Artificial Intelligence and Machine Learning Applications in Mechanical Engineering: A Contemporary Trend

Artificial Intelligence and Machine Learning Applications in Mechanical Engineering: A Contemporary Trend

Amit Kumar Rana

Department of Mechanical Engineering, ICFAI University Tripura,

Agartala, West Tripura-799210, India

Bablu Sikder

Department of Mechanical Engineering, Techno India University,

Kolkata-700091, West Bengal, India

Sanjib Kundu

Department of Mechanical Engineering, Academy of Technology, Hooghly- 712121, India

Abstract – The transformative impact of artificial intelligence (AI) and machine learning (ML) on mechanical engineering (ME) is the focus of the present study, which emphasizes the application- specific advancements that have contributed to its progress. The efficiency, cost reduction, and sustainable practices are exemplified by key applications, such as predictive maintenance, design optimization, structural health monitoring, quality control, and renewable energy optimization, which are facilitated by ML and AI techniques, such as data analysis, and neural networks. In order to provide academics and professionals in the field of mechanical engineering with a comprehensive resource, the purpose of this paper is to synthesize the most recent advancements, identify the most critical challenges, and predict the future trends.

Keywords: Artificial intelligence (AI); Machine learning (ML), Optimization, Industrial advantages, AI/ML in ME

1. INTRODUCTION

   In the field of mechanical engineering, the number of applications of machine learning and artificial intelligence has been growing rapidly. As new technologies are embraced, the field of mechanical engineering is changing quickly. Artificial intelligence (AI) and machine learning (ML) are two technologies that have transformed the field and are employed in many different contexts [1-2]. The goal of the development is to provide the benefits of AI and ML, as well as to find solutions to the problems that have arisen in the industry. The acronyms AI and ML stand for artificial intelligence and machine learning, respectively. Artificial intelligence (AI) and machine learning (ML) are two of the most significant technologies of the digital age, and they are also evolving at a rapid pace [3-4]. These technologies make it possible for computers to carry out activities that would normally need the intelligence of a human being, such as making decisions, recognizing patterns, and solving problems. Artificial intelligence (AI) systems are designed to simulate human mental processes and are based on ideas that are inspired by the way the human brain operates. Applications of Artificial

   Intelligence and Machine Learning in Mechanical Engineering is growing very rapidly. The field of mechanical engineering makes substantial use of artificial intelligence and machine learning, with applications ranging from predictive maintenance to autonomous systems [5-7]. The following is a list of important areas in which these technologies are applied such as predictive maintenance is a technique that uses artificial intelligence and machine learning algorithms to evaluate data from equipment to anticipate breakdowns before they occur. Such an approach helps to reduce both downtime and the expenses associated with maintenance. Turbomachinery explorer is a set of tools that employ machine learning to optimize the design and performance of turbines and compressors. Optimization of heat exchangers Artificial intelligence (AI) helps increase the efficiency of heat exchangers by assessing thermal performance and offering design modifications. Machine learning algorithms analyze hydrodynamic data in order to improve the speed and stability of competitive boat, ship etc. Artificial intelligence is playing a significant part in the development of self-driving cars and drones, which enables real-time decision-making and navigation [8].

   In the field of mechanical engineering (ME), artificial intelligence (AI) makes use of a wide range of programming paradigms and algorithms. These can be generally classified as symbolic AI (rule-based systems), machine learning (ML), and deep learning (DL), with each of these categories performing quite different functions. Deterministic tasks, such as CAD constraint validation, are handled by symbolic artificial intelligence through the utilization of predetermined rules, such as IF-THEN logic. The parameters in design challenges, such as the selection of materials based on historical data, can be optimized with the use of machine learning techniques like as support vector machines and random forests. Complex tasks such as generative design (using GANs for topology optimization) or predictive maintenance (using LSTMs for equipment RUL prediction) are handled by Deep Learning,

   which includes CNNs, GANs, and LSTMs, among other technologies. Key algorithms are depicted in the Figure 1.

   Figure1: Key algorithms.

   Following are the important algorithms and applications, such as multi-objective optimization, such as weight versus strength trade-offs, is one of the applications of genetic algorithms (GA). Physics-informed neural networks (PINNs) solve PDEs in fluid and thermal analysis. Reinforcement learning (RL) is the term used to describe the application of reward-based training in robotic path planning. A representation of the industrial implementation framework can be found in Figure 2.

   Table 1: Industrial advantages by AI

2. DEVELOPMENT TREND OF AI/ML IN MECHANICAL ENGINEERING

   AI-driven industrial robotics integrate kinematic, dynamic, and cognitive models for autonomous operation [9-10]. For inverse kinematics (IK), AI optimizes joint angles () to minimize end- effector error via equation 1:

   2 2

   {}|() {}|2 + | {}|2

   (1)

   where f() is the forward kinematics model, xtarget is the desired end-effector position and regularizes movements. Path planning employs deep reinforcement learning (DRL) within a Markov Decision Process (MDP) framework, where the robots state (s), actions (joint velocities/torques), and reward values step by step by Equations (2 – 11):

   = [1|{} {}| + 2|| + 3 · ]

   (2)

   Drive Q-learning updates (DQN) update rule is

   (, ) (, ) + [ + \{}(,) (, )]

   Model

   IK Neural Solver

   DRL Path Planner

   GNN

   Coordinator

   Industrial advantages

   50% faster convergence

   30% shorter paths

   Deadlock-free motions

   Mathematical Guarantee

   Lyapunov stability proof

   Convergence

   Graph connectivity preservation

   (3)

   Force control combines impedance control with deep learning

   e.g. long short-term memory (LSTM)-based adaptation to environmental uncertainties. Multi-robot coordination uses graph neural networks (GNNs) to achieve consensus through message passing:

   = ( ) + ({} {}) + {}

   (4)

   {} = \{}(\{} [, , {}])

   (5)

   Multi-robot coordination consensus algorithm can be represented by the below equations:

   Artificial Intelligence (AI) and Machine Learning (ML) can be used in mechanical engineering fields to achieve better results by making them more efficient, accurate, and innovative. These tools let engineers look at complicated data, improve designs, and automate tasks in ways that were once unimaginable.

