remain limited. This study therefore seeks to examine how AI- and ML-driven predictive maintenance can be applied to strengthen operational efficiency and service continuity in academic library information service supply chains.

The primary objectives of this study are to:

Examine the effectiveness of predictive maintenance in various supply chain applications.

Explore the role of predictive maintenance within the framework of supply chain management.

Identify areas within the library service supply chain where predictive maintenance can be applied.

Assess the effectiveness of AI and ML techniques in predicting system and equipment failure in academic libraries.

Discuss the results and implications of implementing predictive maintenance strategies for operational efficiency and resilience.

 

2.0 Literature Review

2.1  Artificial Intelligence (AI)

Artificial intelligence (AI) is broadly defined as computer-based systems that perform tasks associated with human cognition, such as perception, reasoning, learning, and decision-making (Cox & Mazumdar, 2022; Kühl et al., 2022). Historically, AI has evolved from symbolic, rule-based systems to modern data‑driven and deep learning approaches (Ambani & Rathod, 2025). In library contexts, AI is treated as an umbrella of technologies (e.g., expert systems, machine learning, natural language processing, robotics, &chatbots) used in technical services, public services, and smart-library infrastructures (Asemi et al., 2020; Das & Islam, 2021; Gasparini & Kautonen, 2022). Early library applications focused on expert systems and rule‑based tools for cataloguing and reference, while current work emphasizes intelligent agents, recommender systems, conversational agents, and AI‑enhanced security and infrastructure (Asemi et al., 2020; Das & Islam, 2021).

 

2.2  Machine Learning (ML)

Machine learning (ML) is commonly defined as a subfield of AI in which algorithms learn from data to improve performance on tasks without being explicitly programmed for each rule (Kühl et al., 2022; Mahesh, 2020; Rahaman, 2024; Shaveta, 2023). Core paradigms include supervised learning (e.g., regression, support vector machines, k‑nearest neighbors), unsupervised learning (e.g., clustering, principal component analysis, association rules), and reinforcement learning, in which agents learn optimal actions via rewards and penalties (Abdullah et al., 2025; Mahesh, 2020; Naeem et al., 2023; Rahaman, 2024; Shaveta, 2023; Tufail et al., 2023). In libraries, ML underpins predictive analysis, such as forecasting resource use, supporting recommender systems, and automating metadata generation and text recognition (Das & Islam, 2021; Gasparini & Kautonen, 2022).

 

2.3  Predictive Maintenance

Predictive maintenance (PdM) is a condition‑based maintenance strategy that uses sensor data and analytical models to detect early signs of degradation and predict failures so that interventions are performed just in time (Moleda et al., 2023; Nunes et al., 2023; Selçuk, 2017; Zonta et al., 2020). Compared with reactive maintenance, which acts only after failure, and time-based preventive maintenance, which follows fixed schedules, PdM aims to reduce unplanned downtime and unnecessary maintenance by basing decisions on the actual health state of equipment (Moleda et al., 2023; Selçuk, 2017; Zonta et al., 2020). In service‑oriented environments, including building and infrastructure systems that support libraries and campuses, PdM improves reliability, safety, energy efficiency, and life‑cycle cost by integrating IoT, AI, and remote monitoring (Moleda et al., 2023; Selçuk, 2017; Zonta et al., 2020).

 

2.4  Academic Library Information Service Supply Chains

In services, supply chain concepts describe the coordinated management of flows of resources, information, and activities from upstream suppliers to end users to enhance value and reduce cost (Mahesh, 2020; Tufail et al., 2023). Academic libraries can be conceptualized as information service supply chains, linking content suppliers (publishers, aggregators, vendors, technology providers) through internal processes (acquisitions, cataloguing, licensing, digital curation) to downstream service delivery (circulation, reference, digital platforms, user training) for stakeholders such as students, faculty, and researchers (Asemi et al., 2020; Cox & Mazumdar, 2022; Das & Islam, 2021; Gasparini &Kautonen, 2022). In this view, managing the library’s information service supply chain involves integrating partners and processes, using data and AI/ML tools to improve responsiveness, personalization, and efficiency across the entire value network (Asemi et al., 2020; Cox & Mazumdar, 2022; Das & Islam, 2021; Es‑sakali et al., 2022; Moleda et al., 2023; Tufail et al., 2023).

 

3.  Methodology

This study employs a qualitative approach, highlighting the effective application of machine learning in predictive maintenance within supply chain management through the use of case studies and real-world examples. These examples demonstrate how predictive maintenance enhances equipment uptime, reduces maintenance costs, optimizes spare parts inventory management, and strengthens overall supply chain resilience.

