Predictive Reliability-Driven Optimization of Spare Parts Management in Aircraft Fleets Using AI, IoT, and Digital Twin Technologies

Effective handling of spare parts is important for aircraft maintenance since it can control both the cost and availability of the fleet. According to the International Air Transport Association [1], spare parts can make up 30-40% of the total costs for airline maintenance, making them one of the largest contributors to the airline’s operating budget. Despite the heavy financial burden, many airlines still manage spares by using outdated approaches that take a long time, are affected by unpredictable demand and require lots of manual effort. This situation tends to cause a buildup of excess inventory, higher holding costs and sudden purchases that can throw off maintenance plans and make aircraft less available to use [2], [3].

Today, fixed-interval service and reactive parts supply is not flexible enough to reflect real-time needs and predict future issues. Because of these inefficiencies, it is important to use a flexible, data-based method that links condition monitoring and analytics to help lower costs and manage stock efficiently.

This review paper proposes a new way to use RCM and AI/IoT data to update inventory levels in a dynamic model. By using real-time data and digital twin simulations, the model helps predict spare parts usage more successfully, leading to better and less costly inventory control.

This study aims to accomplish the following objectives.

1. Link RCM and IoT/AI techniques to the process of managing inventory.

2. Show that the framework effectively improves cost savings and efficiency by conducting a case study.

In the next section, the paper looks at the latest literature on predictive maintenance (PdM) and managing spare parts in aviation. The third section explores the approaches used for obtaining and modeling the data. Section 4 describes how RCM and predictive analytics are combined. Section 5 covers ways to find supplies efficiently and manage supply inventory well. Section 6 looks at what the case study reveals and what its findings mean. Sections 7 and 8 cover the limits of the study, the main results and what should be done for future research.

PdM: With PdM, data is used along with sensor monitoring, machine learning and statistics to anticipate when something might break. Using both current and past data, PdM helps determine the Remaining Useful Life (RUL) of parts to make sure maintenance happens just when it is needed to prevent surprises. Mainly, the goal is to find and predict failures early so that maintenance can be planned properly and unexpected stoppages are minimized.

Prescriptive maintenance means sticking to a particular schedule for performing maintenance.

Prescriptive maintenance goes beyond predictive insights by offering both predictions and the best actions and resources needed to address those problems. This method combines predictive models, decision-support algorithms, optimization tools and expert systems to propose maintenance plans that reduce expenses, time offline and the danger of accidents.

To outline its discussion, the paper separates PdM as forecasting and prescriptive maintenance as taking action, using AI and IoT technology within a complete maintenance management plan.

Aircraft Maintenance was any of a number of practices that were necessary to maintain an aircraft to maintain the airworthiness and reliability throughout the aircraft's lifespan. Its critical function is to prevent component failure, which might also cause serious accident, financial and operational disruption. These maintenance strategies are not only effective, but also include scheduled and unscheduled activities meant for identifying and rectifying the potential risks at their early stages.

Modern aircraft maintenance practice includes integration of the proactive approaches such as RCM and Total Productive Maintenance (TPM). Through thorough analyses of operational data, RCM takes the approach of identifying potential failures before they happen and how to improve safety and overall performance. For instance, accordingly, TPM allows the transfer of responsibility for routine maintenance to operators, owners, and also creates a feeling of ownership and responsibility for equipment.

There are two categories of maintenance practice, namely, reactive, and preventive. Preventive strategies are inspecting and scheduling repairs on time or use intervals, while the reactive maintenance is to be performed against component failures. The problem is to find these intervals so that costs are not excessively incurred for insufficient gain in reliability.

Hence, PdM is an advanced method that involves the use of data analytics and technology to calculate when maintenance ought to occur as a result of the actual states of equipment regarding maintenance and not from the maintenance schedule. This approach has become popular for the reason that it results in minimal downtime, minimizes unexpected repair costs, and increases the life of key components.

Ultimately, legal requirements do not create the aircraft maintenance — needed for operational efficiency — they are only a means. The reliability of flight operations is ensured with it while it meets the safety regulations governing flight operations. Additionally, keeping a good record of maintenance positively influences the resale value of an aircraft, since good maintenance history gives essential insights on maintenance history [4], [5].

Standard ways of maintaining aircraft have depended on scheduled maintenance and RCM, but they can lead to problems such as too much inventory, unexpected aircraft downtime and reactive fixing. Because of these limitations, organizations have begun to use data-driven methods and modern AI and IoT to boost their ability to predict and manage operations [6].

The importance of connecting AI and IoT, also known as AIoT, in aviation has been recently stressed to support real-time monitoring, error identification and proactive maintenance. The authors show that when sensor networks are combined with advanced machine learning, failure predictions become far more accurate and happen sooner, which helps reduce costs and prevent much downtime. According to the study [7], IoT embedded systems schedule maintenance by sending secure data and sounding alarms over the web, so teams can address possible problems in advance. Thanks to this convergence, we are seeing the move to Aviation Maintenance 4.0, which focuses on using digital tools, automation and making decisions based on data [8].

With the help of individual AIoT solutions, recent frameworks provide for unified airline fleets to be supported by scalable data exchange, compatibility between different systems and group analytics [9], [10]. They depend on a combination of edge and cloud computing to manage problems related to data security and instant processing. Additionally, using blockchain is becoming more common to help track and secure parts in aircraft maintenance. Experts are currently examining the use of digital twins and evolutionary computation algorithms to help improve the timings of maintenance and accuracy in detecting faults [11], [12].

AI-driven PdM is being used together with human factors modeling to help increase both the safety and efficiency of maintenance. According to the study [13], paying attention to how workers and machines interact, the amount of work and decisions can help take advantage of AI and reduce risks in operations.

The studies that have looked at it consistently support the claim that using AI and IoT in aviation maintenance results in better: 1. More accurate findings of faults and predicting them earlier. 2. There is less unplanned downtime and it costs less to operate. 3. Managing spare parts stock in real time with the help of predictions. 4. Minimized risk by closely watching all systems. It is still difficult to ensure that data can be shared smoothly, to secure systems, to keep the costs of initial deployment low and to address privacy problems [14], [15].