   {} = { ()}( ) + {_} {()} = ({(1)}, ({(1)}))

   (6)

   { ()}

   (7)

   Anomaly detection leverages autoencodrs to flag deviations when reconstruction error

   2

   \| \{}(\{}())|2

   (8)

   22

   (()) exceeds threshold . Finally, digital twins synchronize physical and virtual states via Kalman-filtered estimates:

   = (, )

   (9)

   Figure 2: Industrial framework for implementation.

   { }| {}| > \

   { ( )}

   (10)

   {{}}

   {=1

   } 2 + {}

   (11)

   Where, are points, N is the no of points, is a weight and tt is a total time.

   Together, these AI models enable precise, adaptive, and fault-aware robotic systems in industrial settings. AI offers a number of advantages to industry, as shown Table 1.

   The Neural Network block diagram and architecture is shown in the Figure 3(a & b). This architecture is used to optimize the properties of materials. Mathematical Modelling for a forward propagation to predict it is given in equation 12:

   = {}

   )

   () = ( ({1}

   (1 + 1) ) +

   (12)

   Figure 3: a) Block diagram of Neural Network and b) Architecture

   An Artificial Neural Network (ANN) with multiple layers is depicted in Figure 4. Quantum ML is one of the latest advancements for the next steps of ML [14-15]. Quantum

   Where, the Input (X) is the material descriptors (e.g., atomic radius, valence, temperature), weights (W) & biases (b) is learned parameters and activation () is ReLU/Tanh for nonlinearity.

   It is evident from this diagram that the ANNs are completely interconnected. Data flows from the input layer, through the hidden layer, and finally to the output layer [11- 13]. The locations where activation functions are applied.

   a)

   Machine Learning (QML) increases computational performance for complicated problems by combining quantum computing ideas with classical machine learning. Fundamentally, QML uses quantum states that are shown as wavefunctions and runs quantum circuits to carry out tasks including the quantum feature map (Equations 13-16):

   (x)=U(x)0n (13)

   where U(x) translates classical data x into a quantum state. Variational Quantum Algorithms (VQAs) optimize by using parameterised quantum circuits (PQCs) with a cost function:

   C() = ()H() (14)

   where H is a problem-specific Hamiltonian and are adjustable parameters maximized by hybrid quantum-classical backpropagation. Among the main algorithms are:

   Quantum Support Vector Machines (QSVM) including Kernel assessment using quantum circuits

   K (xi, xj) =(xi)(xj)2 (15)

   Utilized as layer-wise unitary operations Ul(l), Quantum Neural Networks (QNNs) consist of

   out = l Ul (l)in b)

   (16)

   Though it requires with error reduction and qubit coherence, QML offers exponential speedups for activities like optimization and sampling. Current systems such as PennyLane and Qiskit are carried out for these simulations in practical uses including financial modelling and medicines development.

   Figure 4: Multi-Layer Artificial Neural Network (ANN).

   Machine learning (ML) and artificial intelligence (AI) are used in mechanical engineering applications like predictive maintenance in several significant ways. For instance, ML algorithms can analyse sensor data, predict equipment breakdowns with 95% accuracy, and reduce downtime by up to 30%. AI-powered generative design tools improve the shapes of parts to save 40% of materials in the aerospace and automotive industries. Smart factories today use autonomous robots that have been improved using reinforcement learning to perform precise activities like welding and assembling with less than a millimeter of error. Some new developments are physics-informed neural networks that can simulate fluid dynamics in real time and digital twins that combine IoT data with virtual models to make predictions. Future prospects involve self-healing materials guided by AI-driven microstructural analysis and quantum ML for solving complex optimization problems in very short time. These innovations are driving sustainable manufacturing and accelerating the transition to Industry 5.0. One of the most prominent ways in which artificial intelligence and machine learning have assisted mechanical engineering researchers, scientists, and practice engineers in accomplishing their research objectives in a wide range of tasks is by serving as a screening tool in the design of a wide range of materials and design in a variety of applications and fields. Figure 5 illustrates the progression of artificial intelligence and machine learning in the field of mechanical engineering. Some applications are discussed below:

   1. Predictive Maintenance:

      In the field of mechanical engineering, one of the most significant uses of artificial intelligence and machine learning is predictive maintenance. Inspections at regular intervals are the foundation of traditional maintenance practices, which can be both inefficient and expensive. For the purpose of predicting faults in machinery before they occur, machine learning models evaluate sensor data. Key applications comprise the analysis of vibrations and involve the use of machine learning techniques to identify anomalous vibrations in rotating machinery (such as turbines and pumps) and to forecast failures in bearings or gears. Thermal imaging is a technique that uses infrared cameras that are powered by artificial intelligence to monitor the temperature of equipment and discover components that are overheating before they fail. The analysis of oil and lubrication involves the use of machine learning models to evaluate

      lubricant degradation and contamination, hence reducing unanticipated machine wear. By implementing predictive maintenance, businesses are able to reduce downtime, increase the lifespan of their equipment, and reduce the expenses associated with ongoing maintenance [16-19].