 

The paper further explores potential advancements in the field, including the integration of predictive maintenance with broader supply chain optimization strategies, the development of hybrid machine learning models, and the incorporation of explainable artificial intelligence to improve interpretability of predictive maintenance recommendations.

 

Traditionally, maintenance strategies in supply chains are categorized as either reactive or preventive. Reactive maintenance addresses issues only after they occur, often resulting in unanticipated downtime, increased repair costs, and operational disruptions. Preventive maintenance involves performing scheduled maintenance at regular intervals, regardless of equipment condition. However, overly frequent or insufficient maintenance can either disrupt operations or fail to prevent unplanned failures.

 

Predictive maintenance introduces a proactive alternative. By analyzing historical data, sensor readings, and other relevant parameters, machine learning algorithms can identify patterns and early signs of equipment degradation or potential failure. These models generate real-time forecasts regarding equipment health and the likelihood of future malfunctions, enabling maintenance teams to intervene precisely when necessary. This approach achieves an optimal balance between maintenance costs and equipment availability.

 

The advantages of predictive maintenance in supply chain management are numerous:

Minimized Downtime: Anticipating and preventing equipment failures ensures smoother operations across the supply chain.

Cost Reduction: Proactive maintenance reduces the need for expensive emergency repairs and unnecessary preventive tasks.

Resource Optimization: Maintenance efforts focus on components that require attention, maximizing resource efficiency.

Enhanced Resilience: Supply chains become more capable of handling unexpected events or fluctuations in demand.

Extended Equipment Lifespan: Timely maintenance based on predictive insights prolongs machinery and equipment service life.

Data-Driven Decision-Making: Predictive maintenance provides critical insights into equipment condition, operational patterns, and areas for improvement.

To implement predictive maintenance effectively, organizations must first gather relevant data from equipment sensors and historical maintenance records. The data is then preprocessed, and machine learning models are trained to detect correlations between observed patterns and potential equipment failures. Once trained, these models continuously monitor real-time data to generate actionable predictions.

At its core, predictive maintenance transforms traditional maintenance practices from reactive and time-based approaches to a proactive, data-driven framework, resulting in more efficient, reliable, and robust supply chain operations.

 

4.  Results and Discussion

The application of machine learning in predictive maintenance has demonstrated significant benefits across various sectors of the supply chain, enhancing operational efficiency, reducing costs, and improving overall resilience. By leveraging historical data and real-time monitoring, predictive maintenance models can forecast when equipment is likely to fail, enabling timely interventions and minimizing disruptions.

 

4.1  Applications Across Industries

Industrial and Manufacturing Sector: Automobile manufacturers employ predictive maintenance to monitor assembly line robots and machinery, preventing production interruptions and reducing associated costs.

Aerospace Industry: Airlines and aircraft manufacturers use predictive maintenance to monitor engines and critical components, ensuring safety and adherence to flight schedules.

Power and Utilities: Utility firms apply predictive maintenance to turbines, generators, and transformers, reducing unplanned outages and enhancing operational efficiency.

Oil and Gas Industry: Predictive maintenance monitors pumps, compressors, and pipelines, preventing leaks, reducing downtime, and safeguarding worker safety.

Transportation and Logistics: Fleet management and railway operators use predictive maintenance to monitor vehicles, tracks, and signals, minimizing failures, improving delivery reliability, and ensuring passenger safety.

Construction and Mining: Heavy equipment such as excavators and haul trucks are monitored to maximize operational uptime and reduce maintenance costs.

Healthcare: Hospitals implement predictive maintenance for critical equipment, including MRI scanners and ventilators, ensuring continuous operational readiness.

Retail and E-Commerce: Distribution centers use predictive maintenance to monitor conveyor systems and automated sorting machinery, preventing delays in order processing and deliveries.

Telecommunications: Predictive maintenance ensures continuous operation of cell towers and network infrastructure, reducing service interruptions.

Food and Beverage Industry: Manufacturing and packaging machinery are monitored to maintain product quality and reduce production downtime.

Pharmaceutical Industry: Equipment used in drug manufacturing is monitored to comply with regulatory standards and maintain operational reliability.

 

 

 

 

4.2  Impact on Supply Chain Performance

The use of predictive maintenance contributes to several key outcomes:

Reduced Downtime: By forecasting equipment failures, organizations can prevent unscheduled stoppages, resulting in smoother production schedules, timely deliveries, and higher customer satisfaction.

Cost Savings: Minimizing emergency repairs and unnecessary preventive maintenance reduces overall maintenance expenses while improving resource allocation.

Optimized Maintenance Operations: Maintenance teams can target interventions precisely when needed, improving labor and resource efficiency.

Extended Equipment Lifespan: Proactive maintenance extends the service life of machinery and reduces capital expenditures.