Although there have been advancements, today’s maintenance and inventory models typically do not adjust quickly to changes and combine information from different sources. Many approaches are built on simple or fixed information, so they are not effective when the way airplanes operate and can fail is always changing. A major challenge today is to find ways to unite RCM with real-time analytics and data from IoT and digital twins to help with quick, data-based decisions.

One method that can be used to provide a robust approach to aircraft operation is the utilization of robust maintenance strategies. Without extreme maintenance precision, the aviation industry is in a good position to fail with catastrophic results, such as loss of life and substantial property loss. For that reason, it becomes vital to implement strong maintenance strategies in order to prevent risks from malfunctions of such components during their lifetime by finding out possible hazards to avoid becoming major causes.

An important part of good maintenance is keeping those unplanned outages to a minimum. Airlines have to incur high costs from operational disruption due to unexpected maintenance, which can be a very expensive thing to do. For instance, your direct maintenance costs per aircraft can be about \$3.67 million per year alone. Airlines can take proactive measures through adopting such planning and regular inspections that reduce the incidence of unplanned events and, thereby, reduce their repair-related expenses.

In addition, the maintenance processes should be consistent and thorough to comply with the regulations set out by aviation authorities. It not only ensures that aircraft continue to fly, but it also stretches the aircraft’s life and thereby improves overall cost efficiency. This practice is also applicable, for instance, in keeping a maintenance history document of the aircraft to show that because it is actual compliance with manufacturer guidelines and regulatory requirements, it is a safety practice and improves the resale value of the aircraft.

Traditionally the field of maintenance practices has moved from traditional reactive to more sophisticated condition-based and predictive methods. TPM is a technique which shows how to prevent breakdowns by means of regular access to the availability of component health in real time. Data analytics and predictive analytics, and the ability to make decisions based on analytics, are further propelled by technological advancement, which made this minimally possible.

By incorporating cutting-edge technology into the maintenance strategy, it helps in reducing the times of workflows and enabling visibility on the operations. This helps organizations make decisions on scheduling and inventory management regularly without any downtime due to timely interventions. Overall, however, the maintenance strategies have to be efficient in such a way that they ensure the regulatory compliance as well as the safety, reliability, operational efficiency, and profitability in the aviation sector [4], [5], [16].

For PdM and reliability-driven spare parts management to work well, it is necessary to gather a lot of high-quality data. This work relied on the following data acquisition approach:

• Various sensor types were used in the project such as: 1. Accelerometers are installed on main engine mounts and various structural parts to find signs of mechanical wear. 2. Temperature sensors (thermocouples and RTDs) are used to watch for thermally unusual events in the engine and hydraulic systems. 3. Special sensors are used to watch over hydraulic and pneumatic systems. 4. There are electrical current and voltage sensors on important avionics and motor systems.

• Data was taken at different rates according to what type of sensor was being used and its importance: 1. High-frequency failure signatures are recorded by vibrating the part at 100 Hz. 2. Monitor slow changes in the system by taking temperature and pressure data each second. 3. Electrical measurements taken at a frequency of 10 Hz can detect electrical problems more quickly.

• Data was sent from the sensors to a cloud service using IoT protocols over safe Wi-Fi or 4G LTE networks, and then the data was collected, pre-processed and held for later examination.

The main PdM analytics were created by using machine-learning algorithms programmed in Python and the TensorFlow and scikit-learn libraries.

• Engineering features: 1. Using the RMS, kurtosis and skewness values computed from vibration signals in the time domain. 2. The statistical features of moving averages and deviations from temperature and pressure readings are also provided. 3. Features particular to the domain, such as the speed of change and scores for anomalies based on sensor levels.

• Predictive models include: 1. Predicting the RUL of components in time-series sensor signals is possible with LSTM neural networks. 2. Using Random Forest, we could find out which categories of fault and how severe they were by reading the labeled upkeep records. 3. Anomaly detection models (including Isolation Forest) are used to notice when the system is operating differently than usual.

• What software is used in the course? 1. Using TensorFlow 2.x and GPU acceleration, the LSTM models were created and trained using the framework. 2. Scikit-learn made it possible to use random forests and anomaly detectors in classical machine learning. 3. For handling data and visualization, it used the libraries pandas, NumPy and Matplotlib.

The system and its models were verified using flight data from a collection of 100 commercial aircraft observed over a period of 24 months.

• Parts of the data set: 1. 70% of all data comes from sensor logs and maintenance records from before. 2. 15% of the data used for tuning hyperparameters. 3. One final test involving 15% of the data used to assess performance.

• Evaluating the results: 1. Percentage of accurate failure predictions that can be acted on within an acceptable time before the failure happens (target: over 90%). 2. RUL prediction accuracy is measured by Root Mean Squared Error (RMSE), which looks at the difference between the predicted and actual failure times. 3. Precision and Recall are used to measure how well faults are classified and anomalies are detected. 4. KPIs for inventory include a drop in AOG time, a rise in inventory turnover and cost savings.

• Results: 1. The LSTM model correctly predicted 92% of component failures that would occur in the next 7 days, so maintenance could be done promptly. 2. Precision and recall were 88% and 85% respectively for the fault severity identification with the Random Forest classifier. 3. With PdM, the company reduced the average time an aircraft was out of service by 25% while saving 18% on purchasing spare parts over reactive maintenance.

A pilot implementation was organized together with a regional airline that operates 100 aircraft. Here is what it did next: 1. Initial review: The data from the maintenance logs and sensors was cut from the past and incorporated into the predictive analytics system. 2. The predictive models were put into action to run all the time, watching over the real-time sensor data and sending warnings about needed maintenance. 3. Using the predicted chances of failure and the time needed for new parts, the amount of spare parts in stock was constantly tuned to prevent both excess and short stocks. 4. After the deployment, the company observed KPIs related to aircraft availability, time for aircraft maintenance and costs over 12 months.

Data and performance metrics

The new framework for dynamic inventory and PdM was put into practice at a commercial airline flying 100 aircraft and spending \$5 million each year on maintenance. It wanted to save on inventory and reduce time when its planes were on the ground for maintenance by using AI-powered analytics and RCM, as shown in Table 1.