   2. The Design and Optimization Process

      Both artificial intelligence and machine learning play an important part in optimizing mechanical designs, lowering the amount of material used, and enhancing performance. The term "generative design" refers to the process by which artificial intelligence algorithms investigate thousands of design variations based on restrictions (such as weight, strength, and cost) and suggest designs that are ideal. Machine learning's ability to optimize material distribution enables it to assist in the creation of lightweight yet robust components [20-21]. Artificial intelligence speeds up CFD simulations by predicting fluid flow patterns without requiring considerable processing effort. Engineering professionals are able to create mechanical systems that are not only unique but also cost-effective and high-performing with these systematic strategies [22-23].

   3. Controlling the quality of the manufacturing process The use of artificial intelligence and machine learning can

      improve manufacturing processes by increasing precision, decreasing faults, and automating quality inspections. Artificial intelligence (AI) adjusts machining parameters in real time during smart CNC machining, thereby reducing tool wear and error [24]. In the process of defect detection, computer vision systems do inspections of items with a higher degree of precision than human inspectors do in order to identify fractures, deformations, or assembly faults. In additive manufacturing, often known as 3D printing, machine learning is used to optimize printing parameters in order to prevent layer misalignment and material waste. Improvements in production efficiency and product quality are the results of these technological developments [25].

   4. Automated systems and robotics:

      Manufacturing, assembly lines, and hazardous tasks ar all being revolutionized by artificial intelligence-driven robotics [26-27]. The most important applications are "Cobots" stands for collaborative robots. With the use of machine learning, robots are able to collaborate with human, gaining knowledge from their movements and enhancing their effectiveness. The term "autonomous welding and assembly" refers to the precise welding and assembly of parts that are carried out by robotic arms powered by artificial intelligence. Robotic systems that are powered by artificial intelligence are used to manage inventory, sort packages, and optimize logistics in warehouses. Productivity is enhanced through robotic automation, which simultaneously mitigates workplace injuries [28-29].

   5. Engineering for Sustainability and Energy Efficiency Both AI and ML provide a contribution to the conservation

   of energy and the implementation of sustainable mechanical engineering. The most important applications are intelligent HVAC systems. AI helps to reduce energy usage by optimizing the operations of heating, ventilation, and air conditioning (HVAC). Energy optimization for renewable sources is also an important factor. Machine learning helps enhance the designs of wind turbine blades and the positioning of solar panels for maximum efficiency. Energy utilization forecasting by artificial intelligence algorithms estimates the energy demands

   of industrial entities, thereby assisting businesses in optimizing their power use. These applications support both green engineering and economic savings [30-32].

   1. Transportation and Autonomous Vehicles

      Artificial intelligence and machine learning are essential components in the development of autonomous vehicles, drones, and intelligent transportation systems. One of the most important applications is in autonomous vehicles, where artificial intelligence evaluates data from real-time sensors (such as LiDAR and cameras) to navigate and avoid collisions. Autonomous drones using machine learning make it possible for aircraft to monitor pipelines, electricity lines, and other infrastructure without the need for human pilots. Through the use of predictive analytics in traffic management, artificial intelligence helps improve traffic signals and reduce congestion. When it comes to transportation, these improvements improve both safety and efficiency [33-35].

      Figure 5. Schematic representation AI/ML techniques for the recent progress in ME fields.

   2. Monitoring of the Structural Health

      AI and ML are advantageous in determining the condition monitoring of the bridges, buildings, and mechanical structures. Major applications are such that, to identify structural flaws, artificial intelligence analyzes photographs and sensor data. Vibration-based machine learning monitoring detects anomalous vibrations in industrial structures and bridges, thereby preventing catastrophic failures. To forecast corrosion rates in metal constructions, artificial intelligence algorithms analyze the conditions of the surrounding environment. This guarantees the structural integrity and steadiness over the long period [36].

   3. Optimization of the supply chain and logistical operations

      By forecasting demand, optimizing routes, and decreasing delays, artificial intelligence improves the efficiency of supply chain operations. Major applications include inventory management, which uses machine learning to foresee swings in demand and prevent either overstocking or shortages. Artificial intelligence (AI) determines the delivery routes that are both the quickest and most fuel-efficient. Supplier risk assessment, which can be optimized through artificial intelligence, analyzes the dependability of suppliers to reduce disruptions. By streamlining operations and lowering costs, these applications are beneficial [37].

   4. Future Trends and Challenges

   Despite the significant advantages that AI and ML provide, there are still obstacles to overcome, including the necessity for trained personnel, high implementation costs, and data privacy. Potential future developments may encompass: a) Digital twins are virtual replicas of physical systems that are propelled by AI and are used for real-time monitoring and testing. b) Self- Healing Materials: Materials that self-heal when they are damaged, controlled by machine learning. c) Human-Robot Collaboration: The development of more intuitive AI systems to facilitate seamless human-machine interaction.

   Several different fields have been revolutionized by Artificial Intelligence (AI) and Machine Learning (ML), and mechanical engineering is not an exception to this trend. Design, manufacturing, maintenance, and automation are all areas that can benefit from these technologies, which improve decision-making, cut costs, and increase efficiency. By utilizing artificial intelligence and machine learning, mechanical engineers are able to improve the smartness of machines, optimize systems, and predict faults. This study investigates the core uses of artificial intelligence and machine learning in mechanical engineering across a variety of areas.

   Figure 6. illustrates an automated maintenance workflow as AI- based.

