Ecosystem of Aviation Maintenance: Transition from Aircraft Health Monitoring to Health Management Based on IoT and AI Synergy

Abstract

This paper presents an in-depth exploration of the transformative impact of integrating the Internet of Things (IoT), cloud computing, and artificial intelligence (AI) within the domain of aviation maintenance. It articulates the transition from conventional health monitoring practices to a more advanced, comprehensive health management approach, leveraging these modern technologies. This paper emphasizes the pivotal shift from reactive maintenance strategies to proactive and predictive maintenance paradigms, facilitated by the real-time data collection capabilities of IoT devices and the analytical prowess of AI. This transition not only enhances the safety and reliability of flight operations but also optimizes maintenance procedures, thereby reducing operational costs and improving efficiency. This paper meticulously outlines the implementation challenges, including technological integration, regulatory compliance, and security concerns, while proposing a future research agenda to address these issues and further harness the potential of these technologies in revolutionizing aviation maintenance.

1. Introduction

In the realm of aviation, ensuring the highest levels of safety and operational efficiency is paramount. The integration of the Internet of Things (IoT) and artificial intelligence (AI) into aircraft health monitoring systems represents a significant leap forward in achieving these goals.

The IoT’s contribution to aviation primarily revolves around its ability to facilitate real-time data collection from a multitude of sensors embedded across aircraft systems and components. These sensors continuously gather critical data points, such as engine performance metrics, structural integrity indicators, and systems’ operational status, providing a comprehensive overview of an aircraft’s health in real time. This wealth of data is indispensable for identifying potential issues before they escalate into serious problems, allowing for timely interventions and thereby enhancing flight safety and aircraft reliability.

While the IoT provides the raw data necessary for monitoring aircraft health, AI is the powerhouse that analyzes this data to extract meaningful insights and actionable intelligence. Through machine learning algorithms and advanced analytics, AI can identify patterns and anomalies that may indicate potential failures or areas of concern. This predictive capability is at the heart of modern predictive maintenance strategies, which focus on performing maintenance activities based on the actual condition of the aircraft, rather than on predetermined schedules. By predicting potential issues before they manifest, AI-driven health monitoring systems significantly reduce the risk of unexpected failures, thereby enhancing the safety and reliability of flights.

The synergy between the IoT and AI in aircraft health monitoring facilitates a proactive approach to maintenance, which is instrumental in enhancing flight safety. By identifying potential issues early and enabling maintenance actions to be taken before problems arise, these technologies ensure that aircraft are in optimal condition for safe operation. Furthermore, the ability to predict and prevent failures reduces the likelihood of in-flight malfunctions, significantly contributing to the overall safety of air travel.

The aviation industry’s transition from traditional health monitoring (HM) systems to more comprehensive health management (HMGT) approaches represents a pivotal shift towards predictive and proactive maintenance strategies [1]. This evolution is not just a matter of semantic difference but signifies a deeper transformation in how aircraft health is approached, leveraging cutting-edge technologies like the IoT and AI. The current shift towards health management extends beyond monitoring, aiming for a holistic view of an aircraft’s operational readiness and lifecycle management. It encompasses predictive maintenance, condition-based monitoring, and even extends into logistics and supply chain management for parts and repairs. This approach demands not just data collection but deep, insightful analysis and strategic planning, areas where the IoT and AI are poised to make a significant impact.

One of the primary challenges for researchers is fully understanding the capabilities and limitations of the IoT and AI within the health management paradigm. IoT devices offer unprecedented data collection opportunities, but the sheer volume and variety of data can overwhelm traditional processing and analysis methods. Similarly, while AI has the potential to derive meaningful insights from these data, the complexity and unpredictability of aircraft systems and operations introduce significant challenges in model accuracy and reliability.

The transition to health management also brings new problems to the forefront, such as predicting component lifespan, optimizing repair and maintenance logistics, and even anticipating regulatory and environmental changes. Researchers must explore how the IoT and AI can contribute to these areas, requiring a multidisciplinary approach that spans engineering, data science, logistics, and policy studies.

The overarching goal of this paper is to investigate and develop a comprehensive framework for integrating IoT and AI technologies into aviation health management systems. This framework aims to address the challenges posed by the transition from traditional health monitoring to advanced health management approaches, enhancing predictive maintenance, operational efficiency, and the safety of flight operations.

The paper is organized as follows: Section 2 delineates the research methodologies and analytical procedures employed in this study. Section 3 presents the findings derived from the application of these methodologies, showcasing the impact of the IoT, cloud computing, and AI on aviation maintenance practices within the frame of the new paradigm of aircraft health management. In Section 4, the implications, potential limitations, and practical challenges of the results are examined, setting the stage for future lines of inquiry. Section 5 synthesizes the key takeaways and articulates the broader significance of this research within the context of the aviation industry’s ongoing evolution.

2. Materials and Methods

To achieve the objectives outlined for this study on integrating the IoT and AI into aviation health management systems, a comprehensive approach encompassing various materials and methods will be necessary. This approach will blend theoretical research, technological development, and practical application to address the challenges and leverage the opportunities presented by the transition to advanced health management.

The following is a breakdown of potential materials and methods that could be employed:

• -. Conducting a comprehensive review of existing research and studies on the integration of IoT and AI in aviation and other relevant industries to understand current capabilities, limitations, and best practices.

• -. Grounded theory as research methodology of the transition from health monitoring to health management in aviation through modern IT technologies.

• -. Technical evaluation of the capabilities, limitations, and integration requirements of cloud computing, IoT, digital twins, and blockchain technologies within the context of aviation HM.

• -. Developing integrated system architectures that support the seamless incorporation of the IoT and AI into aviation health management.

Through these materials and methods, this study aims to forge a path toward the effective integration of IoT and AI technologies into aviation health management systems, addressing the identified challenges and fulfilling this study’s objectives. This approach will facilitate the development of robust, efficient, and secure systems capable of enhancing predictive maintenance, operational efficiency, and ultimately the safety of flight operations.

2.1. Grounded Theory as Research Methodology of Transition from Health Monitoring to Health Management in Aviation through Modern IT Technologies

In the ever-evolving landscape of aviation, the shift from traditional health monitoring systems to more integrated health management systems represents a paradigm shift, propelled by advancements in modern information technology (IT) such as the IoT, AI, and others. This study employs a grounded theory approach to explore and justify the decisions necessary for this transition, aiming to develop a comprehensive theoretical framework that guides the integration and implementation of these technologies in aviation health management.

Grounded theory as a research methodology is particularly suited to this study due to its iterative nature and its focus on generating theory from data systematically collected and analyzed [2]. Unlike hypothesis-driven research, grounded theory allows for the emergence of concepts and theories from the data themselves [3]. It is an established methodology for developing conceptual frameworks and models in areas with limited existing research [4]. The iterative process involves concurrently gathering data, constant comparison analysis to find patterns, developing categories and relationships through open and axial coding procedures, and integrative organization of categories into a theoretical formulation [5]. This grounded theory approach is ideal for exploring the complex, multifaceted transition to health management systems in aviation, where predefined theories may not fully capture the nuances of modern IT integration.

The grounded theory approach in this study involves several key methodological steps:

• Gathering qualitative and quantitative data from a variety of sources, including technical reports, IoT device outputs, AI model analyses, and interviews with aviation maintenance and IT experts.

• Employing constant comparative analysis to identify patterns, themes, and discrepancies in the data, facilitating the emergence of grounded concepts related to the integration of the IoT and AI in health management systems.

• Synthesizing the analyzed data into a coherent theoretical framework that explains the transition process, identifies critical factors for success, and anticipates future challenges and opportunities.

Data collection included a structured literature review, analysis of civil aviation authority documents, and focus groups/interviews with aviation maintenance experts. The literature review followed a systematic process of searching scholarly databases, screening based on defined inclusion/exclusion criteria, assessing study relevance and quality, and synthesizing findings related to IoT and AI adoption in aviation maintenance [6]. Civil aviation authority reports were analyzed using qualitative content analysis techniques to extract major themes and trends related to health management strategies and maintenance requirements [7]. Focus groups and interviews with aviation technical staff were conducted using semi-structured protocols to capture the perspectives of IT and digital economy approaches for updating maintenance methods to address AI integration [8].

By employing a grounded theory approach, this study contributes a theoretically grounded framework to the discourse on transitioning aviation maintenance from health monitoring to health management systems using modern IT technologies and underscores the transformative potential of these technologies in enhancing aviation safety, efficiency, and sustainability, paving the way for a new step in aviation maintenance practices.

2.2. Related Works

This subsection is a critical component of our research methodology rooted in grounded theory, which emphasizes the importance of existing knowledge as a substrate for developing new theoretical insights.

Grounded theory, as our guiding research methodology, necessitates a thorough review of prior studies and existing technologies to inform and construct an empirical foundation for our investigation.

This review serves a dual purpose: Firstly, it situates our study within the context of ongoing scholarly conversation, delineating how our work extends or refines the prevailing theoretical frameworks. Secondly, it provides an empirical backdrop against which we can identify patterns, draw inferences, and discover gaps that our research aims to address. In this sense, the literature review is integral to the iterative process of grounded theory, where each piece of evidence contributes to the conceptual scaffolding of our theoretical model.

It assures that our study’s transition from the descriptive realm of health monitoring to the prescriptive domain of health management is not an isolated leap but a methodical progression supported by the confluence of technological advancements and academic inquiry within the aviation industry.

2.2.1. Health Management in Aviation General Approach

The white paper [1] presents an overview of the evolution from aircraft HM to the HMGT approach as a general direction for civil aviation. It emphasizes the importance of utilizing aircraft- and fleet-generated data for efficient maintenance actions, aiming to maintain high aircraft technical availability. The document serves as a call to action for industry stakeholders to embrace the technological advancements and benefits of adopting such a forward-thinking management strategy.

The airplane health management technology building on the prognostics and health management (PHM) system to enhance civil aviation’s safety and economy is introduced in [9]. It emphasizes the importance of extending aircraft operational time without compromising safety, using advanced algorithms and models to prioritize and queue up faults for better decision-making by pilots and airlines. The implementation of this method has been successfully applied to Air China and Singapore Airlines.

Prognostics and health management is crucial for the safety and reliability of aircraft systems, focusing on predicting subsystems’ remaining useful life and mitigating future failures through predictive modeling and real-time data analysis. The paper [10] reviews the current research on PHM in aviation, highlighting key algorithms, applications, challenges, and future research directions.

The research [11] assesses the adoption of condition-based maintenance in aviation on the base of HM systems, identifying challenges and solutions for its practical uptake through a holistic framework. It highlights data, implementation, and future technology integration issues, offering directions for both future research and practical application in aircraft lifecycle management.

2.2.2. Structural Health Monitoring in Aviation

Over the past few decades, the concept of structural health monitoring (SHM) has garnered significant attention from researchers with the aim of developing aircraft that are lighter, safer, and more environmentally friendly. Essentially, SHM seeks to mitigate or eliminate the need for conventional design compromises, such as safety knockdown factors and routine inspections, thereby achieving more cost-efficient maintenance processes [12]. Despite its potential, the adoption of SHM in the industry remains limited, largely due to concerns about the accuracy of reliability assessments and the tangible benefits it offers to the aviation sector. Addressing the reliability of SHM techniques involves close cooperation between SHM specialists and reliability experts [13], while understanding its benefits necessitates a comprehensive, multidisciplinary analysis that examines the impact of SHM on aircraft operations and outlines a strategy for its effective implementation.

Recent studies have shown the value of on-condition maintenance strategies for identifying damage [14] and predicting the remaining useful life of aircraft components [15], which could significantly reduce operational and lifecycle management costs. This is particularly relevant for composite materials, which can sustain damage that is difficult to detect through visual inspection alone, such as delamination caused by high interlaminar shear stresses [16,17,18]. Consequently, aircraft design now often incorporates damage tolerance principles, prioritizing the detection and management of damage over traditional safety-by-design and fail-safe approaches [19,20,21].

The integration of SHM is poised to transform maintenance practices for composite components by enabling the early detection of damage that could otherwise compromise safety. However, reliance solely on visual inspection limits the ability to detect smaller damages, necessitating the definition of acceptable damage limits to ensure structural integrity under operational loads [22]. The promise of SHM extends beyond composite damage, offering potential improvements in the detection and management of various structural issues, thus potentially reducing maintenance costs and downtime while enhancing safety [23].

Nevertheless, the effective implementation of SHM systems requires a thorough understanding of their reliability characteristics. Studies such as those by Cottone et al. [24] and others [25,26] have explored methods to optimize maintenance schedules and reduce costs through reliability-based approaches. Yet, translating these solutions to practical benefits at the aircraft level remains a challenge, with limited research directly addressing the quantifiable advantages of SHM in aviation [27,28].

While SHM promises to streamline aircraft maintenance and improve operational efficiency, its practical application faces hurdles including the need for accurate cost–benefit analysis, standardization for certification, clear requirement definitions, and decision-support tools [29,30,31]. The potential weight savings and design optimization facilitated by SHM systems suggest significant benefits, yet the actual implementation within the aircraft lifecycle is constrained by these challenges. This work aims to bridge the gap in the literature by conducting a multidisciplinary analysis to evaluate the impact of SHM on aircraft operations, ultimately guiding the integration of SHM technologies in the aviation industry.

2.2.3. Health Monitoring of Aircraft Engines

Extensive investigations have been carried out on monitoring the health of gas paths in aeroengines, broadly dividing the approaches into model-based and data-driven methods. Model-based strategies rely on creating an accurate physical–mathematical model to depict the engine’s operational characteristics [32,33]. However, the complexity of and variability in aeroengines, compounded by environmental and operational conditions, often hinder the accuracy of these models. In contrast, data-driven methods leverage actual operational data to uncover insights about the engine’s health, demonstrating superior adaptability and generalization capabilities. This approach has gained significant attention, with researchers like Smart E et al. [34] and Melnyk I et al. [35] developing classification and regression models for anomaly detection in engines. Yet, despite their advantages, data-driven methods face challenges such as the requirement for extensive training data, the complexity of the models, and the handling of high-dimensional, intricately correlated data.

Recent advancements have seen the emergence of information fusion-based health monitoring as a promising direction within data-driven strategies. This approach is categorized into data-level, feature-level, and decision-level fusion, with feature-level fusion particularly noted for its ability to compress information and accurately reflect engine health through the integration of relevant features. Researchers like Jiang W. [36] and Wang H. [37] have demonstrated the successful application of feature fusion techniques in diagnosing and monitoring engine health.

The challenge with utilizing gas-path data directly lies in their high-dimensionality and noise, necessitating dimension reduction or feature extraction for effective monitoring. Methods such as iterative reduced-dimension kernel principal component analysis proposed by Lu F. et al. [38] help in distilling fault features from complex data. While deep learning offers powerful feature extraction capabilities, it suffers from limitations like black-box models and extensive training requirements. Slow feature analysis emerges as a viable alternative, focusing on extracting features that vary slowly over time to distinguish normal operational changes from actual anomalies, as shown in applications by Wang H. [39] and Cheng C. et al. [40].

Given the non-linear and dynamic nature of aeroengine operation, timely and accurate anomaly detection is crucial, requiring methods that minimize false alarms without compromising detection accuracy. The variability in engine states challenges the effectiveness of static thresholds, prompting the need for dynamic threshold adjustment strategies. Research in this area has proposed sliding window models with real-time adaptability, improving the precision of threshold setting and thus enhancing the health monitoring system’s overall performance [41,42,43].

2.2.4. Health Monitoring of Aviation Systems

The evolution of modern aviation hydraulic systems is characterized by trends towards higher pressures, speeds, and power outputs [44], significantly enhancing system efficiency but also introducing pronounced hydraulic pipeline vibrations. These vibrations are primarily induced by the pulsating forces from hydraulic pumps and the structural connections within the system, compounded by fluid–solid interactions within the pipelines that further complexify vibration dynamics.

The study of these vibration characteristics holds substantial value for both scientific research and practical engineering applications [45].

The paper [46] reviews diagnostic methods for the hydraulically powered flight control actuation system in aircraft, focusing on the challenges of diagnosing intermittent and incipient faults and the integration of subsystems for overall system health monitoring. It discusses various diagnostic approaches, including model-based, statistical mapping, and algorithmic methods, and highlights the need for extensive data for effective fault diagnosis. The complexity of subsystem integration increases the likelihood of faults, emphasizing the importance of Integrated Vehicle Health Management tools. However, the scarcity of accessible, detailed data from manufacturers and users complicates the use of data-driven diagnostics, presenting significant opportunities and challenges in the field of aircraft system health monitoring and diagnostics.

The paper [47] introduces a fault detection and isolation method for aircraft gas turbines, utilizing modular convolutional neural networks combined with a physics-driven trend monitoring system. The approach effectively identifies and classifies gas-path faults at the component level, leveraging fault signatures derived from performance degradation monitoring. The study highlights the significant potential of modular neural network architectures in enhancing the accuracy of gas turbine diagnostics.

The study [48] introduces a digital twin framework for aviation systems performance diagnosis, combining component-level mechanism models with data-driven models through a novel integration of particle swarm optimization–extreme gradient boosting and low-rank multimodal fusion, further enhanced by a sparse stacked autoencoder. Highlighting its economic benefits, the framework promises to significantly enhance engine reliability, availability, and efficiency in practical engineering applications.

2.2.5. AI for Aviation Safety

The intersection of AI and aviation safety has generated a significant body of literature, particularly on the application of machine learning (ML).

A comprehensive review [49] delved into the role of natural language processing (NLP) in aviation safety from 2010 to 2022, pinpointing specific NLP methodologies, their efficacy, existing hurdles, and prospective enhancements that could bolster safety and operational efficiency in aviation.

In [50], researchers developed a method using NLP and ML, specifically label spreading and support vector machines, to discern and categorize human factors from aviation incident reports, yielding reliable predictions with a sparse amount of labeled data.

The paper [51] employs ML and data analytics to classify go-around maneuvers at San Francisco International Airport in 2019. This study differentiates between standard and anomalous patterns, finding that the latter often deviate from the norm and present higher energy levels on the initial approach.