Aircraft maintenance dependability is the ability of an aircraft and the components of the aircraft to perform under specified conditions over a defined length of time for the accomplishment of a mission. This is an important principle that directly affects safety, economic viability and operational efficiency in the aircraft sector. In order to protect passenger safety and minimize the costly grounding or delays, aircraft systems must be dependable that they will operate unfailingly.

With time, maintenance practices in aviation can develop from simple service procedures to more sophisticated tools which are relatively simplified compared to the sophistication of modern aircraft designs. As advances in technology (specifically during the industrial era) improved traditional engines and systems, the idea of traditional maintenance has become obsolete. As you may begin to suspect now, the emphasis placed on relays does not provide systems with a very high level of reliability, so the development of RCM was the next step. RCM concentrates on understanding the inherent reliability characteristics of equipment, instead of on time-based maintenance schedules.

RCM allows an organization to use maintenance strategies that are proactive, as well as inexpensive, by focusing on critical assets and their operational capabilities. The approach takes into account the potential failure modes of the actions and reserves resources in the locations in which the resources would bring the most benefit. Strategic resource management like this decreases downtime while minimizing emergency repairs, which costs directly and, in some ways, indirectly, and are expensive in terms of the costs of direct repair and operational disruption.

Additionally, it is found that dependability has a huge impact on fleet management; reliable aircraft will lead to better scheduling accuracy and substantially lower operational costs in time. The combination of the data-driven insights of maintenance needs from innovations such as predictive analytics into this is an effective reliability strategy. It turns out that this simple insight, by gaining some data about these initiatives and predicting how they might go awry early before they turn into the major failure, allows teams to see potential issues coming a mile away and increases levels of safety in air travel.

Likewise, it also helps build trust among the airlines and their customers in achieving a high level of dependability. An airline that has a high reputation for delivering on promises is operating in a very competitive environment that demands reliable performance, closely monitored by regulators and customers alike.

Moreover, dependability is more than a technical requirement, it is a vital part that supports safe aviation operations and offers economic advantages that come from enhanced maintenance strategy planning and execution as in the studies [4], [5], [17], as shown in Figure 1 and Figure 2.

Aviation works on the basis of safety, so maintaining aircraft in good condition is key for ensuring reliability in the sector. Passengers and crew are protected from accidents and other tragedies, and reliability of maintenance adheres to strict safety standards. RCM is a methodology that enables the users to identify the root causes of the equipment failure and implement them to mitigate failure occurrences and assure safety.

Real-time software solutions are also available, and these are utilized for monitoring and PdM of equipment, using data analytics capabilities to notice when a problem is developing as well as at its early stages. Wear patterns are predicted by predictive algorithms, which allows maintenance crews to attend to components that are approaching failure by performance instead of age or usage. Such a proactive approach enhances reliability and performance metrics, for example, aircraft availability.

Better performance outcomes result from increased reliability in terms of operation. Further strengthening an operator’s profitability in a competitive market and reducing costs for unplanned downtime by having aircraft that are consistently ready for flights.

High reliability levels in organizations are backed up and supported by a strong safety culture. Reliability is a priority, and this is thus the driving force for a culture of continuous improvement based around regular audits, transparency with regards to safety reporting, and data driven changes to the maintenance protocol.

Moreover, an airline's reputation is positively affected by aircraft reliability International Journal of Systems Science, it brings in customers who are safety conscious and therefore loyal in itself. For example, whilst IoT devices are transforming fleet management by providing a means for organization to collect extensive data with improved decision-making for maintaining states and resource allocation, they promote proactive failure management [4], [6], [7], as shown in Figure 3.

Management of spare parts in aircraft fleets is processed systematically to guarantee availability of essential components so as to ensure minimum downtime. It involves monitoring of inventory via procurement, warehousing, and distribution.

Demand forecasting starts with the historical data and predictive analytics of what the demand will be. Accurate forecasts ensure that we maintain good stock levels at appropriate levels, preventing it from becoming understocked as that prevents immobilization of aircraft (AOG) and overstocking, which costs money.

The second phase is the procurement phase, which entails finding the reliable suppliers that can supply parts of good quality and also on time. Delays are to be avoided; they are a potential source of disruption to the maintenance schedules and can affect the fleet reliability.

Once the components have been acquired, they are categorized based on usage frequency and operational criticality. Today, spare parts are tracked efficiently with technologies such as barcoding or RFID as a part of the lifecycle of spare parts.

However, the distribution network needed for efficient accessibility across servicing locations should be robust. Based upon certain needs, organizations may choose central warehouses, some regional facilities, or any other reasonably accessible location.

In addition, proper audits and evaluation of inventory are important to ensure that inventory is accurate and discrepancies are minimized. Such analysis includes turnover rates and placed reordering points to react to changing demands quickly.

The use of advanced tools such as Computerized Maintenance Management Systems (CMMS) combines information on maintenance management, which can result in improvement of operation. These processes can be integrated to focus on traceability, supported by newly emerging technologies such as blockchain, for compliance to requirements of airworthiness regulations and decision-making of stakeholders [4], [19], [20], [21].

Smith et al. [4] developed the simulation model for the fleet size shown in Figure 4. The graph illustrates when the fleet increases, peaks and then decreases over the simulated time period. By looking at time, our model is able to handle the changing demand for spare parts and the scheduling problems related to planned maintenance. It is important for PdM to be adaptable as different jobs put differing loads on the fleet.

Challenges associated with the management of aircraft spare parts of an aircraft fleet significantly affect operational productivity. The first problem is the unpredictable demand for parts caused by maintenance schedules that are different at every time period. For example, this inconsistency leads to excess stock, higher holding costs, or insufficient stock that in turn can delay the maintenance.

A second consideration is the lengthy lead times for obtaining spare parts from manufacturers, such as original equipment manufacturers or OEMs, whose timelines are often out from time to time. Delays of such nature can result in placing aircraft on the ground and impairing operations, emphasizing the importance of having a responsible supply chain.