Inventory Optimization: Predictive insights enable better management of spare parts inventory, decreasing excess stock costs while ensuring availability.

Enhanced Supply Chain Resilience: Fewer interruptions caused by equipment failures strengthen the supply chain’s ability to respond to unexpected events and demand fluctuations.

These findings demonstrate that predictive maintenance, powered by machine learning, transforms traditional maintenance approaches from reactive or time-based models into proactive, data-driven strategies. Across industries, this transformation results in more efficient, reliable, and resilient supply chain operations, ultimately supporting business continuity and competitive advantage.

These applications and impacts have been displayed in the following diagram:

 

 

 

5. Conclusion

The increasing globalization of trade and rapid technological advancement have intensified competition among firms, compelling businesses to continuously innovate to maintain market share and enhance revenue. Within this context, automation and robotics have established a critical role across diverse industries by performing repetitive and labor-intensive tasks, thereby supplementing human efforts. Predictive maintenance, powered by machine learning, exemplifies the transformative potential of artificial intelligence in optimizing supply chain operations, enhancing efficiency, and reducing costs.

 

Looking forward, artificial intelligence is expected to achieve even greater levels of sophistication, fostering deeper collaboration between humans and machines. This synergy has the potential to drive groundbreaking improvements not only in supply chain management but also across other critical sectors. By leveraging AI-driven insights, organizations can enhance operational resilience, optimize resources, and maintain a competitive edge in an increasingly dynamic business environment.

 

6.  Recommendations

Academic libraries should adopt predictive maintenance practices proven effective in other service-based supply chains and conduct pilot implementations to compare their performance with traditional maintenance methods.

Predictive maintenance should be integrated into library supply chain management policies, with improved collaboration between librarians, ICT units, vendors, and institutional management.

Libraries should apply predictive maintenance to critical service components such as ILS, servers, OPAC platforms, RFID systems, digitization equipment, and environmental control systems.

Academic libraries should invest in AI- and ML-based tools for analyzing system logs and usage data, alongside staff training and technical partnerships to improve failure prediction accuracy.

Library management should use predictive maintenance insights to enhance operational efficiency, reduce service disruptions, and strengthen the resilience of academic library information service supply chains.

 

References:

Asemi, A., Ko, A., &Nowkarizi, M. (2020). Intelligent libraries: A review on expert systems, artificial intelligence, and robot. Library Hi Tech, 38(3), 1–22.

Cox, A. M., & Mazumdar, S. (2022). Defining artificial intelligence for librarians. Journal of Librarianship and Information Science, 54(4), 779–794.

Das, R. K., & Islam, M. S. U. (2021). Application of artificial intelligence and machine learning in libraries: A systematic review. arXiv.

Es‑sakali, N., Cherkaoui, M., Mghazli, M. O., & Naimi, Z. (2022). Review of predictive maintenance algorithms applied to HVAC systems. Energy Reports, 8, 12601–12626.

Gasparini, A., &Kautonen, H. (2022). Understanding artificial intelligence in research libraries: Extensive literature review. LIBER Quarterly, 32(1), 1–36.

Mahesh, B. (2020). Machine learning algorithms – A review. International Journal of Science and Research, 9(1), 381–386.

Moleda, M., Małysiak‑Mrozek, B., Ding, W., Sunderam, V., & Mrozek, D. (2023). From corrective to predictive maintenance—A review of maintenance approaches for the power industry. Sensors, 23(13), 1–35.

Naeem, S., Ali, A., Anam, S., & Ahmed, M. (2023). An unsupervised machine learning algorithm: Comprehensive review. International Journal of Computing and Digital Systems, 12(1), 1–21.

Nunes, P., Santos, J., & Rocha, E. (2023). Challenges in predictive maintenance – A review. CIRP Journal of Manufacturing Science and Technology, 42, 1–18.

Rahaman, M. J. (2024). A comprehensive review to understand the definitions, advantages, disadvantages and applications of machine learning algorithms. International Journal of Computer Applications, 186(5), 1–10.

Selçuk, Ş. (2017). Predictive maintenance, its implementation and latest trends. Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, 231(9), 1670–1679.

Shaveta. (2023). A review on machine learning. International Journal of Science and Research Archive, 8(6), 1–7.

Tufail, S., Riggs, H., Tariq, M., & Sarwat, A. I. (2023). Advancements and challenges in machine learning: A comprehensive review of models, libraries, applications, and algorithms. Electronics, 12(7), 1–40.

Zonta, T., Costa, C. A., Righi, R. R., Lima, M. J., da Trindade, E. S., & Li, G.‑P. (2020). Predictive maintenance in the industry 4.0: A systematic literature review. Computers & Industrial Engineering, 150, 106889.