A new technique [52] for parsing and categorizing textual data from aviation safety reports via NLP was introduced. When applied to the aviation safety reporting system, this method unearthed significant themes and subthemes, providing deeper insight into aviation safety incidents beyond existing labels.

The paper [53] proposes the Safety Analysis of Flight Events methodology, combining data preprocessing, correlation analysis, supervised learning, and visualization. This approach, when applied to commercial flight data, identifies key factors involved in safety events and underscores the necessity of machine learning interpretations that align with the human understanding of incidents.

Research [54] outlines an approach employing predictive modeling to pinpoint human error in aviation. The results show promise for this methodology’s predictive power, though the study notes the importance of expanding the dataset for future research.

The study [55] evaluates the application of generative language models, including ChatGPT, for summarizing incidents and spotting human factors in aviation safety data. It proposes a human-in-the-loop system for the responsible application of these models, stressing the need for collaborative and iterative refinement in the use of AI for aviation safety.

2.2.6. AI for Aircraft Maintenance

Artificial intelligence has notably become a critical component in aviation maintenance, with its application scope expanding broadly across the sector.

The study [56] delves into the use of convolutional neural networks combined with autonomous drones for streamlining visual inspections in aircraft maintenance. Building upon prior research, the paper presents improved techniques for detecting defects such as dents, enhancing accuracy through targeted image processing and pre-classification methods.

A pioneering approach that marries machine learning with IoT to predict the performance of aircraft wing anti-icing systems is introduced in [57]. Tests show that this artificial neural network-based method surpasses conventional computational fluid dynamics in efficiency and speed, demonstrating its potential utility in aviation maintenance.

The paper [58] provides an overview of statistical and machine learning methodologies applied to analyze aircraft environmental effects, including fuel consumption, emissions, and noise. It synthesizes key studies and points to opportunities for these technologies to contribute to the sustainability of aviation practices.

In [59], deep neural networks and transfer learning are utilized to automatically detect corrosion in aircraft lap joints, with accuracy on par with skilled inspectors, potentially streamlining maintenance workflows and enhancing condition-based maintenance practices.

The study [60] explores active vibration control in helicopters, particularly focusing on individual blade control (IBC) to mitigate hub vibrations. Using fuzzy neural networks among other models, the research underscores IBC’s effectiveness in reducing vibration, offering insights for the development of helicopter vibration control systems.

The paper [61] proposes four data-driven frameworks for estimating the baseline exhaust gas temperature of aeroengines, a critical factor for engine health monitoring and flight safety. Among the tested machine learning models, the Generalized Regression Neural Network was highlighted for its superior accuracy and efficiency, suggesting its suitability for real-world airline deployment.

An examination of machine learning applications in lithium-ion battery research is presented in [62], with particular emphasis on aviation-specific batteries and sustainable aviation technologies. The review assesses the strengths and limitations of various machine learning approaches, paving the way for future innovations in this field.

The paper [63] showcases the application of a multilayer perceptron neural network to model aeroengine transient performance, focusing on thermal management during these periods. Leveraging both simulation data and real-world engine measurements, this model demonstrates accurate replication of thermal transients in engines.

A data-oriented strategy to predict base pressure in suddenly expanded aerodynamic flows is introduced in [64]. This model, trained with existing response data, precisely forecasts base pressure, offering valuable data for optimizing base drag in aerospace applications such as rockets and missiles.

The study [65] brings forward machine learning techniques, particularly deep neural networks, and random forest classifiers, for predicting null motions in control moment gyroscopes, which are pivotal for satellite attitude management. This approach exhibits remarkable accuracy, making it a reliable tool for anticipating null motions beyond the initial training scenarios.

2.2.7. AI in the Documents of Civil Aviation Organizations

Official documents from major civil aviation bodies have increasingly acknowledged AI’s expanding influence in the sector and offer frameworks for its deployment.

The International Civil Aviation Organization (ICAO) emphasizes the urgency of equipping aviation professionals for an AI-centric future [66]. ICAO’s 2017 Training Report underscores the necessity of integrating AI into aviation training programs to foster necessary AI competencies.

In its 2020 artificial intelligence roadmap, the European Union Aviation Safety Agency (EASA) advocates for heightened research into AI’s aviation applications and outlines a strategic oversight framework for AI, considering safety and societal implications [67]. EASA’s roadmap supports the creation, evaluation, and certification of AI systems in aviation.

EUROCONTROL (Brussels, Belgium), in its 2020 Fly AI report, points to the critical need for advancing AI integration, specifically in air traffic management [68]. Its recommendations include the establishment of a unified AI infrastructure and AI-centric educational initiatives. EUROCONTROL also initiated a task force dedicated to propelling AI advancements in aviation.

Looking at the global context, the Federal Aviation Administration (FAA) described the certification for a new course focusing on AI and machine learning in 2022 [69]. This document emphasizes machine learning’s role in maintaining U.S. leadership in aviation systems research and integration.

The International Air Transport Association (IATA), through its 2018 white paper [70], expresses a commitment to promoting AI’s advantages. IATA’s goal is to expedite innovation through research, pilot programs, and partnerships with academia and start-ups.

These organizational initiatives are in harmony with the European Commission’s 2021 strategy [71], which aims to position the EU at the forefront of human-centered, reliable AI, encompassing transportation. The EU’s AI Act marks an acknowledgment by policymakers of AI’s transformative power and the necessity for an ethical framework that fosters excellence and safety.

The EU Commission’s AI package in April 2021 further underscores this commitment, including a communication on a unified European AI approach [72], a review of the AI coordination plan [73], and a proposal for an AI regulation [74].

Globally, ICAO initiatives [75] reflect a commitment to AI knowledge enhancement through partnerships, such as the UN AI for Good Annual Global Summit, AI-focused internships, local AI network support, educational collaborations, workshops, and the potential creation of an AI in Aviation Focus Group.

These documents from aviation authorities not only herald AI’s crucial role in the industry but also emphasize the importance of safety, ethics, and workforce development. There is a clear call for continued R&D, infrastructural updates, and specialized education to cultivate an AI-savvy aviation sector and talent pool, urging academic institutions to align curricula with AI’s rapidly advancing trajectory in aviation.

While the existing literature offers a wealth of insights into the application of information technology within aviation, a notable gap persists in comprehensively understanding and applying these technologies specifically within the realm of aircraft maintenance and repair applications.

The advent of a new concept in aviation health management epitomizes this shift, necessitating a systemic reevaluation of established methodologies. This concept does not merely suggest an incremental change; rather, it demands a holistic rethinking of how IT can be harnessed to enhance aircraft maintenance, repair operations, and, fundamentally, the entire ecosystem of aviation safety and efficiency.

The present study aims to explore these cutting-edge approaches within aviation health management, focusing on the application of modern information technologies in aircraft maintenance and repair. The goal is to cultivate a holistic understanding of IT’s role in aviation, one that respects the legacy of safety and precision while boldly embracing the opportunities presented by technological innovation.

3. Results

3.1. Historical Background of Monitoring and Control Systems’ Development in Aviation

The history of the development of monitoring and control systems in aviation reflects the broader evolution of technology, safety standards, and the increasing complexity of aircraft. This journey from simple mechanical indicators to sophisticated automated systems mirrors the industry’s commitment to safety, efficiency, and reliability. Let us explore this progression in more detail, as follows:

• Individual sensors and simple indicators: In the earliest days of aviation, monitoring and control systems were rudimentary, consisting mainly of mechanical gauges and simple indicators. Pilots relied on these to monitor essential parameters such as speed, altitude, and fuel levels. A basic example is the fuel sensor, which would measure the amount of fuel in the tanks and display this information to the pilot through a mechanical gauge. Light bulbs were used as simple indicators for conditions like gear lock (to show the landing gear was properly engaged) or engine status.

• Dividing more complex systems in aircraft: As aircraft became more complex, it was necessary to divide them into subsystems, each with its sensors for monitoring specific parameters. This division allowed for more specialized and accurate monitoring of the aircraft’s various components. For instance, the hydraulic system would have its pressure sensors and the electrical system its voltage and current sensors. This approach enabled pilots and maintenance crews to diagnose issues more effectively and ensured that each subsystem’s performance could be individually assessed and optimized.

• Automated control systems for individual subsystems: The introduction of automated control systems marked a significant evolution in aviation technology. These systems could automatically adjust the behavior of subsystems without direct pilot intervention, based on sensor data. Autopilots, for example, could control the aircraft’s flight path, altitude, and speed, significantly reducing the pilot’s workload during cruise flight. Engine management systems automatically adjusted fuel flow and engine performance parameters to maintain efficiency and safety.

• Integrated avionics: By the late 20th century, the development of integrated avionics revolutionized how pilots interacted with the aircraft’s systems. Multiple functions were consolidated into multifunction displays, providing pilots with a comprehensive overview of the aircraft’s status, navigation, and communication information through digital screens, replacing the clutter of individual analog gauges.

• Fly-by-wire and further automation: The introduction of fly-by-wire (FBW) systems represented another leap forward. FBW technology transforms traditional manual aircraft controls into a digital interface, where the actions of the pilot are turned into electronic signals sent through wires. These signals are then interpreted by flight control computers, which calculate the necessary adjustments to the actuators on the plane’s control surfaces to achieve the desired response. FBW systems can be designed with mechanical backups for the flight controls or may rely entirely on electronic components. This development allowed for more precise control, reduced weight, and the incorporation of sophisticated flight envelope protection systems, enhancing safety.

• Advanced diagnostics and predictive maintenance: More recently, advanced diagnostics and predictive maintenance have become a focal point. Aircraft are equipped with numerous sensors that not only monitor the status of various systems but also predict potential failures before they occur. This predictive capability is powered by advanced algorithms and big data analytics, enabling maintenance crews to address issues proactively, reducing downtime and improving safety.

• Unmanned aerial vehicles and autonomy: The rise of unmanned aerial vehicles and the move toward fully autonomous flight represent the latest frontier in aviation monitoring and control systems. These systems require sophisticated sensors, actuators, and control algorithms to operate safely without direct human intervention, pushing the boundaries of what is possible in aviation technology.

• Health monitoring systems and predictive maintenance: As aircraft systems became increasingly complex, the need for sophisticated health monitoring systems became apparent. These systems are designed to continuously assess the condition of an aircraft’s critical components, from engines to structural elements, using advanced sensors and data analysis techniques. By monitoring vibration, temperature, and other key indicators, HM systems can identify signs of wear or impending failure in components, allowing for preemptive maintenance actions.

• Integration of HM systems with health management systems: Health management systems take the data provided by health monitoring systems and integrate them with maintenance databases, operational schedules, and logistic support systems. This integration allows for a holistic approach to aircraft maintenance and operations, optimizing the performance and availability of the aircraft while minimizing downtime and maintenance costs. For example, if a sensor detects an abnormality in an engine’s operation, the HMS can automatically schedule a maintenance check, ensuring that the issue is addressed before it leads to a failure.

• Advanced diagnostics and prognostics: Within the realm of health monitoring and management, advanced diagnostic and prognostic capabilities play a vital role. Diagnostics are concerned with identifying the nature and cause of a problem, while prognostics focus on predicting the time until a system or component will no longer perform its intended function. Together, these capabilities enable operators to not just react to problems but to anticipate them, planning maintenance and replacements in a way that maximally preserves the aircraft’s operational availability and safety.

• AI and machine learning: Looking forward, the integration of AI and machine learning into health monitoring and management promises to further revolutionize these systems. AI and ML can analyze vast amounts of data from various sensors more efficiently than traditional methods, identifying patterns and predicting failures with greater accuracy. This capability will enhance the predictive maintenance strategies, further improving the reliability and safety of aviation operations.

3.2. From Health Monitoring to Health Management: Understanding the Shift

The aviation industry is on the brink of a significant transformation, characterized by a paradigm shift from traditional health monitoring systems to a more comprehensive and proactive concept of health management. This transition marks a pivotal change in how the technical condition of aviation equipment is controlled and managed, promising to revolutionize aircraft maintenance, enhance safety, and optimize operational efficiency.

Historically, the aviation industry has relied on health monitoring systems that focus primarily on the detection and diagnosis of issues after they have occurred. These systems, while effective in maintaining a baseline level of safety, operate on a reactive basis. Maintenance and repairs are performed in response to failures or anomalies detected during routine checks or because of in-flight issues. This approach, although tried and tested, leaves room for unexpected downtime, potentially leading to operational inefficiencies and increased costs.

The concept of health management, however, introduces a proactive and predictive approach to the maintenance and operation of aviation equipment. It encompasses not just the monitoring and diagnosis of equipment condition but also the prediction of future states and the management of these conditions to prevent failures before they occur. This paradigm shift is fueled by advancements in information technology, including the IoT, AI, blockchain, and digital twins, which collectively enable a more dynamic, data-driven approach to aircraft maintenance.

The integration of IoT devices allows for the continuous collection of data from various aircraft systems in real time, providing a comprehensive overview of the aircraft’s health. These data, when analyzed through AI algorithms, facilitate predictive maintenance, where potential issues are identified and addressed before they lead to system failures. AI enhances decision-making processes, offering insights that were previously unattainable with traditional monitoring systems.

Digital twins represent another leap forward, offering virtual representations of physical aircraft that can be analyzed in real time to predict wear and tear and optimize maintenance schedules. Meanwhile, blockchain technology promises to revolutionize record-keeping within the health management ecosystem, providing a secure, immutable ledger for maintenance records, part provenance, and regulatory compliance.

The transition to health management carries profound implications for the aviation industry. Firstly, it significantly enhances aircraft safety by enabling issues to be addressed before they compromise operational integrity. This predictive approach also reduces unplanned downtime, increasing aircraft availability and operational efficiency. By optimizing maintenance schedules, airlines can achieve substantial cost savings, reducing the resources expended on unnecessary or emergency repairs.

For regulatory bodies, this shift promises improved oversight capabilities, with access to real-time data and predictive insights enabling more effective monitoring of compliance and safety standards. Manufacturers, too, stand to benefit from the feedback loop provided by health management systems, allowing for continuous improvement of aircraft design and performance based on operational data.

The paradigm shift from health monitoring to health management represents a transformative moment for the aviation industry, promising to redefine how the technical condition of aviation equipment is controlled and managed. By leveraging advanced technologies to adopt a proactive, predictive approach to aircraft maintenance, the industry can achieve unprecedented levels of safety, efficiency, and operational optimization. However, realizing this potential will require careful navigation of technical, regulatory, and organizational challenges, underscoring the importance of a collaborative, industry-wide effort to embrace this new paradigm.

The differences between health monitoring and health management in aviation are summarized in Table 1.

Table 1 highlights the evolution from simply monitoring the health of aircraft systems to actively managing maintenance and repairs to optimize aircraft performance and availability.

3.3. Aviation Health Management Framework

The aviation health management framework is a comprehensive, technology-driven approach designed to optimize the maintenance, safety, and efficiency of aviation operations (Figure 1). At its core, HMGT leverages real-time data analytics, predictive modeling, and integrated communication systems to proactively manage the health of aircraft.

This framework encompasses several key components and processes (Figure 1):

• Data Collection and Integration

Aircraft are equipped with a wide array of sensors and Internet of Things (IoT) devices that continuously monitor various parameters, including engine performance, structural integrity, and system functionality.

Data from these sensors, along with maintenance logs, flight data, and other relevant information, are integrated into a unified data platform. This integration allows for holistic analysis and ensures that all decision-making is based on comprehensive information.

• 2.. Real-Time Monitoring and Analysis

The system continuously monitors the data stream from aircraft sensors, identifying normal operating patterns and detecting any deviations in real time.

Advanced analytics and machine learning algorithms analyze the collected data to diagnose existing issues, predict potential failures, and recommend preventive actions.

• 3.. Predictive Maintenance and Prognostics

AI-driven models predict future aircraft component failures or maintenance needs based on historical data, current performance metrics, and operational conditions.

The system generates maintenance schedules and tasks before issues become critical, allowing for maintenance to be performed more efficiently and with minimal disruption to operations.

• 4.. Decision Support and Actions

Based on the analysis, the system provides detailed recommendations for maintenance actions, including parts replacement and system adjustments.

The AHM framework also supports operational decision-making, offering insights that can optimize flight routes, fuel consumption, and overall aircraft performance.

• 5.. Communication and Coordination

A centralized communication platform ensures that all relevant stakeholders, including pilots, maintenance crews, and operations managers, have access to up-to-date information and recommendations.

The system facilitates seamless coordination with maintenance, repair, and overhaul (MRO) organizations and Original Equipment Manufacturers (OEMs) for parts availability, maintenance scheduling, and addressing systemic issues.

• 6.. Regulatory Compliance and Safety Management

The HM system monitors compliance with aviation regulations and safety standards, automatically documenting maintenance actions and system checks for audit purposes.

Safety management components analyze data for safety trends, contributing to industry-wide safety improvements and helping to prevent incidents before they occur.

• 7.. Continuous Improvement and Learning

The system incorporates feedback mechanisms to continuously update and refine predictive models, ensuring they remain accurate over time.

Insights and best practices derived from the AHM system are shared across the organization and with industry partners to drive continuous improvement in maintenance practices and aircraft design.

By leveraging advanced technologies and integrating data across the aircraft lifecycle, HMGT enhances operational efficiency, improves safety, and reduces maintenance costs, offering a comprehensive strategy for managing the health of aviation assets.

3.4. Analytics of Aviation Health Management Framework

The aviation health management framework represents a paradigm shift in aviation maintenance and operations, moving towards a more proactive, data-driven approach. In this case, the focus on the different types of analytics utilized within the HMGT framework is one of its core components. HMGT systems use a progression from descriptive to diagnostic, predictive, and prescriptive analytics, each adding layers of complexity and value to aircraft maintenance and operations (Figure 2).