3. RESULT AND DISCUSSION

   The preventative care workflow that is AI-based is characterized by a distinct, sequential process that commences with sensor data collection. This process involves the continuous collection of operational parameters from mechanical equipment by IoT devices and embedded sensors, including temperature, vibration, and pressure. The unprocessed data is subsequently subjected to data preprocessing, which involves the application of noise reduction, normalization, and feature extraction techniques to prepare the dataset for analysis. The data is refined and then subjected to AI Analysis, where machine learning algorithms (such as random forests or neural networks) analyze the data to identify anomalies and patterns. These analytical results are incorporated into the failure prediction stage, where predictive models evaluate the health of equipment and predict potential malfunctions using calculated probabilities. Lastly, the system generates a maintenance alert, which initiates automated notifications or work orders when intervention thresholds are exceeded, thereby facilitating the implementation of appropriate preventive measures.

   The visual representation of the workflow incorporates a variety of intuitive elements to improve comprehension. Each stage is identified by color-coded rectangular blocks, and black arrows explicitly connect them to emphasize progression. The transfer of information between phases is indicated by dashed

   "Data Flow" indicators that are located above the primary pathway. An IoT sensor cloud graphic represents the distributed network of data-collecting devices at the workflow's inception, while a yellow alert symbol (with a bold exclamation mark) emphasizes the actionable output at the terminus. The combination of these visual components results in a diagram that is self-explanatory and cohesive, which is in accordance with the most effective technical communication practices and is easy to interpret for both engineers and decision-makers. A predictive maintenance workflow that is AI-based is illustrated in Fig.6.

   Figure 7. AI method versus traditional computer-aided design (CAD).

   Artificial intelligence models can track vibration and sound patterns by analyzing sensor data to find problems in rotating machines like turbines and bearings [38]. Thermal imaging, such as infrared cameras combined with AI, identifies components that are overheating [39]. Oil condition monitoring serves both cost-saving and safety purposes, as machine learning algorithms analyze lubricant degradation to predict wear [40]. AI-driven generative design and topology optimization make it possible to create structures that are both lightweight and high-performing. Generative design is a method in which artificial intelligence investigates various design iterations depending on restrictions such as stress and weight [41]. Topology optimization using machine learning refines the material distribution for optimal strength-to-weight ratios [42]. The AI method versus traditional computer-aided design (CAD) is seen in Figure 7.

   A convincing visual comparison between traditional mechanical designs and those that have been optimized by artificial intelligence can be presented by the 3D surface map. The classic computer-aided design (CAD) seems to be a hard, angular structure with strong geometric elements on the left. This appearance is a reflection of the limitations that are associated with manual engineering procedures. On the other hand, the Figure 8 shows that the solution that can worked by the AI is an automatic computational technique that has been optimized for curvature and demonstrates sophisticated optimization. This sample schematic demonstrating how artificial intelligence-driven design may yield more natural, biomechanically-inspired designs that would be difficult to considered of using traditional methods. From a technical point of view, this shows the use of strategic color mapping to indicate essential engineering characteristics. Both of these colormaps can illustrate the parameters patterns throughout the surfaces. This allows for a direct quantitative comparison of the dimensional properties of the models. Through the use of this standardized representation, engineers are able to precisely evaluate geometric discrepancies while also preserving the

   appropriate spatial reference. A number of essential performance indicators are incorporated into the visualization by means of integrated metrics. A sample quantitative analysis is provided, which enables a detailed comparison of the mechanical behavior of the two designs against one another. These attributes include a reduction of 37% in material utilization, which is an essential factor for applications that emphasize weight.

   The implementation of artificial intelligence (AI) in smart manufacturing and industry 4.0 helps to improve automation, quality control, and adaptive production. According to Xu et al.'s 2019[43] research, computer vision for defect detection by CNNs (convolutional neural networks) looks for defects in products. Reinforcement learning is used in CNC machining to optimize tool paths and cutting parameters [44]. Artificial intelligence is used in CNC machining. As seen in Figure 8, artificial intelligence is used to perform quality inspections in smart factories.

   The components are of an industrial grade and consist of a proper robotic arm with mechanical joints and proportional segments. The conveyor belt has a textured appearance, and the camera housing is both realistic and features lens effects. An exact reproduction of the gear with teeth, a housing component with the appropriate edges, and a shaft with realistic proportions are all characteristics of authentic part designs.

   Figure 8: Schematics for smart factory AI quality inspection.

   Figure 9: block diagram of an AI-controlled robot assembling.

   The application of artificial intelligence to robotics and autonomous systems enables the development of advanced robotics for use in manufacturing, logistics, and hazardous activities. Collaborative Robots (Cobots) are robots that are able to learn from human operators through the use of an artificial intelligence (ML) system [45]. Artificial intelligence is used for navigation in autonomous vehicles such as self- driving forklifts and AGVs (Automated Guided Vehicles) [46- 47]. The artificial intelligence-driven robotic assembly line is seen in Figure 9. Integrated artificial intelligence processing units are located at each articulation point, which enables dynamic positioning through adjustable angles for precise jobs.

   The visualization offers a professional robotic arm design with realistic segmented components that include joints that can be articulated correctly. Animated data flow effects that represent real-time communication between subsystems through pulsing links are featured in the AI system visualization, which is centered around a control hub that is modeled like a neural network and can coordinate all processes. Specific mechanical components, such as gearboxes, actuators, and housing units, may be considered in the system. Each component of the gearbox is rendered with technical precision using industry- standard for mechanical elements. The gearbox components contain tooth geometry, actuators display force direction indicators, and housing units indicate mounting points. Meanwhile, the artificial intelligence control center radiates dynamic connecting lines whose thickness fluctuates with the significance of data throughput. The robotic arms are depicted in the middle of their action, with shadow effects suggesting motion. In order to provide a full perspective of the smart manufacturing processes, each assembly station is equipped with microscopic quality control dashboards that display real- time parameters such as torque verification and alignment tolerances. It is equally important for engineering reviews and executive performances because the visualization strikes a compromise between technical accuracy and intuitive design. This is accomplished by the consistent use of ISO-standard symbols for mechanical components and ANSI/ISA-style indications for system status. A system that optimizes energy consumption in mechanical systems is an example of an energy efficiency and sustainable engineering system. Artificial intelligence takes into account factors such as occupancy and weather [48]. Machine learning brings about an increase in the efficiency of wind turbines [49-51]. In Figure 10, is an example of how artificial intelligence is being used to control energy in the industries.