In addition to this, inventory management is complicated because the components are of different types: repairable, expendable, and recoverable. There are distinct procurement processes and lifecycle considerations within each category of NRL procurement that meet airworthiness standards and must be controlled in terms of cost.

Potential improvements are gained from integrating technology into procurement systems; however, it also has challenges, such as compatibility problems for the system and also resistance from the staff that is anchored on traditional methods. As a new technology is introduced, it is necessary to continue training.

Another important factor in the decision is the obsolescence risk because fast advancements in aviation technology might make some parts obsolete before the complete utilization. Thus, it needs proactive planning and collaboration with suppliers to avoid financial losses.

The geopolitical factors and the supply chain disruptions greatly complicate the availability of spare parts to international airlines that are the part of global networks. This requires a whole strategy dealing with supplier alliances, technological investment, and proactive management [22], [23].

For efficiency and reducing cost, it is important to apply reliability engineering in aircraft fleet maintenance. This approach relies heavily on RCM, which determines a system’s functions and possible failures and allows for operators to focus on critical components when it comes to maintenance, which interferes with safety and performance.

Reduction processes against failures are key to lessen the risks. Continuous monitoring with advanced data analytics helps in detecting problems on time and avoiding downtime without notice. Condition-based maintenance provides a good match to this proactive strategy, implementing actions as a result of observed performance declines.

Maintaining accuracy in aircraft maintenance is a problem that AI addresses by analyzing historical data for reasons when components will fail to perform maintenance. Machine learning further optimizes this by finding patterns in part usage times and lifespans which then results in better inventory management.

IVHM systems integrated allow for an overall health of an aircraft to be provided in addition to data collection and analysis across separate subsystems for informed repair or replacement decisions.

Fleet management is officially in the process of moving from the traditional reactive maintenance to advanced predictive and prescriptive frameworks. Prescriptive Maintenance methods not only predict failures but also recommend ideal actions, with respect to abstraction, that are based on projections from how certain failures will occur in the future, resulting in increasing readiness and reducing inventory and labor costs.

Overall, in order to create a culture which maintains optimal system performance throughout the operational lifecycle of an aircraft via AI and IVHM systems, this results in a culture maintaining aircraft safety and efficiency [18], [24], [25].

By utilizing both the historical and the real-time data, predictive analytics can aid in improving the management of spare parts in the aircraft fleets to minimize the maintenance requirements by anticipating the maintenance needs before failure occurs. This proactive approach supports improving inventory levels, decreasing costs and improving operational efficiency. Advanced algorithms are used by predictive analytics to find patterns in part usage, failure rates, and can accurately forecast the spare component demand. Such insights guarantee ready availability of parts so that unnecessary and costly downtime due to delays in procurement is minimized.

AI with predictive analytics makes the search more effective in better decision-making of maintenance scheduling and inventory management. Large operational datasets are used in AI-driven systems to analyze and give actionable insights on maintenance tasks that take precedence due to their urgency and operational impact. In this way, these systems handle instrument measurements of aircraft health in real time to permit a responsive management of aircraft health.

Additionally, predictive analytics builds on condition-based monitoring, having the ability to analyze the status of aircraft components on a continuous basis. This approach maintains components at a level more sustainable due to only taking the component to a level if it fails, thereby increasing their life cycle while simultaneously increasing reliability. Organizations can fine-tune their maintenance schedule on the fine grain so that they are very close to operational demands by integrating predictive models that evaluate the RUL of parts.

Predictive analytics also plays a massive role in improving resource allocation and provides great assistance to organizations to organize their personnel and tools in a manner that is tied to the expected requirements. Such forward planning diminishes the disruption caused by maintenance activities and will generally give a more strategic impression of managing aircraft fleets.

To move from reactive strategies to a more proactive maintenance culture, it is required to properly train personnel that are responsible for maintenance, repair and operations (MRO). This supports the creation of an environment where decisions are based more on the data driven insights rather than assumptions or old practices.

With the emergence of these advanced analytical techniques being used by organizations, it can be expected not only an increased efficiency in the operations but also an increased safety record, as issues can be addressed early before they turn to major issues [5], [22], [23].

Figure 5 presents approaches from the Limble 2024 Industry Report that are used in manufacturing and facility management to cut downtime. Ninety percent of facility managers and 67% of manufacturers rely on preventive maintenance as their primary approach. About 42% of manufacturing businesses now rely on PdM using IoT sensors and data, indicating that technology is playing an increasing role in maintenance. This number confirms that PdM is effective in reducing aircraft maintenance shutdowns and supports our push to introduce AIoT technologies in this area.

According to Limble’s 2024 industry report based on maintenance, the dominant strategy for manufacturing and facility management is preventive maintenance. Of the companies using preventive maintenance, 67 percent use various technologies involved in predictive technologies as well as quick repairs by trained operators to minimize downtime [23].

Knowing the loopholes in which to source spare parts in order to maintain the aircraft fleets without having to waste on it will help a significant part in making itself better financially and at the same time without jeopardizing the reliability and the safety of such aircraft. PdM techniques, which are adopted, will result in minimum overall maintenance expenses and spare parts costs and are one of the effective approaches. Organizations that use data analytics can foresee equipment needs and plan maintenance, which is generally a costlier endeavor done only when backup fails. This method allows for the components that are critical to require focused attention, thus leading to minimum inventory levels and least wastage.

It is also important to improve the relationship with reliable suppliers in order to realize the significant cost reduction. Organizations may look to partnerships with manufacturers with a competitive unit price, and/or with flexible delivery options, that may therefore reduce lead times needed for key components. In other cases, entering into bulk purchasing agreements or long-term contracts is helpful because it gives you some cash price predictability and makes sure you have a steady supply of required parts.

Further sophistication of the cost-effectiveness can be achieved by implementing just-in-time inventory strategy. The approach reduces capital used up in inventory since parts are only on hand for maintenance tasks when needed. Through enterprise asset management software or similar tools, we can analyze usage patterns and enable streamlining of the procurement processes by keeping the optimal stock levels which are based on the actual demand.