Descriptive analytics involves the examination of historical data to understand what has happened in the past within aircraft systems and operations. This foundational level of analytics focuses on data aggregation, summarization, and the creation of dashboards and reports. In the context of HMGT, descriptive analytics helps maintenance teams identify trends, patterns, and anomalies in aircraft performance data, offering insights into the regular operational status and highlighting areas that may require attention. It is the starting point for more advanced analytical processes, providing a clear picture of the aircraft’s health status over time.

Building on descriptive analytics, diagnostic analytics seeks to determine why something happened. It involves more sophisticated data processing techniques, including drilling down into data, data mining, and causal analysis. In HMGT, diagnostic analytics is used to pinpoint the root causes of observed anomalies or performance deviations. By analyzing correlations and patterns within the data, maintenance teams can identify the underlying factors leading to equipment failure or suboptimal performance. This type of analytics is crucial for effective troubleshooting and ensuring that maintenance actions address the actual causes of issues, not just their symptoms.

Predictive analytics represents a more advanced stage, utilizing statistical models and machine learning algorithms to forecast future events based on historical data. Within HMGT, predictive analytics aims to anticipate potential failures and maintenance needs before they occur. By analyzing trends and patterns in the data, predictive models can identify the likelihood of component failures, enabling proactive maintenance planning. This capability significantly enhances operational efficiency and safety, as it allows for the scheduling of maintenance activities at optimal times, minimizing downtime and preventing disruptions.

Prescriptive analytics is the most sophisticated level of analytics, offering recommendations on what actions should be taken to achieve desired outcomes. It goes beyond predicting future events to suggesting specific maintenance actions that can prevent failures or optimize performance. In HMGT, prescriptive analytics uses optimization algorithms and simulation techniques to recommend the best courses of action under various scenarios. This not only helps in preventing potential issues but also in optimizing the overall maintenance strategy for efficiency, cost-effectiveness, and improved aircraft availability.

The types of analytics depicted in Figure 2 of this study underscore the progressive depth and value they bring to aircraft HMGT. Starting from understanding what has happened and why, through predicting what could happen, to prescribing actions to mitigate or prevent issues, analytics transforms aircraft maintenance from a reactive to a proactive and strategic function. As HMGT systems evolve, the integration of these analytics types promises to significantly enhance the safety, reliability, and efficiency of aviation operations.

The integration of the IoT and AI significantly enhances the capabilities of the different types of analytics in aircraft HMGT. Table 2 describes the role of the IoT and AI across the analytics spectrum.

Table 2 highlights how the IoT serves as the foundation for collecting detailed and continuous data across the aircraft’s systems and components. AI, on the other hand, brings intelligence to the process, turning data into actionable insights, forecasts, and recommendations, thereby enhancing the decision-making processes within HMGT.

For the various types of analytics within HMGT, different AI technologies can be applied to enhance analysis and decision-making processes. The breakdown of AI types and their applicability to each analytics category are shown in Table 3.

By leveraging the appropriate AI tools and techniques for each type of analytics, aviation stakeholders can optimize aircraft performance, improve safety, and reduce operational costs.

3.5. Service Delivery Model for Health Monitoring and Health Management Systems

The service delivery model for HM and HMGT encompasses a range of activities, from data collection to actionable insights for maintenance and operational decisions. Table 4 outlines a comparative service delivery model for each, highlighting their processes, technologies, and outcomes.

Table 4 illustrates how aviation health monitoring serves as a foundational layer, focusing on the collection and basic analysis of data to monitor the health status of aircraft. Aviation health management builds upon this, employing more sophisticated data processing, analytics, and decision-support technologies to proactively manage aircraft health, optimize maintenance, and improve operational efficiencies. Together, these models represent a comprehensive approach to maintaining and managing the health of aviation assets.

HMGT provides a range of unique services designed to optimize aircraft performance, enhance safety, and reduce operational costs through proactive maintenance and management strategies. These services leverage advanced technologies, including the IoT, AI, and big data analytics, to predict potential issues before they occur and prescribe the most effective actions. Table 5 demonstrates some key services provided by health management.

By integrating these services, airlines and aircraft operators can significantly improve the reliability, safety, and efficiency of their fleets, leading to better operational outcomes and cost savings.

The unique services provided by aviation HMGT rely on diverse and sophisticated sources of information to ensure accurate predictions, efficient maintenance planning, and overall operational optimization. An overview of the key sources of information for these services is shown in Table 6.

These sources of information form a complex data ecosystem that feeds into the AI and analytics tools underpinning aviation HMGT services. By effectively integrating and analyzing data from these diverse sources, HMGT systems can deliver actionable insights, enabling proactive maintenance strategies, operational efficiency, and enhanced safety in aviation operations.

3.6. Architecture of Aviation Health Management System

The architecture of an HMGT system, especially in the context of aviation, is designed to handle vast amounts of data from diverse sources, process these data to extract actionable insights, and support decision-making processes that enhance aircraft performance and safety (Figure 3).

There are some layers within this architecture:

• Data collection layer:

  • Sensors and IoT devices, which continuously monitor health and performance metrics such as temperature, pressure, vibration levels, and usage cycles. Each sensor is designed for specific components, from engines to hydraulic systems, ensuring comprehensive coverage.

  • Flight data recorders capture detailed flight operational data, including parameters related to flight controls, engine performance, and environmental conditions, which are crucial for both real-time monitoring and post-flight analysis.

  • Electronic maintenance logs digitally record all maintenance actions performed, providing a detailed history for each aircraft component. This includes information on part replacements, repairs, and routine inspections.

• Data transmission and integration layer:

  • Communication networks, which utilize satellite communications for real-time data transmission during flight and ground-based networks for data offloading post-flight. This ensures that data collected in flight are rapidly available for analysis.

  • Data integration platforms aggregate data from multiple sources, including sensor data, FDR data, maintenance logs, and even external databases such as weather services or manufacturer databases. This integrated data platform is key to providing a holistic view of aircraft health.

• Data processing and storage layer:

  • Data preprocessing tools address data quality issues, normalize data from different sources to a common scale, and fill in gaps in data records. This step is crucial to ensure the analytics engines can accurately interpret the data.

  • Cloud storage and local databases offer a flexible and scalable solution for storing processed data, with redundancy and disaster recovery capabilities. Local storage is used for storing sensitive or critical operational data, ensuring they are accessible even when cloud connectivity is not available.

• Analysis and decision-support layer:

  • Predictive analytics engines employ advanced machine learning algorithms and statistical models to identify patterns and trends in the data that may indicate potential failures or maintenance needs before they become critical issues.

  • Diagnostic tools use algorithms to analyze symptoms and fault codes to pinpoint the underlying causes of observed anomalies, reducing diagnostic time and increasing accuracy.

  • Optimization algorithms apply techniques such as linear programming and genetic algorithms to find the optimal solutions for maintenance scheduling, resource allocation, and other operational challenges, balancing cost, performance, and regulatory compliance.

  • Regulatory compliance modules automatically check maintenance activities and operational data against regulatory standards and requirements, ensuring compliance and simplifying audit processes.

• Actionable insights and interface layer:

  • User dashboards and alerts as customizable interfaces display real-time and historical data, health indicators, and predictive alerts. These dashboards are designed for different user roles, from maintenance technicians to fleet managers, providing relevant information that supports informed decision-making.

  • Maintenance planning and scheduling tools that incorporate predictive maintenance recommendations, allowing maintenance teams to schedule work orders, assign resources, and manage parts inventory in an efficient manner.

• Feedback and continuous improvement layer:

  • Performance monitoring tools track the outcomes of maintenance actions and operational decisions to assess their effectiveness. This includes monitoring aircraft performance, maintenance costs, and downtime, providing data for continuous improvement.

  • Machine learning feedback loops. As new data are collected, they are fed back into the machine learning models, allowing them to learn from the outcomes of previous predictions and decisions. This iterative learning process continually improves the accuracy and reliability of predictive insights.

This architecture is underpinned by several core characteristics common to all the layers:

• Modularity is provided by design with interchangeable components, allowing for upgrades or changes without disrupting the entire system.

• Security and privacy incorporate encryption, secure data transmission protocols, and access controls to protect sensitive data and ensure privacy.

• Interoperability is provided by the capability to integrate with existing aviation systems and standards, facilitating data exchange and collaboration across different platforms and stakeholders.

• User-centric design adapts interfaces and functionalities to meet the needs of various users, from pilots and maintenance crews to operations managers, enhancing usability and adoption.

By integrating these layers and characteristics, the HMGT system architecture provides a robust framework for enhancing aircraft maintenance, safety, and operational efficiency through data-driven insights and proactive management strategies.

3.7. Sensors of HM and HMGT Systems

Sensors are the linchpins of aircraft health monitoring and management systems, essential to the modern aviation maintenance ecosystem. They enhance safety, underpin reliability, and ensure cost-effectiveness by enabling a transition from reactive fixes to proactive, data-driven strategies.

The predictive maintenance paradigm, a leap beyond traditional reactive strategies, relies heavily on sensor data. The power of sensors is amplified through their integration with cloud-based technologies and artificial intelligence. AI algorithms, trained on this information, can make predictive assessments, scheduling maintenance proactively to minimize disruption and cost. This integration also facilitates a dynamic feedback loop where data from executed maintenance inform future AI predictions, creating an ever-improving system of health management. Thus, sensors act not only as data collectors but as the foundational elements of an adaptive learning network, crucial to the aircraft’s continual operational readiness.

Sensors for aviation health monitoring and management systems can be classified based on their function, the type of data they collect, their location on the aircraft, and their role in the health monitoring system (Figure 4).

Sensors play an indispensable role in the health monitoring and management systems of aircraft. By continuously measuring and collecting data on various parameters, sensors enable the detection of potential issues, predictive maintenance, and overall performance monitoring.

3.8. Artificial Intelligence as a Component of HMGT Systems

Artificial intelligence plays a central and transformative role in the architecture of a health management system, especially within aviation. It infuses intelligence across various layers of the system (Figure 3), enhancing data analysis, decision-making processes, and operational efficiencies.

AI integrates into each relevant layer of the architecture:

• In the data collection layer, AI does not directly collect data; it influences the development and deployment of smart sensors and IoT devices. AI algorithms can preprocess data at the edge (close to where data are generated), filtering out noise and reducing the volume of data that needs to be transmitted and processed centrally. This selective data transmission improves efficiency and ensures that only relevant data are sent for further analysis.

• In the data transmission and integration layer, AI plays a role in optimizing data transmission protocols and ensuring data integrity during transmission. It can dynamically adjust data transmission based on network availability and bandwidth, ensuring that critical data are prioritized. Furthermore, AI techniques can be used to integrate and synchronize data from disparate sources, resolving discrepancies and providing a unified data model.

• In the data processing and storage layer, AI, particularly machine learning models, is employed to preprocess and clean the data, identifying and correcting errors, filling in missing values, and standardizing data formats. This layer uses AI to ensure that the data stored, whether in the cloud or locally, are of high quality and ready for analysis.

• In the analysis and decision-support layer, AI is especially effective, offering several functionalities:

  • -. In predictive analytics, machine learning models analyze historical and real-time data to predict future system failures, wear and tear, and maintenance needs.

  • -. As diagnostic tools, AI algorithms diagnose issues by analyzing patterns and anomalies in the data, determining the root cause of system alerts and malfunctions.

  • -. As optimization algorithms, AI uses complex algorithms to optimize maintenance schedules, resource allocation, and operational planning, ensuring optimal outcomes based on a multitude of constraints and objectives.

  • -. In regulatory compliance modules, AI can automate the monitoring and reporting of compliance with aviation regulations, analyzing operational data against regulatory requirements and highlighting potential non-compliance issues.

• In the actionable insights and interface layer, AI enhances user interfaces with intelligent alerts and recommendations, making complex data understandable and actionable for different users. It personalizes dashboards and reports based on user roles and preferences, ensuring that each user receives relevant and timely information.

• In the feedback and continuous improvement layer, AI models continuously learn from new data and outcomes, improving their predictions and recommendations over time. This layer enables the system to adapt to changing conditions and evolve based on performance feedback, ensuring that the health management system becomes more accurate and effective.

AI is integral to enhancing the capabilities of health management systems in aviation, providing the intelligence needed to transform raw data into actionable insights, optimize maintenance and operations, and ensure safety and regulatory compliance.

Deploying AI onboard an aircraft for HMGT tasks presents both opportunities and challenges. While the ground segment of an HMGT system benefits from robust computational resources and comprehensive data access, integrating AI directly onboard introduces real-time processing capabilities that can enhance certain aspects of aircraft operation, safety, and maintenance:

• AI can continuously monitor sensor data from critical aircraft systems (engines, avionics, hydraulics, etc.) in real time, instantly detecting anomalies or deviations from normal operational parameters.

• Once an anomaly is detected, onboard AI can quickly diagnose potential issues, identifying their nature and severity. This immediate diagnostic capability can be crucial for in-flight decision-making and safety.

• By analyzing data trends directly onboard, AI can predict potential failures or maintenance needs before they occur, even without real-time communication with ground systems. This is particularly useful for long-haul flights or operations in remote areas with limited connectivity.

• Based on predictive analysis, onboard AI can suggest optimal maintenance actions, potentially advising the crew to adjust operations to mitigate risks until maintenance can be performed. It can also pre-notify ground maintenance teams about upcoming needs.

• AI can monitor flight systems and environmental conditions to enhance safety, providing pilots with real-time insights into any emerging issues and suggesting corrective actions.

• In critical situations, onboard AI can assist in assessing the situation, providing recommendations for emergency procedures based on the most current system health data and operational parameters.

• AI algorithms can analyze flight data in real time to optimize fuel consumption, advising pilots on adjustments to speed, altitude, or routing that could save fuel without compromising safety or schedule.

• Onboard AI can adjust flight plans in response to changing weather conditions, air traffic, or aircraft performance issues, optimizing for efficiency and safety.

There are some problems with AI integration in onboard systems:

• Onboard systems have limited space, power, and cooling capabilities, which can restrict the computational resources available for AI algorithms. Solutions must be designed to operate within these constraints.

• Introducing AI into critical flight systems raises security and safety concerns. Systems must be designed to be fail-safe, with robust security measures to prevent unauthorized access or interference.

• Any onboard AI system must comply with aviation regulatory standards, which may require extensive testing and certification processes to ensure safety and reliability.

All AI-driven analytics, including predictive maintenance, fault diagnosis, and decision support, are conducted on ground-based servers, where computational resources are less constrained. This setup allows for the use of more sophisticated AI models and analytics tools that may be too resource-intensive for onboard systems. The ground system can integrate data not just from the aircraft in question but from the entire fleet, as well as external data sources like weather services, to enhance the accuracy and efficacy of AI analyses.

Aircraft are equipped with systems capable of transmitting real-time data from onboard sensors to the ground-based AI via satellite or air-to-ground communication networks. This ensures that the AI has access to up-to-date information for its analyses. Pilots and onboard systems have the capability to receive real-time insights, recommendations, and alerts from the ground-based AI, effectively making the AI’s functionality available onboard through communication links.

Transferring the functionality of onboard AI systems to a ground-based component, while maintaining real-time communication with this “quasi-onboard” AI from the aircraft, is an additional strategy. This approach leverages the best of both worlds: the computational power and scalability of ground-based systems and the real-time data collection and communication capabilities of onboard systems.

The above-mentioned “quasi-onboard” strategy has both benefits and limitations:

• Functionalities and benefits:

  • Even with the AI processing occurring on the ground, aircraft can benefit from real-time predictive maintenance insights and fault diagnosis. Recommendations and critical alerts can be communicated back to the aircraft for immediate action.

  • Flight optimization, fuel efficiency recommendations, and adjustments to flight plans based on weather or traffic can be computed on the ground and relayed to the aircraft in real time, aiding in operational decisions.

  • Emergency situations or unexpected operational issues can be quickly addressed by the ground-based AI, providing pilots with immediate support and recommendations based on comprehensive data analysis.

• Potential limitations:

  • The effectiveness of this model depends on the reliability and latency of the communication system. In critical situations, even minimal delays in data transmission or receiving AI insights could be significant.

  • Transmitting detailed sensor data in real time requires significant bandwidth, which may be limited or costly, especially via satellite communication.

  • Data transmission between aircraft and ground systems introduces potential vulnerabilities that must be addressed with robust cybersecurity measures.

  • This model must comply with aviation regulations regarding data transmission, processing, and decision-making support, which may vary across jurisdictions.

Transferring AI functionality to the ground while maintaining real-time communication with aircraft offers a practical and efficient approach to implementing AI in aviation health management. It allows for leveraging advanced AI capabilities without the onboard computational constraints and integrates wider data sources for more comprehensive analyses.

3.9. Internet of Things as a Component of HM and HMGT Systems

The IoT represents a significant advancement over traditional, non-connected sensors in aircraft for several key reasons.

Figure 5 depicts a simplified representation of an aircraft along with various traditional aircraft sensors. These might include sensors for temperature, pressure, flow rates, position, velocity, and other physical parameters. The representation suggests a disconnection between the sensors and any centralized system, highlighting the traditional approach where sensors operate independently without interconnectivity or real-time data analysis. Data collected from these sensors is typically downloaded and analyzed post-flight or during routine maintenance checks. Figure 5 serves an illustrative purpose, providing a schematic representation of an aircraft equipped with a variety of sensors. The symbols used in this figure are symbolic and designed to visually represent a broader set of sensors that are critical to the operation of aircraft health monitoring systems. Figure 4 provides a detailed classification and the functional spectrum of sensors, explaining how each contributes to the overall health management of the aircraft.

The key characteristics of traditional sensors are as follows:

• Each sensor functions independently, monitoring specific parameters without communicating with other systems.

• Data are often retrieved manually and require physical access to the aircraft for download.

• The analysis of data is retrospective, occurring after the flight has ended or during scheduled maintenance.

• Maintenance decisions are reactive, based on the analysis of historical data rather than real-time monitoring.

Figure 6 conceptualizes an aircraft equipped with an IoT-based health monitoring system. The aircraft is surrounded by various sensors representing digital connectivity and data analysis tools as a network of interconnected devices and systems. This reflects the IoT’s capability to enable real-time communication and integration of sensor data for on-the-go analysis and decision-making.