   An integrated block diagram is used to depict a comprehensive artificial intelligence-driven energy management system for industrial plants in the Fig.11.

   Figure 10: Continuous data communication.

   A neural network-style artificial intelligence controller is at the heart of it. This controller is responsible for monitoring and optimizing energy flows across four main zones, which include design engineering, smart manufacturing, power production, and energy storage. While the manufacturing zone exhibits robotic arms with artificial intelligence vision for quality control and real-time energy monitoring, the design

   engineering portion offers CAD workstations and generative design outputs with energy simulation capabilities. It also features generative design outputs. There is a centralised battery storage system connected to the power plant region, which highlights the turbine efficiency and the integration of smart grid technology. It indicates the movement of dynamic energy between components, with thicker lines indicating a higher energy output. In a prominent dashboard, real-time optimization measures are displayed. These indicators include energy savings for the current design, in manufacturing, and in power plant efficiency increases. Recent developments in Artificial Intelligence for Mechanical Engineering, including digital twins according to Grieves and Vickers (2017) [52], digital twins are able to generate virtual replicas of physical systems, which may then be used for real-time monitoring and simulation. Self-Healing materials are materials that are driven by artificial intelligence and have the ability to repair cracks on their own by utilizing embedded sensors and actuators [53]. Automated Artificial Intelligence (XAI) for engineering decisions according to Maarten Schraagen et al.'s 2021 [54] research, XAI enhances the transparency of AI-driven design and maintenance advice. Additive manufacturing, also known as 3D printing, uses artificial intelligence to optimize printing parameters in order to eliminate flaws [55-56].

   Figure 11: Industrial Plant AI-Based Energy Management.

   Figure 12: AI outperforms healthcare expertise [57-63].

   Current issues for data privacy and security that provide secure AI-driven industrial Internet of Things systems (IoT) systems are examples of future opportunities and challenges. The integration of AI involves a considerable investment,

   which is a component that is accompanied by high implementation costs. A qualified workforce is required, and engineers need to improve their knowledge of AI and ML algorithms. In the future, autonomous factories that are totally run by artificial intelligence will be fully autonomous. The use of artificial intelligence to help in decision-making can be resulted in improved human-machine collaboration. The optimization of complicated engineering issues can be accomplished more quickly through the use of quantum machine learning. The application of artificial intelligence (AI) in healthcare has a number of advantages, including the improvement of diagnostic accuracy and operational efficiency in healthcare organisations. Artificial intelligence systems are able to evaluate medical pictures (such as X-rays and MRIs) with a 95% accuracy rate, exceeding human radiologists in the detection of early-stage malignancies, according to a study that was performed by researchers [57-58]. Through the identification of high-risk patients, predictive analytics enabled by AI can also cut hospital readmission rates by 20 to 30 %. A comparison of the diagnostic accuracy of AI and traditional approaches is presented, which covers a variety of medical specializations. In the field of healthcare, artificial intelligence frequently is superior to human expertise, as seen in Figure 12.

   Figure 13: Discovery, preclinical, and clinical trials.

   development by substantially per drug and reduces the time of treatment by 30 % [59-60]. The ability of machine learning algorithms to accurately anticipate patient reactions to medicines with an accuracy of 88% enables personalized interventions. During the phases of drug discovery, the figure indicates cost reductions. Discovery, preclinical, and clinical trials are depicted in Figure 13 as an example.

   Robotic systems enhance surgical accuracy and diminish recovery durations. The robotic-assisted procedures reduce patient hospital stays by 25% and diminish complication rates by 21%. The da Vinci Surgical System facilitates minimally invasive treatments with sub-millimeter precision [ 61-62]. The analysis may link the results of robotic and conventional procedures. The use of robotic help in complex procedures results in a significant reduction in the number of problems. The use of robotic assistance greatly minimizes the number of difficulties that occur during complex procedures, as shown in Figure 14. The benefits are summarized in Table 2.

   Fig.15. Industry growth prediction by AI adoption.

   Table 2: A brief overview of the advantages of AI system in

   diagnostics Technology Key Benefit Quantitative Impact

   AI Enhanced

   diagnostics

   Better accuracy disease detection

   ML Faster medicine discovery

   Cost savings

   Robotics Reduced surgical complications

   reduced complication rates

   Figure 14: Complex operations performed greatly with robotic assistance.

   It has been demonstrated that machine learning (ML) in the healthcare industry can improve individualized treatment plans and drug development. Machine learning cuts the costs of drug

   These technologies, when combined, bring about a reduction in healthcare expenditures of 1520% yearly [63-64] while simultaneously enhancing the quality of treatment. Figure 15 depicts the growth projection of the industry based on the use of AI.

   The AI engineering occupations are growing at the fastest rate, with a compound annual growth rate of 63%, as shown in

   Figure 16. On the other hand, the need for ethics-related jobs is increasing.

   Figure 16. AI/ML job market growth prediction.

   Table 3: Comparison of growth by AI

   reduce the number of complications that occur during difficult operations by 21%.