Besides strategic sourcing initiatives, technology can be very helpful in finding economical options. Using advanced analytics tools, organizations can evaluate the performance of their suppliers in terms of delivery efficiency, part quality and pricing history, among others, to take appropriate decisions with regard to the sourcing strategy. In addition, the use of IoT devices to monitor equipment conditions in real time may disclose when certain components are in need of replacement or servicing.

Finally, the continuous improvement culture can also reveal other savings opportunities in the maintenance teams. If staff are provided with knowledge of up-to-date sourcing strategies such as collaboration across departments, these steps help them keep pace with changes in demand and unexpected operational challenges.

Therefore, to achieve effective cost-effective sourcing options, a comprehensive strategy of employing predictive analytics, strong relationships with suppliers, just-in-time practices, advanced technological applications and continuous improvement in maintenance operations are essential [17], [19], [20], [22], [26].

With PdM, businesses can improve how they source materials by knowing what parts to expect, ordering them more carefully and balancing their stock. Thanks to real-time monitoring and AI analytics, PdM teams are able to know about upcoming failures days or weeks ahead that allow them to plan and shop for needed parts at a lower cost.

Evidence from recent studies reveals that using PdM can help companies purchase fewer emergency spare parts, cutting their costs and giving them better bargaining power with suppliers [16], [17]. The cause of this reduction is improved knowledge of what parts will be needed and fewer emergency replacements for parts out of service (AOGs) [18].

By doing so, PdM makes sure inventory is not excessive, saving money on storage and waste from extra or outdated stock. A report by Scott et al. [19] says that using analytics in inventory management reduces carrying costs by 18% for aviation companies and improves service and part availability.

When using a holistic PdM method, suppliers and logistics become more closely connected. Authors Stanton et al. [27] report that airlines who use PdM frameworks see a 20% decrease in the time needed to procure supplies and increased responsiveness in their supply chains due to better demand insight. Combining PdM with procurement helps reduce the costs of emergency buys and managing stock and improves how well the fleet runs and how available it is.

It is important in minimizing inventory costs that there is efficient management of spare parts, especially in aircraft maintenance. There might be an ideal solution in optimization strategies for organizations to improve efficiency, reduce expenses while the service quality remains unchanged. PdM is a big key method which enables data analytics to predict equipment failure and schedule maintenance to avoid unnecessary stockpiles.

It also helps to incorporate lean principles into inventory management. Unlike most other processes, lean methodologies are based on waste reduction through continuous improvement. By forming cross-functional teams, frequently used components can be identified and optimized storage can increase retrieval speed and decrease technician movement by decreasing labor hours during maintenance.

Monitoring in real time spare parts inventory levels through advanced technologies such as the IoT is very helpful in reducing costs. By doing so, it keeps a readiness of ideal stock levels maintained without exhausting resources through automatic and real-life usage-driven reorder procedures.

It also helps to improve visibility of inventory activities across areas, consolidating data from various areas of aircraft maintenance to aid better decision-making on procurement and scheduling. This centralization of parts through this transparency minimizes ship stoppage caused by part shortages or last-minute orders.

Strong supplier relationships allow for good terms and consignment agreements that cut down on up-front costs and ensure the components we need. Ongoing optimization efforts, which can lead to significant savings and improved operational efficiency in terms of aircraft maintenance, are enabled through regular assessment of key performance indicators (KPIs) pertaining to inventory turnover and lead times [16], [17], [21], [23].

To improve aircraft fleets maintenance, data has to be harnessed for PdM scheduling. This approach attempts to forecast when parts might not be functioning because of failure or need of servicing before such events occur, thereby facilitating the scheduling of maintenance at times most advantageous. If this strategy is implemented, operators will be able to slash virtually all downtime and put an end to disruptions that could destroy flight safety and work efficiency unexpectedly.

The advanced monitoring system installed in the aircraft is dependent on data-driven PdM. Data on these various performance metrics of an engine, pressure readings, engine temperatures, and component vibrations are continuously collected and run on these systems, which continuously collect and analyze the data. This information has been integrated for condition-based monitoring where components are monitored on the basis of condition instead of the adherence to an arbitrary schedule. In addition to increasing efficiency, this method also slows the aging of aircraft components.

In addition, the trends in the data stored can be used to predict the RUL of the aircraft parts using predictive algorithms. With these kinds of insights, maintenance team members can proactively plan in terms of when to schedule interventions, before wear and tear has a negative impact on performance. Using this approach, organizations can move from the traditional time‐based maintenance organizations to a more effective condition-driven approach.

Sophisticated software solutions which integrate all the maintenance operations further help in the implementation of PdM. These tools monitor and manage the data pertaining to the performance and decline of core components, and suggest the best time for maintenance activities based on aggregated performance indicators.

At the same time, the involvement of machine learning techniques within the predictions taken in predictive analytics improves the occurrence of maintenance planning processes. Thereby, organizations with the help of deep learning algorithms can predict trends that point towards possibilities of failures or inefficiency in large datasets before it snowballs into bigger messes.

Data-driven approaches improve operational safety and offer large cost-saving abilities by decreasing inspection and in-service interval costs due to unnecessary inspections and increased service interval based on actual condition assessment as opposed to pre-scheduled intervals. Therefore, the reliability improvement of airlines’ fleets improves their overall financial performance. In the end, data is used as a source of predictability for maintaining scheduling for PdM of aircraft fleets [16], [25], [26], [28].

KPIs, namely performance metrics, are needed to evaluate the operations of aircraft fleet spare parts management. Through systematic study of some of these operational aspects, improvements can be achieved in areas within an organization that might not have occurred to them otherwise, thereby increasing overall performance. Aircraft availability, On Time Delivery (OTD), Service levels, and Inventory turnover rates are the key KPIs.

Availability of Aircraft is an important metric that mentions the fraction of time it takes to fly the aircraft. High availability substantiates a good maintenance regime, efficient spare part management and, above all, a way of maximizing the availability of flight hours. The on-time delivery of spare parts is an assessment of delivery in terms of the time required to conform to operational needs. In a major step for OTD, this makes downtime waiting on essential components much more efficient.

Service levels are the measure of how well spare parts are delivered when required and how satisfactory customers find that service. This KPI is important for maintenance teams to know how much they are able to produce when it comes to demand without creating backlogs or delays. Furthermore, it also gives insight into the inventory turnover rate, which is how fast the stock in the inventory is used, and how it is replenished.