The key characteristics of an IoT-based HM system are as follows:

• Sensors on the aircraft are networked, allowing them to communicate with each other and with centralized systems, enabling integrated health monitoring.

• Sensor data can be transmitted in real time to ground stations or cloud-based platforms for immediate analysis.

• The system can analyze data on the fly to predict maintenance needs, scheduling proactive maintenance actions to avoid failures.

• Pilots and ground personnel receive instant insights into the aircraft’s health, which informs in-flight decisions and ground operations.

Figure 7 represents a continuation of the evolution from traditional sensors and an IoT-based health monitoring system to a cloud-based integral health management system, illustrating the advanced functionalities of the IoT in aviation health management, such as detailed analytics dashboards, automated maintenance alerts, and system-wide integration from the aircraft to maintenance crews and operations centers.

The advanced functionalities of an IoT-based HMGT system are as follows:

• Visual representations and dashboards showcasing the analysis of real-time and historical sensor data.

• Automated systems for issuing maintenance alerts and safety warnings based on predictive analytics.

• A fully integrated operational ecosystem that connects flight operations, maintenance, supply chain logistics, and safety management.

While simple sensors can collect data, IoT technology enhances the scope, efficiency, and utility of data collection and analysis in health monitoring and health management systems. There are some reasons why the IoT is necessary and beneficial in this context:

• IoT devices can continuously monitor aircraft systems in real time, providing a constant stream of data. Unlike simple sensors that might only record data for later retrieval, IoT devices can transmit these data immediately for analysis.

• Data collected by IoT devices can be accessed remotely, allowing maintenance teams, flight operations centers, and other stakeholders to monitor aircraft health in real time, regardless of the aircraft’s location.

• The IoT enables the integration of data from multiple sources and sensors across the aircraft, providing a holistic view of its health. This integrated data can be analyzed more effectively to identify trends, patterns, and anomalies.

• With the IoT, data can be fed into predictive analytics models in real time, allowing for early detection of potential issues before they lead to system failures. This facilitates proactive maintenance, reducing downtime and maintenance costs.

• The IoT provides a scalable infrastructure that can easily accommodate additional sensors and devices as technology advances or as new monitoring requirements emerge.

• IoT devices can be integrated with other systems, such as flight operations, maintenance scheduling, and inventory management systems, creating a seamless ecosystem that supports comprehensive health management.

• Real-time data from IoT devices can provide insights into aircraft performance and operational efficiency, enabling airlines to optimize fuel consumption, route planning, and flight operations.

• Continuous monitoring and predictive maintenance enabled by the IoT reduce the risk of in-flight failures and enhance overall aircraft safety. Maintenance teams can address issues before they become critical, ensuring that aircraft meet the highest safety standards.

• IoT-driven predictive maintenance helps airlines save costs by reducing unplanned downtime, extending the lifespan of aircraft components, and optimizing maintenance workflows.

• By optimizing flight operations and maintenance schedules, the IoT can contribute to reduced fuel consumption and emissions, supporting the aviation industry’s sustainability goals.

Compared to simple sensors, IoT technology in aviation health monitoring and management systems offers enhanced real-time monitoring, advanced predictive analytics, and improved operational efficiency. It enables a more proactive, data-driven approach to aircraft maintenance and operations, ultimately enhancing safety, reducing costs, and supporting sustainability initiatives.

At the same time, the increased connectivity inherent in IoT systems introduces cybersecurity risks that must be managed through robust security protocols and encryption, and the vast amount of data generated by IoT devices require effective data management strategies to store, process, and analyze data efficiently.

Sensors connected to the Internet, as part of the IoT setup in aviation, enable the exchange of information both on the ground and in flight. This capability is facilitated through various communication technologies that provide Internet connectivity to aircraft during flight. Satellite communication (Satcom) is the primary method for providing Internet connectivity to aircraft in flight. Satcom uses geostationary satellites to relay data between the aircraft and ground stations, enabling global coverage. It supports the transmission of data collected by onboard IoT sensors back to ground-based operations centers, maintenance teams, and other relevant parties in near-real time.

3.10. Optimal Construction of Aviation Health Systems

The design and implementation of aviation health systems present a complex optimization challenge, balancing the initial creation costs, operational expenses, and the paramount importance of flight safety. This section delves into a framework aimed at guiding the optimal construction of both health monitoring and health management systems within the aviation sector.

At the core of this approach is the development of objective functions that encapsulate the multifaceted goals of minimizing costs while maximizing system effectiveness and ensuring unwavering adherence to safety standards.

• Health monitoring system optimization

Objective function—minimize the total cost of the health monitoring system, including the costs of creation and operation, while maximizing the system’s effectiveness in monitoring health and safety.

$min F_{HM}=C_{c,HM}+C_{o,HM}-{\gamma }_{HM}{E}_{HM}$

where the following terms are used:

$F_{HM}$ is the objective function for the health monitoring system.

$C_{c,HM}$ represents the initial costs associated with developing and implementing the health monitoring system.

$C_{o,HM}$ includes the ongoing operational costs (maintenance, data analysis, etc.).

$E_{HM}$ quantifies the effectiveness of the health monitoring system in terms of its ability to detect potential issues, contributing to overall safety.

${\gamma }_{HM}$ is a weighting factor that values the system’s effectiveness.

Constraints:

• The system must meet or exceed a predefined effectiveness threshold to ensure adequate safety monitoring.

• Budget constraints for both creation and operational costs.

• b.. Health management system optimization

Objective function—minimize the total cost of the health management system, factoring in the costs of system creation and operation and the benefits of predictive maintenance and enhanced safety.

$\min F_{HMGT}=C_{c,HMGT}+C_{o,HMGT}+C_{PM}+\alpha C_F-\beta S_{HMGT}-{\gamma }_{HMGT}{E}_{HMGT}$

where the following terms are used:

$F_{HMGT}$ is the objective function for the health management system.

$C_{c,HMGT}$ and $C_{o,HMGT}$ represent the initial and ongoing costs, respectively.

$C_{PM}$ is the cost of performing predictive maintenance.

$C_F$ is the cost associated with failures, including direct and indirect costs.

$S_{HMGT}$ quantifies the safety benefits derived from the health management system.

$E_{HMGT}$ represents the system’s overall effectiveness in managing aircraft health.

${\alpha , \beta ,\gamma }_{HMGT}$ are weighting factors for the costs of failures, safety benefits, and system effectiveness.

Constraints:

• The system must achieve a certain level of effectiveness and safety benefits.

• Budget constraints for creation, operation, and maintenance actions.

The depth of integration involves the degree to which the system is embedded within aircraft operations and maintenance processes. A deeper integration implies a more comprehensive system that can offer greater benefits but at potentially higher costs.

To account for depth, one can modify the objective functions to include terms that specifically assess the impact of system depth on costs and benefits, for instance, adding a factor that represents the depth of integration and adjusting the weighting factors to reflect how the depth influences costs and effectiveness.

The presented general description of the problems of optimal construction of aviation health systems offers a structured methodology for evaluating and optimizing the myriad factors that influence the effectiveness and efficiency of health monitoring and management systems and provides a basis for future research in this direction.

3.11. Cloud Technology as a Core Component of HMGT

The implementation of a health management concept leveraging cloud technologies, Internet-connected platforms, and the future integration of digital twins and blockchain fosters a new level of interaction between airlines and aircraft manufacturers. This interaction is driven by enhanced data sharing, transparency, and real-time feedback mechanisms.

The cloud-based HMGT system structured across three hierarchical levels—airplane, MRO, and virtual center—represents a comprehensive approach to aircraft maintenance and operations. These levels interact in one single ecosystem, leveraging the IoT and AI (Figure 8).

Each level leverages AI to perform specific tasks suited to its operational scope and objectives.

3.11.1. Level 1: Airplane

AI algorithms onboard the aircraft analyze sensor data in real time to make immediate operational decisions, such as optimizing flight paths for fuel efficiency or adjusting engine settings for performance optimization.

AI performs continuous monitoring of aircraft systems and components, detecting anomalies and identifying potential faults before they escalate. It uses pattern recognition and machine learning to diagnose issues based on sensor data, providing pilots and maintenance crews with early warnings.

By analyzing data trends, onboard AI can predict when specific components might fail or require maintenance, generating alerts for the crew and automatically notifying ground maintenance teams.

In critical situations, AI assists in evaluating the severity and implications of onboard system failures, supporting the crew with recommendations for emergency procedures or system redundancies to utilize.

3.11.2. Level 2: MRO Level

AI at the MRO level aggregates data from across the fleet to identify common patterns and potential systemic issues, forecasting maintenance needs before they result in operational disruptions. It prioritizes maintenance tasks based on urgency and operational impact.

Utilizing advanced algorithms, AI optimizes maintenance schedules, ensuring aircraft availability and efficient use of resources, including personnel, parts, and maintenance bays. It dynamically adjusts schedules based on changing operational demands and incoming data from aircraft.

AI analyzes trends in part usage and failure rates to predict spare part needs, optimizing inventory levels and logistics to ensure parts are available where and when needed, reducing waiting times and costs.

By analyzing maintenance outcomes and operational data, AI identifies opportunities for improving the reliability of aircraft components and systems, suggesting design modifications or changes in maintenance practices.

3.11.3. Level 3: Virtual Center

At the highest level, AI synthesizes data from individual aircraft and MRO centers to inform long-term strategic decisions, such as fleet renewal, expansion, and long-term budgeting for maintenance and operations.

AI monitors compliance with aviation regulations and safety standards across the fleet, using data analytics to identify areas of risk or non-compliance and suggesting corrective actions.

The AI integrates external data sources, such as weather information, air traffic data, and manufacturer notices, with internal fleet data to enhance operational planning, safety, and efficiency.

Utilizing machine learning and big data analytics, AI at the virtual center level develops sophisticated predictive models that forecast fleet-wide trends, maintenance needs, and operational optimizations.

Across all levels, AI functionality is tailored to the specific needs and operational focus of each tier, from real-time decision-making onboard aircraft to strategic planning at the virtual center. The integration and communication between these levels ensure a cohesive and comprehensive approach to aviation health management, leveraging AI to enhance safety, efficiency, and reliability across the aviation ecosystem. This multi-tiered AI application facilitates a proactive and predictive maintenance strategy, transforming traditional reactive approaches into a more advanced, data-driven paradigm.

The functionality of an aviation HMGT system varies between dynamic (in-flight) operations and stationery (on-ground) activities across the three hierarchical levels: airplane, MRO, and virtual center. Table 7 illustrates this distinction.

This table showcases how the health management system’s functionality adapts to the operational status of the aircraft, ensuring continuous health monitoring, maintenance optimization, and strategic decision-making both in flight and on the ground.

Intermediate landings present a unique opportunity to perform certain health management tasks that are typically categorized under “Stationary Functionality”, albeit in a more time-constrained environment (Table 8). This situation bridges the gap between dynamic and stationary functionalities, utilizing the short ground time for immediate maintenance actions, diagnostics, and data synchronization activities that cannot be performed in flight but are too urgent to wait for extended ground time.

The key considerations for intermediate landings are as follows:

• Activities must be carefully chosen to ensure they fit within the limited time available, prioritizing those that address immediate safety, compliance, or operational efficiency concerns.

• Advanced planning and coordination with ground maintenance teams are essential to execute the necessary tasks efficiently during the stopover. Predictive maintenance insights play a crucial role in this planning.

• Given the limited time, data offloading and uploading activities may need to prioritize critical information that influences immediate operational decisions or maintenance actions.

• Ensuring the availability of maintenance personnel and necessary parts at the stopover location is crucial for addressing identified issues without delay.

Intermediate landings offer a strategic advantage in the health management process, allowing for a blend of dynamic and stationary functionalities to optimize aircraft health, safety, and efficiency. This integrated approach ensures that the aircraft maintains high operational readiness and safety standards throughout its journey, leveraging every opportunity for maintenance and data exchange.

3.12. Blockchain Technology Potential for HMGT

Blockchain technology has several compelling applications in implementing the concept of aviation health management due to its intrinsic properties of decentralization, transparency, security, and immutability.

Blockchain can be used to create an unalterable record of an aircraft’s maintenance history. Each maintenance entry, once recorded on a blockchain, cannot be changed, creating a secure and permanent record. This helps ensure that all parties involved in the aircraft’s operation have access to trustworthy data, which is crucial for safety and compliance.

Blockchain technology can enable end-to-end traceability of aircraft parts. From manufacture through to installation and subsequent maintenance, every transaction associated with a part can be recorded. This provides a clear provenance trail, which is vital for ensuring that only certified and appropriate parts are used, thereby preventing the use of counterfeit parts.

Compliance with aviation regulations can be streamlined using blockchain. It can automatically record all necessary information in a format that is easily auditable, saving time and reducing the likelihood of human error. Smart contracts, self-executing contracts with the terms of the agreement between buyer and seller being directly written into lines of code, could automatically enforce and confirm compliance with regulatory requirements.

Blockchain can enhance the efficiency of supply chain management within aviation health management by providing real-time visibility into the supply chain status. This helps to ensure the right parts are available when needed and reduces the risk of delays caused by parts shortages.

Data sharing between airlines, maintenance teams, manufacturers, and regulatory bodies is essential for effective aviation health management. Blockchain facilitates this sharing in a secure manner, ensuring data integrity and helping to build trust between different stakeholders.

Blockchain-based systems can provide robust identity management and access control for accessing sensitive aircraft data and maintenance records. By using cryptographic keys and permissions, blockchain ensures that only authorized personnel have access to specific datasets.

The trust and transparency offered by blockchain encourage better collaboration between various stakeholders. When all parties can access a shared source of data that is reliable and up to date, they can collaborate more effectively.

While blockchain offers many potential benefits, there are also challenges to its implementation in aviation health management. These include the integration with existing systems, ensuring that the technology scales effectively, and addressing any regulatory challenges associated with adopting a new technology. There is also the need for the aviation industry to develop the required blockchain expertise and to ensure all participants in the network understand and trust the technology.

3.13. Framework for Integrating Cutting-Edge Technologies into Aircraft Health Management

To develop a model for the multi-level aircraft health management system described, it is essential to understand the complex interplay between various factors, including technological and functional elements.

The model can be structured in two layers:

• Functional relationship layer which models the functionality of the system from data collection to predictive maintenance decision-making.

• Technology integration layer which models how various technologies like the IoT, AI, and blockchain interact and support the system.

3.13.1. Functional Relationship Layer

The functional relationship layer of the multi-level aircraft health management system (Figure 8) represents how data flow through various levels of processing and analysis, ultimately supporting maintenance decisions. Below is a description of the model components and equations, elaborating on how each step transforms the data from raw sensor inputs to actionable intelligence:

• Sensor data collection is the lowest level of the system. It involves the collection of raw data from various sensors embedded throughout the aircraft. These sensors monitor a wide range of parameters such as engine health, structural integrity, system performance, and environmental conditions.

• At the local processing level, data collected by sensors are preliminarily processed within the aircraft’s onboard systems. This processing might include filtering noise from data, normalization, and basic anomaly detection. The goal here is to prepare data for more sophisticated analysis either onboard or after transmission to ground-based systems.

• Data that have been locally processed are then sent to the MRO center where they are further integrated and analyzed. This integration may involve aggregating data from individual aircraft over time or combining them with data from other aircraft to detect patterns or trends that may not be apparent from single aircraft data.

• Virtual integrated processing centers are the highest level. At this level, data integrated at MRO centers are further analyzed in a virtual integrated processing center. This center utilizes data from the entire fleet to perform comprehensive analytics, such as predictive maintenance modeling, fleet optimization, and long-term trend analysis. This stage leverages advanced data analytics, machine learning, and possibly simulation models.

For each component, we can define the data transformation:

• $x_{1t}$ represents a single data point collected from sensors at time $t$.

• $\left\{x_{1t}\right\}$ denotes a set of sensor data points, which could include data from various sensors and possibly from various times up to $t$. In the context of fleet-wide analysis, it would encompass sensor data from all aircraft in the fleet.

• $y_{2t}$ symbolizes the processed data at time $t$ after the first level of local processing.

• $\left\{y_{2t}\right\}$ includes all processed data points that have been locally processed on the aircraft and are now being considered for further integrated analysis.

• $z_{3t}$ are data integrated and analyzed at the MRO center at time $t$.

• $w_{4t}$ is the fleet-wide advanced analytics output at time $t$.

• The local processing function $y_{2t}=f_1\left(x_{1t}\right)$ could involve filtering, normalization, and preliminary statistical analysis aimed at preparing data for more detailed analysis.

• The MRO integration function $z_{3t}=f_2(x_{1t},y_{2t})$ could include more sophisticated statistical methods, trend analysis, and integration with historical data to form a comprehensive picture of the aircraft’s health. It may also involve cross-referencing data from similar types or models of aircraft.

• The fleet-wide analysis function $w_{4t}=f_3(z_{3t},\left\{x_{1t}\right\},\left\{y_{2t}\right\})$ represents the use of advanced machine learning algorithms, predictive modeling, and possibly simulation to make strategic maintenance decisions. This function takes into account not just current and past data from one aircraft but integrates information across the entire fleet.

Example Application

For instance, if the sensor data $x_{1t}$ include engine temperature and pressure readings, $y_{2t}$ might represent these readings adjusted for flight altitude and environmental conditions. $z_{3t}$ could then integrate these readings with historical data to assess engine wear, while $w_{4t}$ might predict when the engine will require maintenance based on similar patterns seen across the fleet.

This structured approach allows for each step in the process to be optimized separately for efficiency and accuracy and provides clear pathways for data flow and transformation within the aircraft health management system.

3.13.2. Technology Integration Layer

The technology integration layer of the multi-level aircraft health management system focuses on how various advanced technologies are implemented to support and enhance the functionalities of the aircraft health management system. This layer interlinks technologies such as the IoT, AI, and blockchain with the data processing stages, ensuring the efficient, secure, and intelligent management of aircraft health data:

• IoT devices are embedded sensors and actuators that collect real-time data from various aircraft systems. These devices are critical for real-time data acquisition and communication, enabling the continuous monitoring of aircraft systems.