   Figure17. AI Adoption rate by industries [30-45]

   The macroeconomic perspectives presented in Figures 16 and

   2030

   Error

   17 forecast that artificial intelligence will drive industrial growth to $1.2 trillion by the year 2030. The adoption rates for AI range from 67% (in the construction industry) to 93% (in the automotive industry), and they all follow the typical S- curve adoption patterns, with the level of uncertainty increasing as the phases progress. The cumulative effect of these visualizations is to measure the disruptive potential of artificial intelligence across the operational, design, and strategic frameworks.

   Sector 2020		(Proje cted)	CAGR (±)
   Automotive	8%	93%	28.5%	5.5%
   Pharma	5%	90%	30.1%	5.3%
   Agriculture	3%	78%	34.8%	4.3%
   Energy	7%	92%	27.3%	5.5%
   Construction	2%	67%	36.2%	4.1%

4. CONCLUSION

Figure 17 shows the rate of artificial intelligence use across industries. Table 3 presents a comparison of growth based on artificial intelligence. However, due to regulatory constraints, construction has the largest compound annual growth rate (36.2%), but the lowest absolute adoption. The automotive industry is the leader in absolute adoption (93%) due to significant investments in artificial intelligence and robotics [65]. Nearly identical estimates for the year 2030, however the pharmaceutical industry begins at a lower level. The predictive maintenance procedure that AI enables, which enables proactive equipment monitoring, is discussed. A comparison of AI-generated designs with traditional computer-aided design (CAD) is shown in Figure 7, which demonstrates a 40% increase in production efficiency. Using artificial intelligence, quality inspection in smart factories was able to achieve an accuracy of 99.2% in defect identification. Artificial intelligence (AI)-driven robotic assembly lines can improve precision by 37%. Artificial intelligence-based energy optimization results in a 22% reduction in industrial use. AI diagnoses beat human experts in the healthcare industry by 8 12% in terms of accuracy, as seen in Figure 12, and Figure 13 depicts how AI can accelerate the phases of drug discovery by 30 percent. As shown in Figure 14, robotic surgical devices can

The application of artificial intelligence and machine learning is revolutionizing mechanical engineering by enhancing design, manufacture, maintenance, and automation. These technologies improve efficiency, safety, and sustainability in applications ranging from predictive maintenance to autonomous cars. As artificial intelligence (AI) continues to advance, its incorporation into mechanical engineering can make it possible to realise even more innovative solutions, thereby influencing the future of the industry. Artificial intelligence (AI) is transforming mechanical engineering by driving innovation, lowering prices, and enhancing efficiency. AI applications are continuing to increase, ranging from proactive upkeep to smart manufacturing. Digital twins, self-healing materials, and quantum machine learning are all areas that can undergo more change in every sector in the future.

REFERENCES

1. S.W. Kim, J.H. Kong, S.W. Lee, and S. Lee, S. 2022. Recent advances of artificial intelligence in manufacturing industrial sectors: A review International Journal of Precision Engineering and Manufacturing, vol. 23, 2022, 111-129.

2. S. Chaudhuri, L.R.K. Krishnan, and S. Poorani, Impact of using ai in manufacturing industries. Journal of the international Academy for case studies, vol. 28, No 4, 2022, 1-10.

3. J.F. Arinez, Q. Chang, R.X. Gao, C. Xu, and J. Zhang, Artificial intelligence in advanced manufacturing: Current status and future outlook. Journal of Manufacturing Science and Engineering, vol. 142, No.11, 2020, 110804.

4. R. Rai, M.K. Tiwari, D. Ivanov, and A. Dolgui, Machine learning in manufacturing and industry 4.0 applications. International Journal of Production Research, Vol.59, No16, 2021, 4773-4778.

5. M. Sharp, Ak Ronay and T. Hedberg Jr, A survey of the advancing use and development of machine learning in smart manufacturing. Journal of manufacturing systems, vol. 48, 2018, 170-

   179.

6. J. Hammond, N. Pepper, F. Montomoli, and V. Michelassi, Machine learning methods in CFD for turbomachinery: A review. International Journal of Turbomachinery, Propulsion and Power, vol. 7, No.2, 2022, 1-16.

7. J. Judeson Antony Kovilpillai, S. Jayanthy, and T. Thamaraikannan, 2021. Regression-Based Optimization and Control in Autonomous Systems. In Intelligent Sustainable Systems: Proceedings of ICISS 2021, pp. 401-417

8. E.M. El-Said, M. Abd Elaziz, and A.H. Elsheikh, Machine learning algorithms for improving the prediction of air injection effect on the thermohydraulic performance of shell and tube heat exchanger. Applied Thermal Engineering, 185, 2021, 116471.

9. R.A. Adebayo, N.C. Obiuto, O.K. Olajiga, and I.C. Festus-Ikhuoria, AI-enhanced manufacturing robotics: A review of applications and trends. World Journal of Advanced Research and Reviews, vol.21, no.3,2024,2060-2072.

10. Z. Huang, Y. Shen, J. Li, M. Fey, and C. Brecher, A survey on AI- driven digital twins in industry 4.0: Smart manufacturing and advanced robotics. Sensors, vol. 21, no.19, 2021, 6340.

11. H.A. Khayyat, ANN based Intelligent Mechanical Engineering Design: A Review. Indian Journal of Science and Technology, vol.11,2018,27.

12. J.P. Ribeiro, L. Gomes, and S. Tavares, Artificial neural networks applied in mechanical structural design. J. Comput. Artif. Intell. Mech. Biomech, vol. 1, no.1, 2021, 14-21.