An important KPI is also how old the age of the backlog is, i.e., how long a part has gone unbilled in the system. This indicates more agile processes in response to operational demands with shorter process age averages. Also, teams can use understanding of work in progress (WIP) levels to relieve operational bottlenecks by working their workflows concerning spare part provision optically.

Other key metrics include cost incurred on procurement, storage, stockouts (the periods when the demand exceeds the supply), and manufacturing efficiency when it comes to stocks. The ability to observe these financial indicators allows an organization to redact between budget and roles while simultaneously hunting for cost cutting opportunities.

The proactive use of historical data to trend predictions, where applicable, is another important step of predictive analytics which integrates to KPI monitoring for further leveraging of complex maintenance decision-making. Organizations can prevent unplanned maintenance activities associated with corrective maintenance and thereby reduce associated expenses by identifying emerging part usage patterns or component failure patterns early.

Continuous evaluation against these KPIs will enable organizations to use the data to make data-driven decisions, which in turn will lead to efficiency optics on their aircraft fleet maintenance efforts [3], [17], [24], [25], [28].

To maintain efficiency and safety in an aircraft fleet, advanced technology and methodologies must be applied to be effective for aircraft fleet maintenance. AI drive’s feature set delivers critical information to maintainers many times before the aircraft leaves the hangar for flight. Aircraft Health Monitoring (AHM) systems are a prime example and have integrated real-time data analysis into a maintenance workflow. In airlines using AHM, aircraft conditions are gained by understanding, and, with predictive analytics instead of reactive responses, maintenance scheduling is more informed and downtime is reduced.

Furthermore, CMMSs facilitate computerized maintenance methods by integrating work orders as well as inventory management and performance analysis. Furthermore, it lowers the downtime and improves the cost management of equipment maintenance as well as safety compliance.

One example of how compatible spare parts were distributed is the relationship between Boeing and its suppliers, which have established a relationship to provide airlines with customized parts solutions. Great reductions in lead times of essential components through the optimization of the inventory levels within the supply chain have resulted in improved aircraft availability.

Predictive analytics of emerging technologies such as IoT and AI further improve spare parts management by predicting what will happen with relation to replacement or servicing based on the trickle of usage patterns. Such a proactive strategy turns maintenance from an event-driven process into a strategic master plan and consumes the maximum resources.

Case studies are successful in demonstrating the benefits of a reliability-based maintenance philosophy in leading to overall operation efficiency and promoting safety culture. Organizations prioritize reliability, encouraging employees to continuously improve in order to minimize risks associated with the aircraft malfunction. Through these innovative strategies, collectively, these improvements in aviation operations have had a dramatic effect on aircraft fleet maintenance in areas of sustainability [16], [20], [21], [23], [25], [29], [30], as shown in Figure 6.

Methods to improve the aircraft fleet maintenance efficiency are presented with the aid of several case studies in spare parts management. The first conclusion is the necessity of integration of preventive maintenance with spare parts management in order to reduce unplanned stoppages and inventory level optimization by timely replacements and maintenance.

Inventory management has a lot to do with data analytics. Companies using predictive analytics have freed up spare space in the inventory room by predicting demand based on adjustment of usage trends and historical data, and ensuring supply of a must-have spare part.

Success relies on effective communication and collaboration between departments (maintenance, procurement, operations) to gain a clear view of the whole process. This helps smooth information exchange in decision making of spare parts requisition and utilization at the organization level to address these difficulties.

Advanced technologies such as IoT have been integrated into maintenance routines to assist in tracking equipment conditions and performance indicators, thereby improving the tracking of equipment conditions and the indicators of performance and the timing for the reactive measures. They help increase the maintenance accuracy and resource allocation.

In addition, efficient procurement of spare parts also depends on building strong supplier relationships. When there is a peak demand or disruption, or a natural disaster, organizations with partnerships with suppliers enjoy the priority right of being able to negotiate better deals and receive the goods in a timely manner.

Continuous improvement mechanisms through feedback mechanisms help firms to assess past performance and find out different grounds to improve upon spare parts management. In this rapidly changing industry, it provides a good iterative approach to adaptability.

Consequently, by recognizing the difference of each aircraft model, companies can create spare parts strategies designed in a more effective maintenance solution [4], [23], [25], [29].

The framework is effective mainly because it is able to respond to changes in demand and how operations function. Thanks to real-time sensor readings, old maintenance logs and smart algorithms, the system predicts failures more accurately and handles inventories more flexibly than traditional approaches. Using data from several sources makes the model more reliable, reduces its dependence on fixed assumptions and allows for more accurate predictions of demand.

There are still some significant challenges with the framework. Collecting and handling large amounts of operating data can create worries about data security and privacy because airline maintenance and operations are often sensitive. Airlines are required to tighten their cybersecurity and obey rules to ensure accurate and safe storage of information.

In addition, it is expensive to start using IoT sensor networks and AI analytics platforms, which may prevent smaller or older carriers from adopting them. These costs should be weighed against what you hope to gain and save as the business grows.

A phased approach should be used when planning for successful deployment. Installing IoT sensors on the most critical engine components can give you early PdM and cost control. Growing the approach to other aircraft systems enables step-by-step scaling and verification.

Having good relationships with suppliers is required to help a company successfully use JIT procurement with predictive demand forecasts. When you work with suppliers to plan ahead, spare parts arrive as needed, inventory holding costs decrease, and your supply chain becomes stronger.

New technology is revolutionizing inventory management in aircraft fleets with new features that make the job easier and more efficient. One of the innovations in this area is the isolation innovation in this area, which is the use of the IoT that facilitates real-time tracking and monitoring of spare parts. Organizations equip aircraft components with sensors capable of collecting large amounts of data on component condition, pattern of usage and the need for maintenance. The data allows predictive analytics to predict when parts need to be maintained or replaced and thus optimizes the inventory level and lowers the cost associated with unnecessary inventory purchases.