• AI involves the use of machine learning models and algorithms to analyze large volumes of data, identify patterns, make predictions, and provide actionable insights. AI technologies are crucial in processing and analyzing data at all stages, from local processing on the aircraft to integrative analyses at MRO centers and fleet-wide analytics.

• Blockchain technology is used to ensure the integrity and security of the data across the system. It provides a secure, immutable ledger for recording and sharing information about aircraft maintenance and operations, which helps in maintaining transparency, traceability, and compliance.

The integration model describes how each technology affects the different stages of data processing and decision-making:

• $T_{IoT}\left(x\right)$—function representing the enhancement or modification of data xx by IoT capabilities, primarily data collection and initial data transmission.

• $T_{AI}\left(y\right)$—function representing the processing and analysis enhancements brought by AI technologies to data yy.

• $T_{BC}\left(z\right)$—function representing the security and integrity enhancements applied to data zz through blockchain technology.

• The IoT impact on data collection and transmission $x_{1t}^{\prime }=T_{IoT}\left(x_{1t}\right)$ represents IoT-enhanced data collection at time $t$. IoT devices ensure real-time, accurate data collection and secure transmission to processing units.

• The AI impact on data processing and analysis $y_{2t}^{\prime }=T_{AI}\left(y_{2t}\right)$, $z_{3t}^{\prime }=T_{AI}\left(z_{3t}\right)$, and $w_{4t}^{\prime }=T_{AI}\left(w_{4t}\right)$ analyze and process the data at various stages, from local preprocessing on the aircraft $y_{2t}^{\prime }$, through integrative analysis at MRO centers $z_{3t}^{\prime }$, to comprehensive fleet-wide analytics $w_{4t}^{\prime }$.

• The blockchain impact on data integrity and security $z_{3t}^{″}=T_{BC}\left(z_{3t}^{\prime }\right)$ and $w_{4t}^{″}=T_{BC}\left(w_{4t}^{\prime }\right)$ ensures that the data processed and the insights gained remain tamper-proof and traceable, enhancing security, particularly in data integration and fleet-wide decision-making stages.

• The highest level of data analysis and decision-making, which incorporates all technologies, can be represented as $w_{4t}^{″}=g\left\{T_{IoT}\left(x_{1t}\right),T_{AI}\left[T_{BC}\left(z_{3t}\right)\right], T_{BC}\left[\left(T_{AI}\left(w_{4t}\right)\right)\right]\right\}$ where $g$ is a function that models the combined influence of IoT, AI, and blockchain technologies on the overall health management decisions. This function takes into account how real-time data collection, advanced analytics, and secure data handling converge to provide optimal maintenance strategies and predictions.

Example Application

Consider the process of predicting engine maintenance needs:

• IoT devices collect engine performance data.

• AI analyzes trends and predicts potential issues based on both current and historical data.

• Blockchain secures the predictions and maintenance records, ensuring that all subsequent actions are verifiable and compliant with regulatory standards.

This model facilitates a robust framework for integrating cutting-edge technologies into aircraft health management, enhancing each layer’s effectiveness from data collection to strategic decision-making.

To provide a comprehensive view of how the technologies (IoT, AI, Blockchain) enhance the functionalities of the aircraft health management system, Table 9 maps out the specific actions performed by each technology across different health management functions.

This table outlines a multifaceted approach where each technology plays a crucial role in enhancing the functionality and reliability of the aircraft health management system.

To offer a clear view of how the combined use of the IoT, AI, and Blockchain supports various decision-making actions within the aircraft health management system, Table 10 outlines the specific contributions of each technology to different decision-making scenarios.

This table demonstrates how the IoT, AI, and Blockchain work synergistically to enhance decision-making capabilities in aircraft health management, improving safety, efficiency, and compliance across all operations.

3.14. Digital Twin Technology Potential for HMGT

Digital twin technology is a game-changer in the realm of aviation health management, offering a multitude of benefits that can enhance the efficiency, safety, and reliability of aircraft operations.

Digital twins enable predictive maintenance by creating virtual models of aircraft that are continuously updated with real-time data from onboard sensors. These models can simulate the aircraft’s performance under various conditions and predict when parts might fail or require service. By anticipating maintenance needs, airlines can plan maintenance activities proactively, minimizing downtime and extending the life of aircraft components.

With a digital twin, every aspect of the aircraft’s health can be monitored in real time, providing an immediate snapshot of its status. This capability allows for in-depth simulations of potential scenarios, helping to identify potential issues before they occur and testing solutions virtually before implementing them in the real world.

Digital twins facilitate more informed decision-making by providing a comprehensive and accurate representation of the aircraft’s current and predicted future states. Maintenance teams can use this information to make better decisions about when and how to perform maintenance, and flight crews can use it for operational decisions.

Digital twin technology allows for the customization of maintenance and operational strategies based on the specific conditions and usage patterns of each aircraft. This customization leads to more efficient operations and can help optimize fuel consumption, flight paths, and other operational parameters to improve efficiency and reduce costs.

From design and manufacturing to operation and decommissioning, digital twins support the entire lifecycle management of an aircraft. Manufacturers can use them to improve design and production, while airlines can use them throughout the operational life of the aircraft to monitor its health and performance, ultimately leading to more sustainable lifecycle management.

Digital twins can be used to create realistic training environments for pilots and maintenance crews, allowing them to gain experience with different types of scenarios in a risk-free virtual setting. This training can improve safety and ensure that crews are prepared for a wide range of situations.

Digital twins can be integrated with other technologies such as AI, the IoT, and blockchain to create a holistic aviation health management system. AI can be used to analyze data from the digital twin to make predictions and recommendations, IoT devices can provide the real-time data needed to keep the digital twin updated, and blockchain can secure the data and ensure that all records are immutable and traceable.

While digital twins are a powerful tool, there are challenges to their implementation. These include the need for high-fidelity data, computational resources to run simulations, and integration with existing systems. Additionally, there are considerations around data security and privacy, particularly when handling sensitive operational data.

3.15. Aviation Technical Support as a Service

The transformation towards a distributed economy model, with Aviation Technical Support as a Service (ATSaaS), represents a significant evolution of the health management concept. In this model, specialized organizations take over various functions traditionally managed by airline MRO operations.

The Aviation Technical Support as a Service concept represents a paradigm shift in the way aircraft health management is approached, offering a distributed and specialized model that challenges traditional MRO operations. This innovative concept is driven by the increasing complexity of aircraft systems, the need for specialized expertise, and the demand for cost-effective and efficient maintenance solutions.

At the core of the ATSaaS model is the recognition that aircraft health management encompasses a wide range of specialized functions, each requiring a unique set of skills and resources. Instead of airline MRO operations attempting to manage all these functions in-house, the ATSaaS approach advocates for the outsourcing of specific tasks to specialized organizations that can provide superior expertise and economies of scale (Figure 9).

The ATSaaS model reimagines aircraft health management by creating a distributed, specialized service network that leverages cloud technologies to offer airlines a more agile, expert-driven maintenance support structure.

The ATSaaS center, staffed by specialists, provides dedicated support and advanced technical services, leveraging AI and big data to monitor aircraft health. Data from each participating airline’s fleet are streamed to a cloud platform, enabling real-time health management and predictive maintenance planning. The system facilitates a seamless exchange of data and actionable insights between the ATSaaS hub and MRO facilities, ensuring that physical maintenance activities are data-driven and strategically timed.

The ATSaaS model is scalable and capable of servicing multiple airlines simultaneously, pooling resources and expertise to offer cost-efficient, high-quality technical support. Continuous data flow and real-time analytics enable a dynamic response to maintenance needs, with MRO actions feeding back into the system to enhance predictive accuracy.

One of the key advantages of the ATSaaS model is the ability to leverage the expertise of specialized organizations that focus solely on specific aspects of aircraft HMGT. These organizations can invest heavily in research, development, and cutting-edge technologies specific to their area of specialization, ensuring that they remain at the forefront of their respective fields. This level of specialization is difficult to achieve within a traditional airline MRO operation, which must balance various priorities and responsibilities.

Another significant benefit of the ATSaaS model is the potential for cost savings and improved efficiency. By outsourcing specific functions to specialized organizations, airlines can avoid the need to maintain extensive in-house capabilities and resources for every aspect of aircraft health management. This can lead to reduced overhead costs, optimized resource allocation, and the ability to scale services up or down as needed, depending on the airline’s operational requirements.

The ATSaaS model also enables greater flexibility and agility in responding to changing market conditions and technological advancements. Specialized organizations can adapt and evolve their services more rapidly, incorporating the latest innovations and best practices in their respective domains. This agility can be challenging for traditional airline MRO operations, which may be constrained by legacy systems, processes, and organizational structures.

Moreover, the ATSaaS model fosters collaboration and knowledge sharing among specialized organizations, airlines, and other stakeholders within the aviation ecosystem. By leveraging a distributed network of expertise, the industry can collectively advance aircraft health management practices, share insights, and drive continuous improvement.

However, the successful implementation of the ATSaaS model requires careful consideration and effective collaboration between airlines and specialized service providers. The standardization of data formats, protocols, and interfaces is essential to ensure seamless integration and data exchange among various stakeholders. Additionally, robust cybersecurity measures and data protection protocols must be in place to safeguard sensitive aircraft health data and maintain operational integrity.

The Aviation Technical Support as a Service concept represents a significant evolution in aircraft health management, offering a distributed and specialized model that leverages the expertise of specialized organizations. By outsourcing specific functions to these organizations, airlines can benefit from improved expertise, cost savings, flexibility, and agility. While the implementation of the ATSaaS model presents challenges, it holds the potential to drive innovation, collaboration, and continuous improvement in the aircraft maintenance and health management domain.

The next evolution in aviation health management moves towards a more interconnected, intelligent, and collaborative system. This system not only optimizes maintenance and improves aircraft availability but also leverages the collective expertise and data for industry-wide advancements, driving innovation and maintaining high safety standards.

Figure 10 illustrates the progressive evolution of ATSaaS, showcasing a tiered approach to aviation health management.

A three-tiered cloud-based architecture of HMGT with ATSaaS approach can be proposed.

3.15.1. Tier 1: Aircraft

At the foundation of this model lies the individual aircraft, equipped with a constellation of IoT devices and sensors. Each aircraft is a node in a broader network, generating a continuous stream of data on its health and performance.

Sensors collect data on various parameters, from engine health to structural integrity.

Initial data analysis can occur onboard, utilizing edge computing to provide real-time diagnostics and to identify potential issues.

The aircraft communicates this data to the next tier for more extensive analysis and integration.

3.15.2. Tier 2: Airline Fleet Management

Each airline operates as a centralized node for its respective fleet, aggregating data from individual aircraft.

At this level, data from the entire fleet are analyzed to identify trends, optimize maintenance schedules, and improve overall operational efficiency.

Machine learning models analyze aggregated data to predict maintenance needs across the fleet.

Decision-support systems help manage and allocate resources effectively, ensuring that maintenance activities are timely and disruptions are minimized.

3.15.3. Tier 3: ATSaaS

At the apex of the model is the ATSaaS hub, a sophisticated operations center that interfaces with multiple airlines.

Specialized engineers and data scientists provide advanced analytical services, drawing on a deep understanding of aircraft systems and predictive analytics.

By servicing multiple airlines, the ATSaaS provider achieves economies of scale, driving down costs and enhancing the quality of support.

The centralization of data and expertise facilitates ongoing improvements in maintenance practices and aircraft design.

This tiered ATSaaS model represents a holistic approach to aircraft maintenance and health management, characterized by its layered yet integrated structure:

• The tiered approach ensures that decisions at every level are informed by the most comprehensive and current data available.

• Each tier can customize the services and analysis it provides or receives, allowing for flexibility in meeting specific needs.

• The model promotes collaboration across the industry, with data sharing between airlines, ATSaaS providers, MROs, and OEMs, ultimately leading to a more efficient and resilient aviation ecosystem.

This advanced framework reflects the next step in the transformation of maintenance and HMGT support paradigms within the industry.

4. Discussion

4.1. The Promise of HMGT: A Leap towards Proactivity and Efficiency

The previous section effectively illustrates the dual facets of implementing aircraft health management—the benefits and the complexity it introduces into the aviation maintenance ecosystem. This serves as a foundation for understanding how HMGT, while promising significant advantages, also necessitates careful consideration of the challenges and intricacies it brings along.

The primary allure of HMGT lies in its potential to transform aircraft maintenance from a reactive to a proactive discipline. By leveraging real-time data analytics, HMGT allows for the early detection of potential issues before they escalate into serious problems. This predictive capability can significantly enhance operational safety by ensuring that maintenance needs are addressed promptly and efficiently, thereby reducing the likelihood of in-flight failures.

HMGT can lead to considerable economic benefits for airlines and aircraft operators. The ability to predict maintenance needs allows for better planning and resource allocation, which in turn can lead to reduced downtime for aircraft. This is crucial in an industry where the cost implications of unscheduled maintenance and groundings can be substantial. By optimizing maintenance schedules and reducing the need for emergency repairs, HMGT can contribute to improved fleet availability and operational reliability, factors that directly impact an airline’s profitability and service quality.

However, the transition to an HMGT-driven maintenance paradigm is not without its challenges. The complexity introduced by HMGT encompasses technological, organizational, and regulatory dimensions. Technologically, the implementation of HMGT requires a robust infrastructure capable of collecting, transmitting, processing, and analyzing vast amounts of data generated by aircraft systems. This necessitates significant investment in IT systems and capabilities, including advanced analytics, machine learning algorithms, and cybersecurity measures to protect sensitive data.

Organizationally, the adoption of HMGT demands a shift in mindset from traditional maintenance practices to a data-driven approach. This can involve restructuring maintenance teams, developing new competencies among personnel, and fostering a culture of continuous learning and adaptation. The change management aspect of implementing HMGT cannot be underestimated, as it requires buy-in from all levels of an organization.

From a regulatory standpoint, the introduction of HMGT presents challenges related to compliance with existing aviation maintenance standards and practices. Regulators must balance the need to ensure safety and reliability with the desire to encourage innovation and the adoption of new technologies. Developing and updating regulatory frameworks that accommodate HMGT while maintaining rigorous safety standards is a complex but necessary process.

The benefits of HMGT, including enhanced safety, operational efficiency, and economic gains, make a compelling case for its adoption. However, navigating the complexity it introduces requires a thoughtful approach that addresses technological, organizational, and regulatory challenges. By doing so, the aviation industry can fully harness the power of HMGT to achieve a higher standard of maintenance and operational excellence. As the industry moves forward, it will be essential to continue fostering collaboration among all stakeholders to ensure that the transition to HMGT is both successful and sustainable.

4.2. Implementation of HMGT in Aviation Enterprise

Health management can be implemented at the level of a single enterprise for the operation and maintenance of aviation assets. Implementing HMGT within an individual enterprise, such as an airline, a maintenance, repair, and overhaul (MRO) provider, or an aircraft leasing company, involves developing a system that integrates with the existing operational framework to enhance efficiency, safety, and reliability.

Implementing HMGT at the enterprise level offers several benefits, including enhanced aircraft availability, reduced maintenance costs, improved safety, and compliance with regulatory standards. It allows the enterprise to proactively manage its fleet’s health, optimizing maintenance schedules based on actual aircraft condition rather than fixed intervals and making informed operational decisions based on predictive insights.

The roadmap for the successful implementation of HMGT in an aviation company can include the next steps:

Phase 1. Assessment and planning:

1.1 Needs assessment: evaluate the current state of maintenance operations, identify pain points, and define specific goals for the HMGT system (e.g., reducing unplanned maintenance, enhancing aircraft availability).

1.2 Feasibility study: conduct a technical and financial feasibility study to assess the potential benefits, costs, and ROI of implementing an HMGT system.

1.3 Stakeholder engagement: engage key stakeholders across the organization, including maintenance teams, flight operations, IT, and executive leadership, to gather input and build support for the project.

1.4 Regulatory compliance review: review relevant aviation regulations and standards to ensure the HMGT system will comply with all regulatory requirements.

Phase 2. System design and development:

2.1 Requirements specification: develop detailed specifications for the HMGT system, including data collection needs, predictive analytics capabilities, user interface requirements, and integration with existing systems.

2.2 Technology selection: select appropriate technologies and platforms for data collection, storage, analytics, and user interfaces, considering scalability, security, and interoperability.

2.3 System architecture design: design the system architecture for the HMGT system, outlining the data flow, processing logic, analytics models, and integration points with existing systems.

2.4 Development and testing: develop the HMGT system according to the specifications, followed by rigorous testing to ensure accuracy, reliability, and compliance with regulatory requirements.

Phase 3. Implementation and integration:

3.1 Pilot implementation: implement the HMGT system on a small scale or in a pilot program to test its effectiveness in a controlled environment and gather feedback for improvement.

3.2 Integration with existing systems: integrate the HMGT system with existing IT infrastructure, maintenance management systems, flight operations systems, and other relevant platforms.

3.3 Data collection and model training: begin collecting data from aircraft sensors, maintenance logs, and other sources, and use these data to train and refine the predictive analytics models within the HMGT system.

3.4 User training and change management: conduct comprehensive training sessions for all users of the HMGT system, including maintenance technicians, operations managers, and IT staff, and implement change management strategies to facilitate adoption.

Phase 4. Operational deployment:

4.1 Full-scale deployment: roll out the HMGT system across the entire fleet and all relevant operational areas, monitoring closely for any issues or challenges that arise.

4.2 Performance monitoring and optimization: continuously monitor the performance of the HMGT system, using data analytics to assess its impact on maintenance operations, aircraft availability, and operational efficiency.

4.3 Feedback loop and continuous improvement: implement a feedback loop to collect input from users and performance data and use this information to continuously improve the HM system, adjusting predictive models, workflows, and user interfaces as needed.

Phase 5. Review and expansion:

5.1 Impact assessment: conduct a comprehensive review of the HMGT system’s performance, evaluating its impact on maintenance costs, aircraft availability, safety, and regulatory compliance.