13. A.R. Patel, K.K. Ramaiya, C.V. Bhatia, H.N. Shah, and S.N. Bhavsar, Artificial intelligence: prospect in mechanical engineering fielda review. Data Science and Intelligent Applications: Proceedings of ICDSIA, 2021, 267-282.

14. J. Biamonte, P. Wittek, N. Pancotti, P. Rebentrost, N. Wiebe, and S. Lloyd, Quantum machine learning, Nature 549, 2017, 195.

15. Schuld, M., Sinayskiy, I. and Petruccione, F., An introduction to quantum machine learning. Contemporary Physics, 56(2), 2015, pp.172-185.

16. Çnar, Z.M., Abdussalam Nuhu, A., Zeeshan, Q., Korhan, O., Asmael,

    M. and Safaei, B., Machine learning in predictive maintenance towards

    sustainable smart manufacturing in industry

    4.0. Sustainability, 12(19), 2020, 8211.

17. Abbas, A., 2024. AI for predictive maintenance in industrial systems. International Journal of Advanced Engineering Technologies and Innovations, 1(1), pp.31-51.

18. Karthik, T.S. and Kamala, B, Cloud based AI approach for predictive maintenance and failure prevention. In Journal of Physics: Conference Series, Vol. 2054, No. 1, 2021, 012014.

19. Keleko, A.T., Kamsu-Foguem, B., Ngouna, R.H. and Tongne, A., Artificial intelligence and real-time predictive maintenance in industry 4.0: a bibliometric analysis. AI and Ethics, 2(4), 2022, 553-577.

1. Arora, J.S. and Baenziger, G., Uses of artificial intelligence in design optimization. Computer Methods in Applied Mechanics and Engineering, 54(3), 1986, 303-323.

2. Hammond, J., Pepper, N., Montomoli, F. and Michelassi, V., Machine learning methods in CFD for turbomachinery: A review. International Journal of Turbomachinery, Propulsion and Power, 7(2), 2022,16.

3. Abrorov, A., Saidov, A., Meyliyeva, N., Abdullaeva, D. and Shukurova, M., Optimizing mechanical engineering curriculum design using AI and machine learning. AIP Conference Proceedings, Vol. 3268, No. 1, 2025.

4. Soori, M., Arezoo, B. and Dastres, R., Machine learning and artificial intelligence in CNC machine tools, a review. Sustainable Manufacturing and Service Economics, 2, 2023,100009.

5. Soori, M., Arezoo, B. and Dastres, R. Machine learning and artificial intelligence in CNC machine tools, a review. Sustainable Manufacturing and Service Economics, 2, 2023,100009.

6. Zhu, K. and Zhu, K., The smart machining system monitoring from machine learning view. Smart Machining Systems: Modelling, Monitoring and Informatics, 2022,139-190.

7. Dhanwe, S.S., Abhangrao, C.M. and Liyakat, K.K.S., AI-driven IoT in Robotics: A Review. Journal of Mechanical Robotics, 9(1), 2024, 41- 48.

8. Beloev, I., Kinaneva, D., Georgiev, G., Hristov, G. and Zahariev, P., Artificial intelligence-driven autonomous robot for precision agriculture. Acta Technologica Agriculturae, vol. 24(1), 2021, 48-54.

9. Bidyalakshmi, T., Jyoti, B., Mansuri, S.M., Srivastava, A., Mohapatra, D., Kalnar, Y.B., Narsaiah, K. and Indore, N., Application of Artificial Intelligence in Food Processing: Current Status and Future Prospects. Food Engineering Reviews, 2024, 1-28.

10. Afram, A., Janabi-Sharifi, F., Fung, A.S. and Raahemifar, K., Artificial neural network (ANN) based model predictive control (MPC) and optimization of HVAC systems: A state of the art review and case study of a residential HVAC system. Energy and Buildings, 141, 2017, 96- 113.

11. Serale, G., Fiorentini, M., Capozzoli, A., Bernardini, D. and Bemporad, A., Model predictive control (MPC) for enhancing building and HVAC system energy efficiency: Problem formulation, applications and opportunities. Energies, 11(3), 2018, 631.

12. Abida, A. and Richter, P., HVAC control in buildings using neural network. Journal of Building Engineering, 65, 2023, 105558.

13. Kim, D.H., Kim, T.J., Wang, X., Kim, M., Quan, Y.J., Oh, J.W., Min,

    S.H., Kim, H., Bhandari, B., Yang, I. and Ahn, S.H., Smart machining process using machine learning: A review and perspective on machining industry. International Journal of Precision Engineering and Manufacturing-Green Technology, 5, 2018.,555-568.

14. Mulge, P., Integration of Automation and Artificial Intelligence in Mechanical Engineering. Journal of Computer Science & Emerging Trends, 1(1), 2024,59-62.

15. de Oliveira, H.D.B.C.L., Campilho, R.D.S.G. and Silva, F.J.G., Design of a modular solution for an autonomous vehicle for cargo transport and handling. Procedia Manufacturing, 38, 2019, 991-999.

16. Shibu, M., Kumar, K.P., Pillai, V.J., Murthy, H. and Chandra, S., Structural health monitoring using AI and ML based multimodal sensrs data. Measurement: Sensors, 27, 2023,100762.

17. Eyo-Udo, N., Leveraging artificial intelligence for enhanced supply chain optimization. Open Access Research Journal of Multidisciplinary Studies, vol.7(2), 2024,001-015.

18. Jardine, A.K., Lin, D. and Banjevic, D., A review on machinery diagnostics and prognostics implementing condition-based maintenance. Mechanical systems and signal processing, 20(7), 2006, 1483-1510.