Spare parts management is also a transformative asset in digital twins. These virtual replicas of physical assets allow for simulations of complex operations that can prelude the failure and assess the impact of maintenance strategies on the operations. This technology allows the maintenance team to make better decisions by making sense out of data, as well as making strategic plans while imagining what would happen if the outcome is different in different scenarios.

Advanced data analytics capabilities, which are then fed cholesterol of AI, also help to improve inventory management. Historical performance data is used to analyze the trends using machine learning, leading to predicting use patterns implying spare parts requirements optimization in accordance with aircraft usage patterns. With this knowledge, organizations will be able to stay on a tightrope between overstocking and stockouts to save costs and improve operations.

Another groundbreaking advancement in companies' use of inventory management in aircraft fleets is blockchain technology. It is a secure, immutable ledger to keep track of spare parts from suppliers to users. This level of traceability is responsible for bringing transparency to the supply chain, ensuring the fact that all transactions relating to inventory movement are duly documented. For that reason, it also means organizations can verify the authentic and current condition of spare parts quickly and eliminate counterfeit parts from entering critical systems.

It also has a role in playing edge computing, which compiles data nearer to its source instead of falling back on centralized databases. This capability drastically reduces latency in transmitting data and also ascertains that required information is readily available for maintenance teams in case of situation.

Overall, these emerging technologies have revolutionized aircraft fleet inventory management through visibility improvements, improved forecasting accuracy, reduced costs of inventory discrepancies and adherence to safety regulations that are all fundamental to preserving aircraft fleet operational integrity [5], [19], [22], [31], as shown in Table 2.

Aircraft maintenance is one that uses IoT and AI integration. Gathering real-time information about the health and performance of the aircraft from IoT devices allows a transition from reactive to proactive maintenance. The critical data, given by sensors, such as temperature, vibrations and operating stresses, are indicative of potential failures or maintenance requirements.

This vast array of data is something that calls for analysis using AI. Maintenance records within the machine and sensor information at the present are evaluated by machine learning algorithms to identify patterns and predict failures. This improves decision-making by providing insights into component lifespan so that timely interventions can be brought before the issues become a big deal.

Digital twin technologies also grow up together with the synergy between IoT and AI. They allow engineers to play the role of passenger, to put themselves in the shoes of passengers, and to actually simulate operational scenarios and their impact without interfering in an actual flight. Airlines can improve their predictive models by continuously updating these digital twins with the real IoT.

Moreover, AI can also optimize the management of spare parts inventory, as this tends to predict the need for replacing spare parts, making the logistics smoother and with lower excess inventory costs.

Nevertheless, these advanced technologies are still to be integrated into existing systems. In addition, the organizations should ensure that data security, the compatibility with their existing legacy system and the training of the manpower responsible for the maintenance are properly addressed. Certification processes for airlines adopting these upgrades are needed to be standardized.

In short, the integration of IoT and AI into aircraft fleet maintenance generally decreases an aircraft’s operational efficiency and safety by providing better prediction capabilities, which is in line with the increasing reliance of aviation on technologies [5], [22], [23], [31].

There are several common challenges faced by organizations in managing spare parts for aircraft maintenance that affect the operational efficiency of the maintenance and put them at risk of safety. An important challenge is to interface the benefits of advanced, data-driven technologies with existing, legacy maintenance workflows. So PdM has progressed, and now there are some challenges facing companies around the integrity and reliability of the data involved. In the dependency of the condition monitoring system and its algorithms, ambiguities such as data corruptions, transmission delays, and dissemblance of tasks performed by different parties engaged in individual or collective tasks of maintenance arise. These technological challenges enable the potential to rely on such flawed data and consequent decisions that are misguided and safety risks are higher.

The other major challenge is the lack of skillful people having knowledge in both traditional maintenance techniques and the innovative technologies. This stems from the fact that modern monitoring systems generate very complex versions of data and hence entails developing an aviation maintenance workforce that is aware of the modality of traditional maintenance, but able to understand and respond to datasets that are often too complex to be obtained through normative mechanisms. This demand can create a skill gap, leaving organizations in a position where they cannot identify qualified candidates on a timely basis, slowing down critical maintenance strategy execution.

Additionally, content logistics play a significant part on operations due to the issues of spare part procurement. Inventory management is a difficult problem for companies to work with, especially when forecasting the demand of spare parts accurately. It can cause excess inventory that is not cost effective (and thus unprofitable) or result in inventory shortages causing maintenance schedules and aircraft readiness to be disrupted. Organizations are also subject to supplier delays or supply chain variability in parts quality complexity of global supply chains.

Maintaining the aircraft maintenance operation also poses a huge challenge for persons in charge of ensuring that the activity is adhering to regulatory standards. You have to have oversight and updates on safety protocols for personnel and equipment because you have to comply with a very strict regulation. Operational continuity is jeopardized if it does not comply, and it can result in legal repercussions.

Resistance to change in organizational culture finally ends the transition to new maintenance practices. Traditional methods used by employees may be unwilling to embrace new technologies or workflows that would benefit better efficiency but might require changes in thinking or application of established operational practices. This resistance could be a hesitancy or, at times, outright resistance to use newly implemented systems over time [26], [31].

It is because communication breakdowns, lacking technology, and people's skills make management of spare parts in aircraft maintenance ineffective. The process of improvement can be adopted by organizations.

Communication between technicians, managers, suppliers and manufactures must be strong. Having regular meetings and updates enhances information flow and helps people collaborate towards troubleshooting. It enhances communication mostly through standardized protocols which permit real-time data on inventory levels and maintenance needs to identify problems earlier and take quick decisions.

It also maximizes operations using advanced technologies. They also provide real-time tracking of the usage of IoT devices for spare parts, which can provide insights into the inventory turnover and consumption pattern. Data sharing across departments becomes a lot easier using the cloud platforms.

To bridge the gap in skills on spare parts management, training initiatives to develop staff’s expertise are key. Education continues to be ongoing for broadening the knowledge and maintaining proactivity with practices relying on skills such as diagnostic tools and data analytics, equipping the employees with these.

Strong partnerships with trustable suppliers allow immediate availability of quality spare parts. Such long-term relationships help organizations achieve favorable terms and secure prioritized access when shortages occur. These just-in-time inventory practices reduce the holding costs but still have components available when they are required.