5.2 Lessons learned: document lessons learned during the implementation and operational phases and share these insights across the organization to drive continuous improvement.

5.3 Expansion and scaling: based on the success of the HMGT system, consider expanding its capabilities or applying the system to additional areas of operation, such as flight operations optimization or supply chain management.

5.4 Ongoing development: maintain an ongoing development plan for the HMGT system to incorporate new technologies, data sources, and analytical methods, ensuring the system remains at the cutting edge of health management innovation.

Implementing a health management system is a complex, multi-phase project that requires careful planning, stakeholder engagement, and continuous improvement. By following this roadmap, an aviation company can successfully implement an HMGT system that enhances maintenance operations, improves safety, and ensures regulatory compliance.

4.3. Limitations of IoT and AI Implementation in Aviation Health Management Systems

The study of integrating the IoT and AI into aircraft health monitoring and safety is a rapidly evolving field, facing several limitations and challenges that must be addressed to fully realize its potential. These limitations encompass technological, regulatory, and practical aspects, each influencing the pace and direction of advancements in this area.

There are some key limitations of IoT and AI implementation in aviation health management systems:

• One of the most significant concerns with the IoT and AI in aviation is the security and privacy of the data collected. Aircraft generate vast amounts of data, and ensuring these data are securely transmitted, stored, and processed is paramount. There is also the challenge of protecting these data against cyberattacks, which could potentially compromise the safety of flight operations.

• Integrating new IoT and AI technologies into existing aviation systems poses significant challenges. These include compatibility with legacy systems, ensuring interoperability among diverse systems and components, and standardizing protocols and data formats. The aviation industry operates globally under stringent regulatory standards, making the harmonization of new technologies with existing frameworks a complex and time-consuming process.

• The sheer volume of data generated by IoT devices on aircraft is overwhelming. As noted in recent discussions by professional organizations such as Capgemini and AWS, by mid-life, a large passenger aircraft can amass hundreds of thousands of documents, underscoring the critical role of sophisticated data management strategies for lifecycle optimization platforms with additional capabilities for image recognition and analysis of aviation data [76]. By applying complex algorithms and AI, it is possible to predict potential system failures before they occur, thus allowing for timely interventions that can prevent costly downtimes and enhance aircraft performance and safety [77,78].

• The aviation industry is heavily regulated to ensure the utmost safety and reliability of flight operations. Introducing new technologies like the IoT and AI into aircraft health monitoring systems involves navigating a maze of regulatory approvals and certification processes. These processes can be lengthy and costly, potentially hindering rapid innovation and implementation.

• While AI has made significant strides in predictive analytics, there are limitations to the reliability and accuracy of its predictions, particularly in complex environments like aviation. Ensuring that AI algorithms can accurately predict potential issues without generating false positives or missing critical signals is an ongoing challenge. This requires continuous refinement of AI models and validation against real-world scenarios.

• The integration of advanced technologies into aviation maintenance and operations also brings human factors into play. There is a need for extensive training for maintenance personnel and flight crews to effectively interact with and interpret the information provided by IoT and AI systems. Balancing the reliance on automated systems with the need for human oversight and decision-making is a nuanced challenge that requires careful consideration.

• The cost of implementing and maintaining advanced IoT and AI systems can be prohibitive for some operators, particularly smaller airlines or those in developing regions. Additionally, the infrastructure required to support these technologies, both on aircraft and on the ground, requires significant investment, which can be a limitation for widespread adoption.

Despite the promise IoT and AI hold for transforming aircraft health monitoring and enhancing flight safety, these limitations highlight the complexities and challenges that need to be addressed. As the industry continues to advance, a collaborative approach involving regulators, manufacturers, airlines, and technology providers will be crucial in overcoming these hurdles, paving the way for safer, more efficient air travel in the future.

4.4. Cross-Sector Technological Transformations: Insights for Aviation

The advent of IoT and AI technologies has transformed various sectors by enhancing efficiency, improving safety, and reducing operational costs. By examining these industries, we can extrapolate potential impacts, benefits, and challenges these technologies may pose in aviation health monitoring systems.

In healthcare, AI and the IoT have revolutionized patient monitoring and diagnostics. AI algorithms analyze vast amounts of medical data to predict patient health outcomes, facilitating early interventions that can save lives and reduce treatment costs [79].

For example, AI-driven systems in hospitals monitor patient vitals in real time, predicting acute events such as heart attacks before they occur [80], which parallels the predictive maintenance needs in aviation for anticipating mechanical failures before they lead to system breakdowns. Just as AI predicts critical health events from medical data [81], similar algorithms can predict aircraft component failures, enabling proactive maintenance strategies that prevent downtime and enhance safety. Healthcare’s integration of diverse data types—from patient records to real-time monitoring—highlights the importance of a unified data architecture [82], which is equally critical in aviation for a holistic view of aircraft health.

The automotive industry has long embraced the IoT and AI to streamline production lines and implement predictive maintenance of equipment. IoT sensors track parts and assemblies throughout the manufacturing process, ensuring precise quality control and inventory management [83]. AI models predict equipment failures, scheduling maintenance that minimizes production disruption, closely mirroring the logistical needs of aviation maintenance. The automotive industry’s use of the IoT for tracking parts [84] can be adapted to aviation for enhancing parts logistics, ensuring the right components are available for maintenance without excess inventory. The AI-driven automation of production processes in automotive manufacturing [85] provides insights into how similar technologies could optimize aircraft assembly and maintenance workflows, reducing costs and increasing throughput.

While the healthcare and automotive sectors differ significantly from aviation, the core applications of the IoT and AI provide valuable lessons:

• Both sectors are heavily regulated with stringent safety standards, much like aviation. The successful integration of the IoT and AI in these regulated environments offers a blueprint for aviation to follow, particularly in handling regulatory compliance and ensuring data security.

• The scalability of AI solutions in healthcare and automotive sectors demonstrates the potential for similar technologies to adapt to the complex and dynamic nature of aviation operations.

• Both sectors have documented reductions in operational costs and improved efficiencies as a result of IoT and AI integration.

Detailed cost–benefit analyses from these sectors can help aviation stakeholders make informed decisions regarding technology investments while keeping the strategic approach needed to tailor these technologies to aviation’s unique requirements.

4.5. Data Privacy, Security, and Cybersecurity Challenges in HMGT Systems

The integration of IoT and AI technologies into aviation health monitoring systems presents new opportunities for enhancing safety and operational efficiency. However, the adoption of these technologies also introduces significant challenges related to data privacy, security, and cybersecurity.

On the one hand, these technologies themselves have protective mechanisms for solving these problems. IoT devices and AI systems use sophisticated encryption techniques to protect data transmitted across networks. This encryption ensures that data, from sensor outputs to maintenance logs, remain confidential and tamper-proof during transmission. AI algorithms are adept at monitoring network traffic for unusual patterns that may indicate a security breach or attempted intrusion. By continuously learning from network data, these AI systems can quickly detect and alert human operators to potential threats, often before any harm is done.

On the other hand, there are potential risks and vulnerabilities in IoT and AI applications. Despite advances in encryption, data transmitted from IoT devices can be intercepted by unauthorized entities if not properly secured. This risk is particularly acute when devices use unsecured communication channels or when encryption protocols are outdated. AI systems themselves can be targets for cyberattacks. Adversarial attacks, where malicious inputs are designed to deceive AI algorithms, can lead to incorrect system outputs or even system failure, posing serious risks to aircraft safety. IoT devices are inherently connected, making them vulnerable to cascading failures if one device is compromised. The interconnected nature of these devices means that a single point of failure can have widespread consequences.

As the number of connected IoT devices within aviation systems grows, scaling security measures appropriately becomes a significant challenge. Ensuring consistent security across all devices requires substantial resources and continuous oversight. The absence of standardized security protocols across the IoT industry complicates the implementation of effective security measures. Without industry-wide standards, each device might require unique handling, increasing the complexity of securing aviation systems.

Integrating new security technologies with existing aviation infrastructures is both technically challenging and costly. Older systems may not support the latest security measures, requiring either costly upgrades or custom solutions that may not offer the same level of protection. This involves not only implementing advanced technological solutions but also continuously evaluating and adapting these solutions to address emerging threats. Collaboration across the aviation industry to develop standardized security protocols will be crucial in mitigating risks effectively. As these technologies continue to evolve, so too must the strategies to protect the data they handle, ensuring that the benefits of the IoT and AI are realized without compromising the safety and privacy of aviation operations.

4.6. Future Research Activities in Study Domain

The integration of the IoT and AI into aviation health management systems presents vast potential for revolutionizing aircraft maintenance, safety, and operational efficiency. However, realizing this potential requires targeted research efforts to address existing gaps and harness these technologies fully. There are several directions for future research in this field, as follows:

• IoT device integration and standardization.

Research focus: developing standards for the integration of IoT devices in aircraft systems to ensure interoperability, reliability, and data accuracy across different platforms and manufacturers.

Objective: create a unified framework for IoT device communication, data exchange protocols, and security standards that can be adopted industry-wide.

• 2.. AI algorithms for predictive maintenance.

Research focus: designing and refining AI algorithms specifically tailored for predictive maintenance in aviation, capable of processing vast amounts of data from IoT sensors to accurately predict potential failures before they occur.

Objective: enhance the predictive capabilities of HMGT, reducing unplanned downtime and extending the lifespan of aircraft components.

• 3.. Real-time data processing and analytics.

Research focus: investigating efficient methods for the real-time processing and analysis of the massive data streams generated by IoT devices in aircraft, leveraging edge computing and AI techniques.

Objective: enable immediate, actionable insights directly from data collected in flight, improving decision-making processes for maintenance and operational efficiency.

• 4.. Cybersecurity for aviation IoT and AI systems.

Research focus: developing robust cybersecurity measures to protect the integrity and confidentiality of data within HMGT, addressing the unique challenges posed by the integration of the IoT and AI in aviation.

Objective: ensure that HMGT is secure from cyber threats, preserving the safety and reliability of aviation operations.

• 5.. Regulatory compliance and certification.

Research focus: exploring the implications of IoT and AI integration on regulatory compliance and certification processes for aircraft maintenance and operations.

Objective: provide guidelines and frameworks that align with aviation regulations, facilitating the adoption of IoT and AI technologies in AHMS while ensuring safety and compliance.

• 6.. Human–AI interaction in maintenance operations.

Research focus: studying the dynamics of human–AI interaction within the context of aviation maintenance, focusing on user interfaces, decision-support systems, and the impact of AI recommendations on maintenance practices.

Objective: optimize the collaboration between human operators and AI systems in HMGT, enhancing the efficiency and effectiveness of maintenance activities.

• 7.. Scalability and adaptability of AI models.

Research focus: investigating methods to ensure the scalability and adaptability of AI models in HMGT, allowing them to evolve with changing aircraft technologies, operational practices, and emerging data patterns.

Objective: develop AI models that can be easily updated and scaled, maintaining their accuracy and relevance over time.

• 8.. Impact of AI and the IoT on aviation sustainability.

Research focus: assessing how the implementation of AI and the IoT in AHMS can contribute to aviation sustainability efforts, including fuel efficiency, emissions reduction, and resource optimization.

Objective: identify strategies in which HMGT can support the aviation industry’s sustainability goals through more efficient operations and maintenance practices.

These research directions aim to address the technical, operational, and regulatory challenges associated with implementing the IoT and AI in aviation health management systems. By focusing on these areas, the aviation industry can move closer to realizing the full potential of these technologies, enhancing safety, efficiency, and sustainability.

5. Conclusions

The convergence of the IoT, AI and cloud computing within the ecosystem of aviation maintenance signifies a revolutionary leap towards transforming aircraft health monitoring into a comprehensive health management paradigm. This paper has meticulously examined the symbiotic relationship between these technologies and their collective potential to redefine the maintenance strategies employed in the aviation industry. Through an extensive review, the significant benefits of this integration were identified, including enhanced predictive maintenance capabilities, improved operational efficiency, and elevated safety standards, all of which are pivotal for the sustainable growth and reliability of air travel.

However, the implementation of such advanced technologies is not devoid of challenges. Security concerns, the complexity of integrating new systems with legacy infrastructures, regulatory hurdles, and the need for substantial investment in both technology and training represent significant barriers to adoption. Despite these challenges, the potential benefits far outweigh the obstacles, provided that a strategic, collaborative approach is employed.

Future research within this domain is essential to further refine these technologies, making them more accessible and effective. Investigations into the development of more robust AI models for predictive maintenance, enhanced cybersecurity measures for IoT devices, and more efficient data management and analysis frameworks are critical. Moreover, there is a pressing need for the establishment of universal standards and protocols to facilitate the seamless integration of these technologies across the global aviation industry.

The integration of the IoT, cloud computing, and artificial intelligence heralds a new era in aviation maintenance, shifting the focus from mere health monitoring to comprehensive health management. This transition is not only a testament to the advancements in technology but also a crucial step towards ensuring the utmost safety, reliability, and efficiency of air travel. It is imperative for all stakeholders in the aviation industry to collaboratively embrace these changes, overcoming the inherent challenges and driving innovation to new heights.

Author Contributions

Conceptualization, I.K.; methodology, I.K.; software, V.P.; validation, I.K. and V.P.; formal analysis, I.K.; investigation, I.K.; resources, V.P.; data curation, V.P.; writing—original draft preparation, I.K.; writing—review and editing, I.K. and V.P.; visualization, I.K.; supervision, I.K.; project administration, I.K.; funding acquisition, V.P. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding author.

Conflicts of Interest

Author Vladimir Perekrestov was employed by the company Sky Net Technics. The remaining authors declare that the re-search was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Tables

Figure 1. Aviation health management framework.

Figure 2. Analytics of aviation health management.

Figure 3. Architecture of aviation health management system.

Figure 4. General taxonomy of aviation sensors.

Figure 5. Aircraft with various traditional aircraft sensors.

Figure 6. IoT-based health monitoring system.

Figure 7. Cloud-based integral health management system.

Figure 8. Ecosystem of cloud-based aviation HMGT.

Figure 9. Ecosystem of HMGT based on ATSaaS concept.

Figure 10. Cloud-based architecture of HMGT with ATSaaS approach.

Table 1

Differences between HM and HMGT in aviation.

Aspect	Health Monitoring	Health Management
Definition	The technique of monitoring the output of one or more condition indicators during aircraft operation, to diagnose faulty states and predict future degradation.	Using aircraft- and fleet-generated data to promptly identify maintenance needs and trigger effective and efficient maintenance actions.
Process stages	Sense, acquire, transfer, and analyze (SATA).	Sense, acquire, transfer, analyze, and act (SATAA).
Objective	To observe and report on the health status of aircraft systems and components, often in real time.	To use observed data to make informed decisions on maintenance actions that prevent failures and optimize aircraft availability.
Maintenance approach	More reactive, identifying potential issues before they lead to failure but not necessarily prescribing action.	Proactive and prescriptive, suggesting specific maintenance actions based on data analysis to prevent potential failures.
Involvement in decision-making	Limited to providing data on and insights into the health of the aircraft’s systems and components.	Direct involvement in the decision-making process for maintenance, suggesting when and what type of action should be taken.
Technology utilization	Primarily focused on data collection and analysis to identify potential issues.	Beyond data collection and analysis, it includes decision-making algorithms and automation to initiate maintenance actions.
Outcome	Identification of potential issues and monitoring of system performance over time.	Optimized maintenance schedules, reduced unscheduled maintenance, and improved aircraft availability and reliability.
Benefits	Enhanced monitoring and early detection of potential issues, improving the reliability of aircraft components and systems.	Reduced operational disruptions and maintenance costs and improved safety by preventing failures before they occur.
Examples	Engine vibration monitoring, hydraulic fluid level monitoring, and temperature monitoring of critical components.	Predictive maintenance scheduling based on engine wear patterns, optimization of parts replacement schedules, and targeted inspections.

Table 2

The role of the IoT and AI across the analytics spectrum of HMGT.

Type of Analytics	Role of IoT	Role of AI
Descriptive analytics	IoT devices collect and transmit real-time data from various aircraft components and systems, providing a comprehensive dataset for analysis.	AI processes this vast amount of data to identify patterns, trends, and anomalies, offering insights into past and current aircraft performance.
Diagnostic analytics	IoT sensors monitor specific parameters that can indicate the health status of components, feeding data into diagnostic processes.	AI algorithms analyze the data to diagnose the root causes of issues by identifying correlations and patterns that may not be obvious to human analysts.
Predictive analytics	IoT systems continuously gather data that can be used to forecast future conditions and performance of aircraft components.	AI leverages machine learning models to predict future failures or maintenance needs based on historical data, improving accuracy over time with more data.
Prescriptive analytics	IoT provides the real-time operational data required for making informed decisions on the best courses of action.	AI uses optimization and simulation algorithms to prescribe specific actions to prevent failures or optimize performance, considering multiple variables.

Table 3

AI types and their applicability to analytics of HMGT.

Type ofAnalytics	Types of AI Used
Descriptive analytics	Data visualization tools: AI-enhanced visualization tools help in presenting data in an easily understandable format, identifying patterns and trends.Clustering algorithms: used to group similar data points, revealing patterns or anomalies in large datasets.
Diagnostic analytics	Classification algorithms: AI models that categorize data into predefined classes, useful for identifying the type or category of a problem.Root Cause Analysis (RCA) algorithms: AI-driven RCA can automate the identification of root causes behind observed anomalies.
Predictive analytics	Machine learning models: supervised learning algorithms, such as regression models, decision trees, and neural networks, can forecast future events based on historical data.Time-series analysis: AI techniques for analyzing time-series data to predict future data points.
Prescriptive analytics	Optimization algorithms: AI methods that find the best solution from a set of possible options, considering constraints and objectives.Reinforcement learning: AI learns to make decisions by performing actions in an environment to achieve a goal, optimizing decision-making over time.

Table 4

Service delivery model for HM and HMGT.