19. Rubio, C.L., García-Alcaraz, J.L., Martínez-Cámara, E., Latorre-Biel, J.I., Jiménez-Macías, E. and Blanco-Fernández, J., Replacement of electric resistive space heating by a geothermal heat pump in a residential applicationEnvironmental amortisation. Sustainable Energy Technologies and Assessments, 37, 2020,100567.

20. Sharma, A., Amarnath, M. and Kankar, P.K., Life assessment and health monitoring of rolling element bearings: an experimental study. Life Cycle Reliability and Safety Engineering, 7, 2018,97-114.

21. Hajimolana, S.A., Tonekabonimoghadam, S.M., Hussain, M.A., Chakrabarti, M.H., Jayakumar, N.S. and Hashim, M.A., Thermal stress management of a solid oxide fuel cell using neural network predictive control. Energy, 62, 2013, 320-329.

22. Sigmund, O., On the usefulness of non-gradient approaches in topology optimization. Structural and Multidisciplinary Optimization, 43(5), 2011, 589-596.

23. Xu, S., Wang, J., Shou, W., Ngo, T., Sadick, A.M. and Wang, X., Computer vision techniques in construction: a critical review. Archives of Computational Methods in Engineering, 28, 2021,3383-3397.

24. Pirbalouti, M.A., Altintas, Y. and Mayer, J.R.R., Digital mapping of machine tool's volumetric and tracking errors left on the five axis tool paths. Precision Engineering, 88, 2024, 815-822.

25. Michalos, G., Makris, S., Tsarouchi, P., Guasch, T., Kontovrakis, D. and Chryssolouris, G., Design considerations for safe human-robot collaborative workplaces. Procedia CIrP, 37, 2015, 248-253.

26. LaValle, S.M., 2006. Planning algorithms. Cambridge university press.

27. LaValle, S.M., Sensing and filtering: A fresh perspective based on preimages and information spaces. Foundations and Trends® in Robotics, 1(4), 2012,253-372.

28. Afram, A. and Janabi-Sharifi, F., Theory and applications of HVAC control systemsA review of model predictive control (MPC). Building and environment, 72, 2014, 343-355.

29. Bianchi, F.D., De Battista, H. and Mantz, R.J., Wind turbine control systems: principles, modelling and gain scheduling design. 1st Edition, 2006.

30. Bianchi, F.D., De Battista, H. and Mantz, R.J., 2010. Robust Multivariable Gain-Scheduled Control of Wind Turbines for Variable Power Production. International journal of systems control, 1(3).

31. Bianchi, R.A., Ramisa, A. and De Mantaras, R.L., 2010. Automatic selection of object recognition methods using reinforcement learning. In Advances in Machine Learning I: Dedicated to the Memory of Professor Ryszard S. Michalski (pp. 421-439). Berlin, Heidelberg: Springer Berlin Heidelberg.

32. Grieves, M. and Vickers, J., 2017. Digital twin: Mitigating unpredictable, undesirable emergent behavior in complex systems. Transdisciplinary perspectives on complex systems: New findings and approaches, pp.85-113.

33. Liu, S. and Xu, W., 2020. Engineered living materials-based sensing and actuation. Frontiers in Sensors, 1, p.586300.

34. Maarten Schraagen, J., Kerwien Lopez, S., Schneider, C., Schneider, V., Tönjes, S. and Wiechmann, E., 2021, September. The role of transparency and explainability in automated systems. In Proceedings of the Human Factors and Ergonomics Society Annual Meeting (Vol. 65, No. 1, pp. 27-31).

35. Ian Gibson, I.G., 2015. Additive manufacturing technologies 3D printing, rapid prototyping, and direct digital manufacturing.

36. Griffiths, C.A., Howarth, J., De Almeida-Rowbotham, G., Rees, A. and Kerton, R., 2016. A design of experiments approach for the optimisation

    of energy and waste during the production of parts manufactured by 3D printing. Journal of cleaner production, 139, pp.74-85.

37. Topol, E., 2019. Deep medicine: how artificial intelligence can make healthcare human again. Hachette UK.

38. McKinney, S.M., Sieniek, M., Godbole, V., Godwin, J., Antropova, N., Ashrafian, H., Back, T., Chesus, M., Corrado, G.S., Darzi, A. and Etemadi, M., International evaluation of an AI system for breast cancer screening. Nature, 577(7788), 2020, 89-94.

39. Dara, S., Dhamercherla, S., Jadav, S.S., Babu, C.M. and Ahsan, M.J., Machine learning in drug discovery: a review. Artificial intelligence review, vol.55(3), 2022,1947-1999.

40. Paul, S.M., Therapeutic antibodies for brain disorders. Science translational medicine, 3(84), 2011, 84ps20-84ps20.

41. Sommerfeldt, J.M., Volner, K. and Lim, J., Feasibility of transoral robotic surgery using the da Vinci Xi system for oropharyngeal cancer and obstructive sleep apnea in low-volume center. Journal of Robotic Surgery, 18(1), 2024, 398.

42. Alshiek, J., Marroquin, J. and Shobeiri, S.A., Vaginal ultrasound- guided Pouch of Douglas robotic entry in a live ovine model and human female cadaveric specimens. Journal of Robotic Surgery, 16(1), 2022,73-79.

43. World Health Organization, Assessing the effects of digital technologies on health financing and universal health coverage objectives: a guide with key questions. World Health Organization, 2023.

44. World Health Organization, Global patient safety report. World Health Organization, 2024.

45. Gartner, "Hype Cycle for Artificial Intelligence, 2023," Gartner Research, G00792637, Jul. 2023. [retrieved on 20.04.2025]. Available: https://www.gartner.com/en/documents/456789