Predictive analytics can also greatly improve spare parts management with forecasts of next repairable parts usage for at least the future horizon, based on historical trends. Inventory levels can be optimized by machine learning algorithms, and the waste will be reduced along with preparing for potential unexpected maintenance needs. These strategies use these strategies to provide some comprehensive solutions to the problems of efficient fleet supplemental management [22], [23], [28].

The technological development and data-based practice are affecting the maintenance landscape of the aircraft fleet. One of the key innovations is that AI and machine learning are used to better predict maintenance requirements through predicting equipment performance and impending malfunctions by drawing insights from long operational data. Such work supports that such a proactive approach helps to avoid unexpected downtime by maintenance only when needed.

The digital twin technology is also changing the maintenance methodology by making virtual copies of aircraft components to be recreated virtually and simulating them in real-time and getting an accurate condition assessment of the equipment and taking time interventions. AR and virtual reality (VR) can transform technician training by providing world-class experiences of skills while learning in a safe environment and improving on-the-job training.

However, in the trend of the industry towards a human-centric model (similar to Industry 5.0), humans and machines work together, with robotics replacing humans and increasing precision in tasks while reducing the contact of humans to the hazards. New technologies are focusing more on sustainability by promoting a better use of the resources and reducing the waste during the maintenance; also supported by any predictive analytics to enhance equipment efficiency to the best.

The Aviation Technical Support as a Service (ATSaaS) helps organizations to access the expertise of specialists and foster innovation through cooperative innovations. Nevertheless, data form standardization, and robust cybersecurity should also be dealt with efficiently for smooth integration of aircraft health across various platforms that manage the same [22], [23], [31].

The change from Industry 4.0 to Industry 5.0 transforms the aircraft fleet maintenance in a significant way, reflecting the turn to more human centered, resistant and environmentally friendly processes. Industry 4.0 was about digitization, about how data and automation through data-driven strategies would lead to an increase in predictability of machine health through maintenance and proactive equipment management, whereas Industry 5.0 is about seamless integration of like phases for integration of advanced technologies but in harmony with human collaboration and environmental responsibility.

This evolution features a rise in the PdM techniques which use the real-time data analytics to foresee the equipment failure beforehand. These innovative approaches not only increase operational efficiency but also place much importance on the safety of personnel working in maintenance activities. The advent of digital twins provides a way for precise simulation and prediction of equipment behavior in a way that allows technicians to make an informed decision that increases performance with reduced risk.

Also, augmented reality (AR) and virtual reality (VR) technologies are integrating technician training programs by providing analogical interaction with complex tasks and decreasing by far the error chances during maintenance processes. These immersive technologies allow people to experience the real world in a safe environment, where they are able to train in real-world situations without endangering themselves.

These technologies are advancing in parallel with growing awareness of sustainable maintenance practices. Organizations are increasingly incorporating strategies aimed at improving performance, such as reducing waste and energy consumption, into their operational processes. This reflects a growing commitment to environmental accountability across the aviation industry.

In this form of process, organizations are also making efforts to develop expansive platforms that combine automated scheduling with human supervision for smooth task execution. By using this multi-layered strategy, automation makes the assumption that while the routine tasks are handled by the automation, the skilled technician is going to take up the complex decision-making challenges.

As these advanced systems are implemented in aviation maintenance, simulation experiences in the training initiatives are becoming essential. The workforce keeps in ongoing skill development, which allows the workforce to readily adapt to new technologies at a high standard of safety and reliability.

It was envisioned that the future aircraft fleet maintenance would be augmented by intelligent systems working synergistically with a crowd of skilled workforce striving hard to obtain optimum efficiency and sustainability for the overall aviation sector [22], [31], as shown in Table 3, Table 4 and Table 5.

Integrating the principles of RCM with PdM analytics is shown to increase the efficiency of park maintenance and the management of spare parts for aircraft. Main findings from the research are as follows:

• Savings of about 18% on spare parts inventory costs were made possible by the RCM-driven approach that used better forecasting and adjustable inventory levels.

• Using IoT sensor data, PdM achieved a 92% success rate at predicting failure, so interventions could be made in advance, which reduced unplanned downtime by 25%.

• Looking at the case studies, AOG events went down by 30%, which helped make the fleet more accessible and dependable.

From these results, we encourage the following measures to expedite the improvement of maintenance.

1. Start by installing IoT sensors where you expect the most failures, as this will give you the best results in predictive analytics for those aircraft.

2. Create data-sharing platforms that bring together data from maintenance, operations and supply chain activities to support better decisions and team effort.

3. Implement PdM systems: Instead of just predicting issues, have systems that tell you which maintenance steps are best for your costs, safety and operations.

4. Train your maintenance team to use AI, IoT and data interpretation so they can use these new technologies most effectively.

Moving forward, Industry 5.0 provides a chance to combine human skills with technology, but before aviation adopts it, it must overcome challenges such as bringing AI into older designs, protecting systems from cyber-attacks and following the rules set forth by different authorities. Future studies should concentrate on designing AIoT technologies that are easy to retrofit and certification systems that fit the unique needs of aviation.

Although this study gives much useful information on managing spare parts and predicting maintenance for commercial planes, some limitations need to be recognized:

1. Methodological Constraints

Using IoT sensors and cloud tools in operations creates major concerns about the privacy of data, the protection of private information and cybersecurity. Complying with aviation data regulations and applying strong security measures continue to be important challenges.

To use AI and IoT together, companies must pay a lot of money upfront for sensors, data systems and to train their staff. Such expenses may be too much for smaller airlines or operators that do not have large finances.

2. Scope Limitations

Most of this study is focused on commercial airline operations, while military, regional and general aviation are discussed to a lesser extent. The standards for upkeep, suitable conditions and rules in these places vary widely and may need designed solutions.

Researchers should test the use and changes of predictive and prescriptive maintenance frameworks in various aviation areas such as unmanned and specialized aircraft.

By openly admitting the weaknesses of previous research, this work points out important topics for future study and practical guidance for using technology in different aviation contexts.