ServiceComponent	Aviation HealthMonitoring	Aviation Health Management
Data collection	- IoT devices and sensors collect real-time data from aircraft systems. - Flight data monitoring systems.	- Utilizes the same IoT and sensor infrastructure as HM for data collection, with added emphasis on integration across different data sources for comprehensive analysis.
Datatransmission	- Data are transmitted in real time or near-real time via satellite or ground-based communication networks.	- Like HM but with enhanced security protocols for data integrity and confidentiality during transmission.
Data processing	- Basic processing to clean and structure data for analysis.	- Advanced processing techniques, including data integration from multiple sources and contextual analysis.
Analysis	- Descriptive analytics to understand current and historical aircraft performance. - Diagnostic analytics to identify issues.	- Predictive analytics to forecast potential failures or maintenance needs. - Prescriptive analytics to recommend specific maintenance actions.
Actionableinsights	- Alerts and notifications about current health status and potential issues.	- Detailed maintenance recommendations, including optimal timing and required actions to prevent failures.
Decisionsupport	- Provides data for manual review and decision-making by maintenance personnel.	- AI and decision-support systems offer automated recommendations, optimizing maintenance schedules and operational planning.
Implementation and feedback	- Implementation of maintenance actions based on manual planning and decision-making.	- Automated scheduling and implementation of maintenance actions. - Continuous feedback loop for system improvement based on outcomes and performance data.
Technologies used	- IoT sensors, data loggers, and basic data analytics tools.	- Advanced AI and machine learning models, optimization algorithms, and integrated data platforms.
Outcome	- Improved situational awareness and early detection of potential issues.	- Optimized aircraft availability, reduced maintenance costs, enhanced safety, and proactive maintenance strategies.

Table 5

Key services provided by health management.

Service	Description
Predictivemaintenance	Utilizes historical and real-time data to predict when components might fail or require maintenance, allowing for maintenance to be scheduled at the most opportune times, thereby minimizing downtime and avoiding unexpected failures.
Condition-based monitoring	Continuously monitors the condition of aircraft components to determine their health and functionality, ensuring that maintenance and repairs are performed based on actual need rather than on a fixed schedule, improving efficiency and extending the lifespan of parts.
Fault diagnosis	Employs diagnostic algorithms to identify the root causes of issues, reducing troubleshooting time and ensuring that maintenance efforts are directly targeted quickly and accurately at solving the underlying problems.
Optimization of maintenance schedules	Uses advanced analytics to optimize maintenance schedules, balancing operational demands with maintenance needs to ensure aircraft are available when needed while maintaining high safety standards.
Lifecyclemanagement	Offers comprehensive management of aircraft components’ lifecycle, from installation through to repair and replacement, ensuring optimal performance throughout their service life and informing procurement and inventory decisions.
Operationaldecision support	Provides decision-makers with actionable insights and recommendations to optimize flight operations, maintenance planning, and resource allocation, enhancing operational efficiency and effectiveness.
Safety analysis and risk management	Analyzes data to identify potential safety risks and implements predictive models to mitigate these risks before they impact operations, enhancing overall safety.
Regulatorycompliancemonitoring	Ensures that aircraft maintenance and operations comply with all relevant regulations and standards, leveraging data analytics to streamline compliance processes and reduce the risk of non-compliance penalties.

Table 6

Sources of information for services provided by health management.

Service	Sources of Information
Predictive maintenance	- Flight data monitoring systems: capture performance data during flights. - Maintenance records: historical maintenance actions and outcomes. - Sensor data: real-time health data from aircraft components. - Environmental data: conditions that might affect component wear and tear, like weather.
Condition-basedmonitoring	- IoT sensors: collect real-time data on various parameters such as temperature, pressure, and vibration. - Aircraft health monitoring systems: system-specific health data. - Operational data: usage patterns, flight hours, cycles, etc.
Fault diagnosis	- Diagnostic tools: software tools that analyze error codes and system alerts. - Flight crew reports: manual observations and issues reported by the flight crew. - Maintenance logs: detailed records of past inspections, issues, and fixes.
Optimization ofmaintenance schedules	- Historical maintenance data: trends and patterns from past maintenance activities. - Operational schedules: planned flight operations and aircraft utilization data. - Component lifespan data: manufacturer data on expected component lifecycles. - Regulatory requirements: mandatory maintenance intervals.
Lifecycle management	- Component tracking systems: detailed histories of individual parts. - Warranty information: data on part warranties and service life. - Supplier and manufacturer databases: specifications and advisories on components. - Cost records: historical data on repair vs. replacement costs.
Operational decision support	- Flight operations data: schedules, flight paths, and efficiency metrics. - Crew availability: staffing and crew qualification records.
Safety analysis and risk management	- Incident and accident reports: analyses of past incidents and near-misses. - Safety audit results: findings from internal and external safety audits. - Regulatory safety directives: advisories and mandates from aviation authorities. - Operational data analysis: trends and anomalies that indicate risks.
Regulatory compliance monitoring	- Regulatory databases: up-to-date requirements and changes in aviation regulations. - Compliance checklists and tools: software that matches operational data against regulatory requirements. - Audit reports: documentation from previous regulatory inspections and audits.

Table 7

HMGT functionality across different levels in dynamic and stationary activities.

Level	Dynamic Functionality (In Flight)	Stationary Functionality (On Ground)
Airplane	- Real-time monitoring of aircraft systems and components. - Immediate fault detection and preliminary diagnostics. - Predictive alerts for maintenance needs during the flight. - Emergency handling and recommendations.	- Detailed system diagnostics and data download for in-depth analysis. - Maintenance activities based on predictive alerts issued in flight. - Software updates and system calibrations. - Data synchronization with ground systems for historical analysis.
MRO	- Receiving real-time alerts and health data from aircraft. - Preliminary planning of maintenance activities and resource allocation. - Adjusting maintenance schedules based on in-flight data. - Remote diagnostics and support for in-flight issues.	- Detailed data analysis for predictive maintenance planning. - Execution of scheduled and unscheduled maintenance tasks. - Inventory management and spare parts logistics based on predictive needs. - Review and adjustment of maintenance protocols based on historical data.
Virtual center	- Fleet-wide monitoring for safety and compliance. - Integration of in-flight data with external data sources for enhanced insights. - Real-time strategic decision support based on fleet status. - Coordination of responses to widespread issues or emergencies detected in flight.	- Strategic analysis and planning based on comprehensive datasets. - Regulatory compliance reviews and safety performance analysis. - Long-term predictive modeling and trend analysis for fleet optimization. - Development and dissemination of best practices and maintenance protocols.

Table 8

HMGT functionality during intermediate landings.

Level	Functionality during Intermediate Landings
Airplane	- Quick diagnostics and fault isolation based on in-flight data. - Implementation of minor maintenance or repairs identified through predictive alerts. - Software updates or recalibrations that require the aircraft to be stationary but can be completed quickly. - Data offloading for systems that require large bandwidth, not feasible in flight.
MRO	- Rapid deployment of maintenance teams for urgent tasks identified in flight. - Quick restocking of essential supplies or parts anticipated through predictive maintenance data. - Adjustment of maintenance schedules in real time to accommodate immediate needs without significant disruption. - Immediate analysis of downloaded data for critical issues needing swift resolution.
Virtual center	- Real-time strategic decision-making support based on data received from the aircraft’s recent flight leg. - Coordination of logistics and resources to address immediate maintenance needs during the stopover. - Updating fleet-wide predictive models based on the latest data uploads from intermediate landings. - Rapid dissemination of critical information or advisories to other aircraft in the fleet based on findings during the stopover.

Table 9

Health management functions and supported technologies.

Functionality	IoT Actions	AI Actions	Blockchain Actions
Data collection	- Collects real-time sensor data from aircraft systems. - Transmits data securely to processing units.	- Filters and preprocesses sensor data for anomalies. - Prioritizes data based on criticality.	- Records timestamps and origins of data entries to ensure traceability and integrity.
Data transmission	- Provides real-time, secure data communication. - Uses encryption for data being transmitted.	- Compresses and optimizes data for transmission. - Performs error checking and correction.	- Ensures that data transmissions are secure and verifiable across networks.
Local processing	- Facilitates initial processing like data filtering at the source.	- Analyzes data locally to detect immediate anomalies. - Supports edge computing models for faster local decision-making.	- Logs processing actions and decisions for audit purposes.
Data integration at MRO	- Aggregates data from various flights for MRO analysis.	- Integrates diverse datasets. - Uses complex algorithms to identify patterns across multiple data sources.	- Provides a secure platform for integrating data from multiple sources.
Fleet-wide analysis	- Enables collection and aggregation of fleet-wide data for analysis.	- Performs predictive analytics to forecast maintenance needs. - Applies machine learning to optimize fleet operations.	- Secures fleet-wide data analysis results. - Ensures compliance with data regulations.
Predictive maintenance	- Monitors system health to predict hardware failures. - Provides operational data for analysis.	- Uses historical data to model and predict future failures. - Optimizes maintenance schedules based on predictive models.	- Ensures that maintenance records are immutable and accessible for audit.
Maintenance scheduling	- Assists in scheduling by providing real-time operational status.	- Determines optimal maintenance times using AI optimization algorithms. - Predicts future maintenance windows.	- Records and verifies scheduled maintenance against actual actions.
Regulatory compliance	- Ensures all sensor and operational data needed for compliance are collected accurately.	- Analyzes data to ensure compliance with safety and operational standards. - Reports potential compliance issues.	- Provides a secure and immutable record of all compliance-related actions and data.
Decision support	- Provides real-time data inputs for decision-making processes.	- Analyzes real-time data to provide actionable insights and recommendations. - Supports simulations and what-if analyses.	- Secures and records decision-making processes and outcomes.

Table 10

Decision-making actions and supported technologies.

Decision-Making Actions	IoT Actions	AI Actions	Blockchain Actions
Anomaly detection	- Provides real-time sensor data indicating deviations.	- Analyzes data patterns to identify anomalies. - Alerts maintenance and operational teams to unusual readings.	- Logs and timestamps anomaly detections for audit trails.
Maintenance type recommendation	- Supplies detailed operational data to inform decision-making.	- Uses predictive models to recommend specific types of maintenance based on data analysis. - Prioritizes maintenance actions based on severity and impact.	- Records recommendations ensuring compliance and verification.
Scheduling of maintenance operations	- Offers availability status of aircraft systems.	- Optimizes maintenance schedules to minimize downtime. - Predicts the best time for maintenance to ensure high availability and operational efficiency.	- Ensures maintenance schedules are securely shared and immutable.
Resource allocation	- Monitors resource usage and availability.	- Dynamically allocates resources (e.g., parts, crew) using optimization algorithms. - Predicts future resource needs based on operational trends.	- Provides a secure record of resource allocations and adjustments.
Failure prediction	- Continuously monitors equipment to capture performance data.	- Employs machine learning algorithms to predict potential failures before they occur. - Provides risk assessment based on current and historical data.	- Secures predictions and related data for transparency and traceability.
Operational efficiency improvement	- Collects comprehensive performance data across all flights.	- Analyzes performance data to identify inefficiencies. - Recommends adjustments for optimizing operational parameters.	- Certifies that all operational adjustments are recorded and compliant with regulations.
Compliance verification	- Ensures data required for compliance are accurately collected.	- Automatically checks compliance against regulatory standards using real-time data. - Generates compliance reports and highlights discrepancies.	- Immutable logging of compliance checks and outcomes ensures accountability.
Safety enhancements	- Provides data on environmental and operational conditions.	- Analyzes safety-critical data to forecast potential safety issues. - Implements machine learning to enhance predictive safety measures.	- Creates an unalterable record of safety incidents and responses.
Cost optimization	- Tracks cost-related data from operations.	- Analyzes cost data to identify areas for potential savings. - Recommends cost-effective strategies based on AI-driven insights.	- Blockchain ensures transparency in financial transactions and cost-related decisions.

References

1. From Aircraft Health Monitoring to Aircraft Health Management; White Paper on AHM IATA: Montreal, QC, Canada, 2022.

2. Charmaz, K. Constructing Grounded Theory; SAGE: Thousand Oaks, CA, USA, 2014.

3. Birks, M.; Mills, J. Grounded Theory: A Practical Guide; 2nd ed. SAGE: Thousand Oaks, CA, USA, 2015.

4. Aldiabat, K.M.; Le Navenec, C.-L. Data Saturation: The Mysterious Step in Grounded Theory Methodology. Qual. Rep.; 2018; 23, pp. 245-261. [DOI: https://dx.doi.org/10.46743/2160-3715/2018.2994]

5. Creswell, J.W.; Poth, C.N. Qualitative Inquiry and Research Design Choosing among Five Approaches; 4th ed. SAGE: Thousand Oaks, CA, USA, 2017.

6. Snyder, H. Literature Review as a Research Methodology: An Overview and Guidelines. J. Bus. Res.; 2019; 104, pp. 333-339. [DOI: https://dx.doi.org/10.1016/j.jbusres.2019.07.039]

7. Elo, S.; Kääriäinen, M.; Kanste, O.; Pölkki, T.; Utriainen, K.; Kyngäs, H. Qualitative Content Analysis: A Focus on Trustworthiness. SAGE Open; 2014; 4, 2158244014522633. [DOI: https://dx.doi.org/10.1177/2158244014522633]

8. Kallio, H.; Pietilä, A.-M.; Johnson, M.; Kangasniemi, M. Systematic Methodological Review: Developing a Framework for a Qualitative Semi-Structured Interview Guide. J. Adv. Nurs.; 2016; 72, pp. 2954-2965. [DOI: https://dx.doi.org/10.1111/jan.13031]

9. Li, S.; Dan, M.; Yang, L. An Airplane Health Management approach for civil aviation. Proceedings of the 2011 Prognostics and System Health Managment Confernece; Shenzhen, China, 24–25 May 2011; pp. 1-4. [DOI: https://dx.doi.org/10.1109/PHM.2011.5939473]

10. Fu, S.; Avdelidis, N.P. Prognostic and Health Management of Critical Aircraft Systems and Components: An Overview. Sensors; 2023; 23, 8124. [DOI: https://dx.doi.org/10.3390/s23198124] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37836954]

11. Verhagen, W.J.C.; Santos, B.F.; Freeman, F.; van Kessel, P.; Zarouchas, D.; Loutas, T.; Yeun, R.C.K.; Heiets, I. Condition-Based Maintenance in Aviation: Challenges and Opportunities. Aerospace; 2023; 10, 762. [DOI: https://dx.doi.org/10.3390/aerospace10090762]

12. Boller, C.; Chang, F.; Fujino, Y. Encyclopedia of Structural Health Monitoring; John Wiley & Sons: Hoboken, NJ, USA, 2009.

13. Tschöke, K.; Mueller, I.; Memmolo, V.; Moix-Bonet, M.; Moll, J.; Lugovtsova, Y.; Golub, M.; Venkat, R.S.; Schubert, L. Feasibility of Model-Assisted Probability of Detection Principles for Structural Health Monitoring Systems based on Guided Waves for Fibre-Reinforced Composites. IEEE Trans. Ultrason. Ferroelectr. Freq. Control; 2021; 68, pp. 3156-3173. [DOI: https://dx.doi.org/10.1109/TUFFC.2021.3084898] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34057890]

14. Memmolo, V.; Boffa, N.D.; Maio, L.; Monaco, E.; Ricci, F. Damage Localization in Composite Structures Using a Guided Waves Based Multi-Parameter Approach. Aerospace; 2018; 5, 111. [DOI: https://dx.doi.org/10.3390/aerospace5040111]

15. Coppe, A.; Pais, M.J.; Haftka, R.T.; Kim, N.H. Using a simple crack growth model in predicting remaining useful life. J. Aircr.; 2012; 49, pp. 1965-1973. [DOI: https://dx.doi.org/10.2514/1.C031808]

16. United States Government Accountability Office. Aviation Safety: Status of FAA’s Actions to Oversee the Safety of Composite Airplanes; United States Government Accountability Office: Washington, DC, USA, 2011.

17. Abrate, S. Impact on Composite Structures; Cambridge University Press: Cambridge, UK, 1998.

18. Maio, L.; Monaco, E.; Ricci, F.; Lecce, L. Simulation of low velocity impact on composite laminates with progressive failure analysis. Compos. Struct.; 2013; 103, pp. 75-85. [DOI: https://dx.doi.org/10.1016/j.compstruct.2013.02.027]

19. Alderliesten, R.C. Introduction to Aerospace Structures and Materials; TU Delft: Delft, The Netherlands, 2018; [DOI: https://dx.doi.org/10.5074/t.2018.003]

20. U.S. Department of Transportation—Federal Aviation Administration. AC No. 20-107B, Composite Aircraft Structures; U.S. Department of Transportation—Federal Aviation Administration: Oklahoma City, OK, USA, 2009.

21. U.S. Department of Transportation—Federal Aviation Administration. AC No. 25571-1D, Damage Tolerance and Fatigue Evaluation of Structures; U.S. Department of Transportation—Federal Aviation Administration: Oklahoma City, OK, USA, 2011.

22. European Union Aviation Safety Agency (EASA). Acceptable Means of Compliance 20–29 Amendment 6. 2010; Available online: https://perma.cc/KQX3-KBZL (accessed on 20 March 2024).

23. Giurgiutiu, V. Structural Health Monitoring with Piezoelectric Wafer Active Sensors; 2nd ed. Academic Press: Cambridge, MA, USA, 2014.

24. Cottone, G.; Golliwitzer, S.; Heckenberger, U.; Srauss, D. Reliability-oriented optimization of replacement strategies for monitored composite panels for Aircraft Structures. Proceedings of the 9th International Workshop on Structural Health Monitoring (IWSHM); Stanford, CA, USA, 10–12 September 2013; Volume 2, pp. 118-125.

25. Apicella, A.; Cottone, G.; De Marchi, L.; Heckenberger, U.; Marzani, A. ROC and cost curves for SHM performance characterization in a multilevel damage classification Framework: Application to Impact Damage in Aircraft Composites Structures. Proceedings of the 10th International Workshop on Structural Health Monitoring (IWSHM); Stanford, CA, USA, 1–3 September 2015; Volume 2, pp. 118-125.

26. Ali Shan, S.A.; Solangi, Y.A.; Ikram, M. Analysis of barriers to the adoption of cleaner energy technologies in Pakistan using Modified Delphi and Fuzzy Analytical Hierarchy Process. J. Clean. Prod.; 2019; 239, pp. 1037-1050. [DOI: https://dx.doi.org/10.1016/j.jclepro.2019.07.020]

27. Pattabhiraman, S.; Gogu, C.; Kim, N.H.; Haftka, R.T.; Bes, C. Skipping unnecessary structural airframe maintenance using an on-board structural health monitoring system. J. Risk Reliab.; 2012; 226, pp. 549-560. [DOI: https://dx.doi.org/10.1177/1748006X12459157]

28. Fioriti, M.; Pavan, G.; Corpino, S.; Fusaro, R. Impacts of a prognostics and health management system on aircraft fleet operating cost during conceptual design phase by using parametric estimation. Proceedings of the 5th CEAS Air & Space Conference; Delft, The Netherlands, 7–11 September 2015; Available online: http://hdl.handle.net/11583/2627572 (accessed on 20 March 2024).

29. Buderath, M.; Adhikari, P.P. Simulation Framework and Certification Guidance for Condition Monitoring and Prognostic Health Management. Proceedings of the European Conference of the Prognostics and Health Management Society; Dresden, Germany, 3–5 July 2012.

30. Adhikari, P.P.; Buderath, M. A Framework for Aircraft Maintenance Strategy including CBM. Proceedings of the PHM Society European Conference; Bilbao, Spain, 5–8 July 2016; Volume 3, [DOI: https://dx.doi.org/10.36001/phme.2016.v3i1.1622]

31. Adhikari, P.P.; Makhecha, D.; Buderath, M. A Certifiable Approach towards Integrated Solution for Aircraft Readiness Management. Proceedings of the PHM Society European Conference; Nantes, France, 8–10 July 2014; Volume 2, [DOI: https://dx.doi.org/10.36001/phme.2014.v2i1.1485]

32. Jin, P.; Lu, F.; Huang, J.; Kong, X.; Fan, M. Life cycle gas path performance monitoring with control loop parameters uncertainty for aeroengine. Aerosp. Sci. Technol.; 2021; 115, 106775. [DOI: https://dx.doi.org/10.1016/j.ast.2021.106775]

33. Chang, X.; Huang, J.; Lu, F.; Sun, H. Gas-Path Health Estimation for an Aircraft Engine Based on a Sliding Mode Observer. Energies; 2016; 9, 598. [DOI: https://dx.doi.org/10.3390/en9080598]

34. Smart, E.; Brown, D.; Denman, J. Combining multiple classifiers to quantitatively rank the impact of abnormalities in flight data. Appl. Soft Comput.; 2012; 12, pp. 2583-2592. [DOI: https://dx.doi.org/10.1016/j.asoc.2012.03.059]

35. Melnyk, I.; Banerjee, A.; Matthews, B.; Oza, N.; Assoc Comp, M. Semi-Markov Switching Vector Autoregressive Model-Based Anomaly Detection in Aviation Systems. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD); San Francisco, CA, USA, 13–17 August 2016; pp. 1065-1074.

36. Jiang, W.; Xu, Y.; Chen, Z.; Zhang, N.; Xue, X.; Zhou, J. Measurement of health evolution tendency for aircraft engine using a data-driven method based on multi-scale series reconstruction and adaptive hybrid model. Measurement; 2022; 199, 111502. [DOI: https://dx.doi.org/10.1016/j.measurement.2022.111502]

37. Wang, H.; Jiang, W.; Deng, X.; Geng, J. A new method for fault detection of aeroengine based on isolation forest. Measurement; 2021; 185, 110064. [DOI: https://dx.doi.org/10.1016/j.measurement.2021.110064]

38. Lu, F.; Jiang, J.; Huang, J.; Qiu, X. An Iterative Reduced KPCA Hidden Markov Model for Gas Turbine Performance Fault Diagnosis. Energies; 2018; 11, 1807. [DOI: https://dx.doi.org/10.3390/en11071807]

39. Wang, H.; Wang, X.; Wang, Z. Performance assessment method of dynamic process based on SFA-GPR. J. Process Control; 2022; 111, pp. 27-34. [DOI: https://dx.doi.org/10.1016/j.jprocont.2022.01.001]

40. Cheng, C.; Liu, M.; Chen, H.; Xie, P.; Zhou, Y. Slow feature analysis-aided detection and diagnosis of incipient faults for running gear systems of high-speed trains. ISA Trans.; 2022; 125, pp. 415-425. [DOI: https://dx.doi.org/10.1016/j.isatra.2021.06.023]

41. Zhao, W.; Guo, Y.; Sun, H. Research on an Adaptive Threshold Setting Method for Aero-Engine Fault Detection Based on KDE-EWMA. J. Aerosp. Eng.; 2022; 35, 04022087. [DOI: https://dx.doi.org/10.1061/(ASCE)AS.1943-5525.0001483]

42. Ammiche, M.; Kouadri, A.; Bensmail, A. A Modified Moving Window dynamic PCA with Fuzzy Logic Filter and application to fault detection. Chemom. Intell. Lab. Syst.; 2018; 177, pp. 100-113. [DOI: https://dx.doi.org/10.1016/j.chemolab.2018.04.012]

43. Song, B.; Shi, H.; Tan, S.; Tao, Y. Serial correlated-uncorrelated concurrent space method for process monitoring. J. Process Control; 2021; 105, pp. 292-301. [DOI: https://dx.doi.org/10.1016/j.jprocont.2021.07.016]

44. Chen, Z.; Liu, Y.; Zhou, P. A comparative study of fractal dimension calculation methods for rough surface profiles. Chaos Solitons Fractals; 2018; 112, pp. 24-30. [DOI: https://dx.doi.org/10.1016/j.chaos.2018.04.027]

45. Chen, Z.; Zhang, X.; Zhou, P. Vibration fatigue life analysis of engine piping system based on multi-point random excitation. J. Propuls. Technol.; 2019; 40, pp. 1620-1627.

46. Iyaghigba, S.D.; Ali, F.; Jennions, I.K. A Review of Diagnostic Methods for Hydraulically Powered Flight Control Actuation Systems. Machines; 2023; 11, 165. [DOI: https://dx.doi.org/10.3390/machines11020165]

47. Fentaye, A.D.; Zaccaria, V.; Kyprianidis, K. Aircraft Engine Performance Monitoring and Diagnostics Based on Deep Convolutional Neural Networks. Machines; 2021; 9, 337. [DOI: https://dx.doi.org/10.3390/machines9120337]

48. Wang, Z.; Wang, Y.; Wang, X.; Yang, K.; Zhao, Y. A Novel Digital Twin Framework for Aeroengine Performance Diagnosis. Aerospace; 2023; 10, 789. [DOI: https://dx.doi.org/10.3390/aerospace10090789]

49. Yang, C.; Huang, C. Natural Language Processing (NLP) in Aviation Safety: Systematic Review of Research and Outlook into the Future. Aerospace; 2023; 10, 600. [DOI: https://dx.doi.org/10.3390/aerospace10070600]

50. Madeira, T.; Melício, R.; Valério, D.; Santos, L. Machine Learning and Natural Language Processing for Prediction of Human Factors in Aviation Incident Reports. Aerospace; 2021; 8, 47. [DOI: https://dx.doi.org/10.3390/aerospace8020047]

51. Kumar, S.G.; Corrado, S.J.; Puranik, T.G.; Mavris, D.N. Classification and Analysis of Go-Arounds in Commercial Aviation Using ADS-B Data. Aerospace; 2021; 8, 291. [DOI: https://dx.doi.org/10.3390/aerospace8100291]

52. Rose, R.L.; Puranik, T.G.; Mavris, D.N. Natural Language Processing Based Method for Clustering and Analysis of Aviation Safety Narratives. Aerospace; 2020; 7, 143. [DOI: https://dx.doi.org/10.3390/aerospace7100143]

53. Lee, H.; Madar, S.; Sairam, S.; Puranik, T.G.; Payan, A.P.; Kirby, M.; Pinon, O.J.; Mavris, D.N. Critical Parameter Identification for Safety Events in Commercial Aviation Using Machine Learning. Aerospace; 2020; 7, 73. [DOI: https://dx.doi.org/10.3390/aerospace7060073]

54. Nogueira, R.P.R.; Melicio, R.; Valério, D.; Santos, L.F.F.M. Learning Methods and Predictive Modeling to Identify Failure by Human Factors in the Aviation Industry. Appl. Sci.; 2023; 13, 4069. [DOI: https://dx.doi.org/10.3390/app13064069]

55. Tikayat Ray, A.; Bhat, A.P.; White, R.T.; Nguyen, V.M.; Pinon Fischer, O.J.; Mavris, D.N. Examining the Potential of Generative Language Models for Aviation Safety Analysis: Case Study and Insights Using the Aviation Safety Reporting System (ASRS). Aerospace; 2023; 10, 770. [DOI: https://dx.doi.org/10.3390/aerospace10090770]

56. Doğru, A.; Bouarfa, S.; Arizar, R.; Aydoğan, R. Using Convolutional Neural Networks to Automate Aircraft Maintenance Visual Inspection. Aerospace; 2020; 7, 171. [DOI: https://dx.doi.org/10.3390/aerospace7120171]

57. Abdelghany, E.S.; Farghaly, M.B.; Almalki, M.M.; Sarhan, H.H.; Essa, M.E.-S.M. Machine Learning and IoT Trends for Intelligent Prediction of Aircraft Wing Anti-Icing System Temperature. Aerospace; 2023; 10, 676. [DOI: https://dx.doi.org/10.3390/aerospace10080676]

58. Gao, Z.; Mavris, D.N. Statistics and Machine Learning in Aviation Environmental Impact Analysis: A Survey of Recent Progress. Aerospace; 2022; 9, 750. [DOI: https://dx.doi.org/10.3390/aerospace9120750]

59. Brandoli, B.; de Geus, A.R.; Souza, J.R.; Spadon, G.; Soares, A.; Rodrigues, J.F., Jr.; Komorowski, J.; Matwin, S. Aircraft Fuselage Corrosion Detection Using Artificial Intelligence. Sensors; 2021; 21, 4026. [DOI: https://dx.doi.org/10.3390/s21124026]

60. Yang, R.; Gao, Y.; Wang, H.; Ni, X. Fuzzy Neural Network PID Control Used in Individual Blade Control. Aerospace; 2023; 10, 623. [DOI: https://dx.doi.org/10.3390/aerospace10070623]

61. Wang, Z.; Zhao, Y. Data-Driven Exhaust Gas Temperature Baseline Predictions for Aeroengine Based on Machine Learning Algorithms. Aerospace; 2023; 10, 17. [DOI: https://dx.doi.org/10.3390/aerospace10010017]

62. Chen, J.; Qi, G.; Wang, K. Synergizing Machine Learning and the Aviation Sector in Lithium-Ion Battery Applications: A Review. Energies; 2023; 16, 6318. [DOI: https://dx.doi.org/10.3390/en16176318]

63. Baumann, M.; Koch, C.; Staudacher, S. Application of Neural Networks and Transfer Learning to Turbomachinery Heat Transfer. Aerospace; 2022; 9, 49. [DOI: https://dx.doi.org/10.3390/aerospace9020049]

64. Quadros, J.D.; Khan, S.A.; Aabid, A.; Alam, M.S.; Baig, M. Machine Learning Applications in Modelling and Analysis of Base Pressure in Suddenly Expanded Flows. Aerospace; 2021; 8, 318. [DOI: https://dx.doi.org/10.3390/aerospace8110318]

65. Papakonstantinou, C.; Daramouskas, I.; Lappas, V.; Moulianitis, V.C.; Kostopoulos, V. A Machine Learning Approach for Global Steering Control Moment Gyroscope Clusters. Aerospace; 2022; 9, 164. [DOI: https://dx.doi.org/10.3390/aerospace9030164]

66. ICAO. Attracting, Educating and Retaining the Next Generation; ICAO: Montreal, QC, Canada, 2017; Available online: https://www.icao.int/safety/ngap/Documentation/icao_training_report_vol7_No3.pdf (accessed on 20 March 2024).

67. EASA. EASA Artificial Intelligence Roadmap 1.0; EASA: Cologne, Germany, 2020; Available online: https://www.easa.europa.eu/en/downloads/109668/en (accessed on 20 March 2024).

68. European Aviation Artificial Intelligence High Level Group. The FLY AI Report Demystifying and Accelerating AI in Aviation/ATM; EUROCONTROL: Brussels, Belgium, 2020; Available online: https://www.eurocontrol.int/publication/fly-ai-report (accessed on 20 March 2024).

69. FAA. Technical Discipline: Artificial Intelligence—Machine Learning; FAA: Washington, DC, USA, 2022; Available online: https://www.faa.gov/aircraft/air_cert/step/disciplines/artificial_intelligence (accessed on 20 March 2024).

70. IATA. AI in Aviation: Exploring the Fundamentals, Threats, and Opportunities of Artificial Intelligence (AI) in the Aviation Industry; White Paper IATA: Montreal, QC, Canada, 2018; Available online: https://www.iata.org/contentassets/2d997082f3c84c7cba001f506edd2c2e/ai-white-paper.pdf (accessed on 20 March 2024).

71. European Commission. EU AI Act: First Regulation on Artificial Intelligence 2023; European Commission: Brussels, Belgium, 2023; Available online: https://www.europarl.europa.eu/news/en/headlines/society/20230601STO93804/eu-ai-act-first-regulation-on-artificial-intelligence?&at_campaign=20226-Digital&at_medium=Google_Ads&at_platform=Search&at_creation=RSA&at_goal=TR_G&at_advertiser=Webcomm&at_audience=ai%20europe&at_topic=Artificial_intelligence_Act&at_location=LV&gclid=CjwKCAjw8symBhAqEiwAaTA__IQ5BDyPcf7q7qibfaj0mvGRxZjRhUIE4DI6gri2phJOQt3Jw7L_rRoCqR8QAvD_BwE (accessed on 7 September 2023).

72. European Commission. Communication on Fostering a European Approach to Artificial Intelligence; European Commission: Brussels, Belgium, 2021; Available online: https://digital-strategy.ec.europa.eu/en/library/communication-fostering-european-approach-artificial-intelligence (accessed on 20 March 2024).

73. European Commission. Coordinated Plan on Artificial Intelligence; European Commission: Brussels, Belgium, 2021; Available online: https://digital-strategy.ec.europa.eu/en/policies/plan-ai (accessed on 20 March 2024).

74. European Commission. Proposal for a Regulation Laying Down Harmonised Rules on Artificial Intelligence; European Commission: Brussels, Belgium, 2021; Available online: https://digital-strategy.ec.europa.eu/en/library/proposal-regulation-laying-down-harmonised-rules-artificial-intelligence (accessed on 20 March 2024).

75. ICAO. ICAO’s Activities in the Field of AI; ICAO: Montreal, QC, Canada, 2023; Available online: https://www.icao.int/safety/Pages/Artificial-Intelligence-(AI).aspx (accessed on 20 March 2024).

76. Capgemini and AWS Announce the Launch of a Technology Platform to Improve the Lifespan of Aircraft Parts. Press Release. Available online: https://prod.ucwe.capgemini.com/wp-content/uploads/2023/06/2023_06_21_Capgemini_AWS_technology_platform_to-improve_the_lifespan_of_aircraft_parts.pdf (accessed on 17 May 2024).

77. Khalid, S.; Song, J.; Azad, M.M.; Elahi, M.U.; Lee, J.; Jo, S.-H.; Kim, H.S. A Comprehensive Review of Emerging Trends in Aircraft Structural Prognostics and Health Management. Mathematics; 2023; 11, 3837. [DOI: https://dx.doi.org/10.3390/math11183837]

78. Tavares, S.M.O.; Ribeiro, J.A.; Ribeiro, B.A.; de Castro, P.M.S.T. Aircraft Structural Design and Life-Cycle Assessment through Digital Twins. Designs; 2024; 8, 29. [DOI: https://dx.doi.org/10.3390/designs8020029]

79. Karalis, V.D. The Integration of Artificial Intelligence into Clinical Practice. Appl. Biosci.; 2024; 3, pp. 14-44. [DOI: https://dx.doi.org/10.3390/applbiosci3010002]

80. Junaid, S.B.; Imam, A.A.; Shuaibu, A.N.; Basri, S.; Kumar, G.; Surakat, Y.A.; Balogun, A.O.; Abdulkarim, M.; Garba, A.; Sahalu, Y. et al. Artificial Intelligence, Sensors and Vital Health Signs: A Review. Appl. Sci.; 2022; 12, 11475. [DOI: https://dx.doi.org/10.3390/app122211475]

81. Rojek, I.; Kotlarz, P.; Kozielski, M.; Jagodziński, M.; Królikowski, Z. Development of AI-Based Prediction of Heart Attack Risk as an Element of Preventive Medicine. Electronics; 2024; 13, 272. [DOI: https://dx.doi.org/10.3390/electronics13020272]

82. Berros, N.; El Mendili, F.; Filaly, Y.; El Bouzekri El Idrissi, Y. Enhancing Digital Health Services with Big Data Analytics. Big Data Cogn. Comput.; 2023; 7, 64. [DOI: https://dx.doi.org/10.3390/bdcc7020064]

83. Allioui, H.; Mourdi, Y. Exploring the Full Potentials of IoT for Better Financial Growth and Stability: A Comprehensive Survey. Sensors; 2023; 23, 8015. [DOI: https://dx.doi.org/10.3390/s23198015] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37836845]

84. Mashayekhy, Y.; Babaei, A.; Yuan, X.-M.; Xue, A. Impact of Internet of Things (IoT) on Inventory Management: A Literature Survey. Logistics; 2022; 6, 33. [DOI: https://dx.doi.org/10.3390/logistics6020033]

85. Mendi, A.F. A Digital Twin Case Study on Automotive Production Line. Sensors; 2022; 22, 6963. [DOI: https://dx.doi.org/10.3390/s22186963] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36146313]