Introductory Chapter: Welding in the Era of Industry 5.0

1. Introduction

The outstanding advancement of manufacturing and industrial processes may be seen in the journey from Industry 1.0 to Industry 5.0 [1, 2, 3, 4, 5], as shown in Figure 1. With the emergence of Industry 1.0 in the late 1700s, agricultural economies saw a dramatic shift to mechanized industry based on steam and water power. Revolutionary innovations like steam engine and mechanized looms were introduced during this time. During the late nineteenth and early twentieth centuries, Industry 2.0 saw the rise of electrification and mass manufacturing. Developments like electricity and the assembly line made it easier to produce standardized items on a large scale in an efficient manner [7]. The Digital Revolution, or Industry 3.0, began in the late twentieth century with the advent of computer technology, early robots, and automation. Information technology and computer-controlled manufacturing saw a paradigm change as a result [8].

Figure 1.

Evolution of industry, from Industry 1.0 to Industry 5.0 [6].

The early twenty-first century marked the advent of Industry 4.0, the Fourth Industrial Revolution. It centered on automation, data exchange, Internet of Things (IoT), and cyber-physical systems [9, 10, 11]. This era facilitated the development of smart factories, predictive maintenance, and the concept of digital twins. In the emerging era of Industry 5.0, the focus is on the collaboration of humans and machines [7]. This phase introduces collaborative robots (COBOTS) working in collaboration with humans. The concept behind this collaboration is to combine human creativity and decision-making skills with advanced technologies to develop a more innovative, efficient, and adaptive industrial environment [12]. COBOTS assist humans in tasks that require dexterity, precision, and rapid adaptation [13]. Even in complicated, nonroutine circumstances, this collaboration helps in quick problem-solving and improving productivity with enhanced product quality. As a means of bridging the gap between human expertise and technological capabilities, Industry 5.0 envisions manufacturing as a future where innovation and adaptation are critical.

2. A comparison between current (Industry 4.0) and proposed (Industry 5.0) industrial revolution

The comparison between current (Industry 4.0) and proposed (Industry 5.0) industrial revolution based on various aspects is tabled in Table 1 [2, 18]:

Aspect	Current industrial revolution, Industry 4.0	Proposed industrial revolution, Industry 5.0
Emphasis	Automation and digitization	Human–machine collaboration [10]
Key technology	IoT, big data, AI, and automation [2]	Collaborative robotics [10], AI decision support, and hyper-automation [14]
Human involvement	Humans in a supervisory role	Humans actively collaborate with machines
Role of automation	Extensive automation of processes	Automation complements human capabilities
Customization	Focus on customization and batch size	Individualization of products and “mass customization” [15]
Complexity and problem-solving	Limited human intervention in complex tasks	Humans actively address complex problems
Real-time decision support	Limited, often machine-driven	AI provides real-time decision support [16]
Interoperability	Emphasized for machine-to-machine	Collaboration between human and machine
Data analytics	Big data and analytics for process optimization	Data analytics for insights and decision support
Transition	Transition from manual to automated processes	Transition from full automation to collaboration [17]

Table 1.

Status of technology development in Industry 4.0 and proposed Industry 5.0.

3. The increasing role of AI and ML in manufacturing

The technologies of Artificial Intelligence (AI) and Machine Learning (ML) are revolutionary, and their goal is to give machines the capacity to think like humans do, that is, to learn, reason, and make judgments. The detailed classification of Artificial Intelligence is shown in Figure 2. AI is the larger area that includes a variety of approaches that allow machines to emulate intelligent behavior. ML, a subset of AI, is the study of creating algorithms that let computers learn from and make predictions or judgments based on data [11]. The information of different modes such as physical, simulation or data basis, response of machines can be seen in Figure 3 in detail.

Figure 2.

Artificial intelligence.

Figure 3.

Data-driven manufacturing. (1) particular manufacturing process, (2) relevant process data collection, (3) using data as a basis for training an ML model, and (4) use of trained model, to perform quality estimations for decision support [19].

The integration of AI and ML into manufacturing is a transformative phenomenon that is revolutionizing the industry on multiple fronts. Complex product assembly is made possible by AI-driven robotics and automation, with excellent accuracy and efficiency [20]. AI-powered quality control systems enable real-time defect detection by analyzing sensor data and images, which allows for immediate corrective action and guarantees high-quality output. ML improves manufacturing processes by evaluating large datasets to find patterns and trends [21], and AI improves predictive maintenance, which reduces the downtime and maintenance expenses [3]. With the help of these technologies, manufacturing systems will become more responsive, flexible, and agile.

4. The importance of welding in manufacturing

Welding is an important manufacturing process that creates durable joins in materials, typically metals or thermoplastics, by melting their edges and blending them together (Figure 4). The purpose of joint is to join two similar and dissimilar grades of materials for fulfilling the desired mechanical and physical properties in application. Sometimes the resulting weld is often stronger than the original components due performed thermal cycle and filler materials. Beyond manufacturing, welding plays a vital role in repair and maintenance, especially in sectors like aerospace and defense, to ensure safety and precision. Welding serves as the backbone of various industries, including manufacturing, construction, automotive, energy, and shipbuilding, for the fabrication of structures and components [23].

Figure 4.

Different development stages of arc welding [22].

The development of welding spans centuries, beginning with early manual techniques. The Industrial Revolution introduced mechanized approaches, followed by twentieth-century advancements. The transformative leap occurred with robotic welding, offering unparalleled accuracy, precision, and safety. Integration of AI and ML in recent years has further elevated robotic welding, enabling adaptive control, real-time optimization, defect detection, and predictive maintenance.

The evolving welding industry demands innovation to meet modern industrial demands. Advanced techniques are crucial for enhancing production and weld quality along with worker’s safety. Also, enhancing sustainability involves the use of resource optimization and energy-efficient techniques. By overcoming these challenges and taking advantage of these opportunities, welding productivity, quality, and sustainability are all increased.

5. Robotic welding and automation

Automation and robotic welding both refer to the automated and controlled use of robotic devices for welding processes as shown in Figure 5. These systems employ robotic arms or manipulators that are equipped with welding instruments to perform welding operations with enhanced output, better quality, and less costs. This technique is used in a variety of industries, such as manufacturing, construction, and automotive, where robotic systems can effectively manage complex or repetitive welding processes [24, 25, 26].

Figure 5.

(a) Steps in robotic welding [24] and (b) the robotic welding system [25].

Robotic welding and automation have revolutionized the welding industry by enhancing not only accuracy and precision but also efficiency and productivity. The choice of robotic welding type depends on material types, joint configurations, and usages. Arc welding robots create robust joints through electric arcs on metal surfaces and are commonly used in automotive and construction industries. Spot welding robots create joints on metal sheets at specific points. Plasma, laser, and friction welding robots provide high-precision needs. MIG (Figure 6) and TIG welding robots offer high-quality welds, with MIG suitable for automotive applications and TIG for precision in aerospace.

Figure 6.

Robotic MIG welding [27].

Benefits of robotic welding and automation:

1. Robotic welding ensures precision and high-quality welds consistently.

2. Robots work at a steady and efficient pace, which leads to faster production rates.

3. Complex and intricate welding tasks, which are challenging for manual welding, can be easily achieved with robots.

4. Modern robotic welding systems offer easy reprogramming, adaptability to diverse tasks, and scalability to efficiently meet changing production requirements.

5. Robots ensure workplace safety by handling tasks in extreme environments, such as welding fumes, extreme heat, and so on.

6. How AI and ML can revolutionize welding processes

Welding procedures become more efficient and cost-effective when AI and ML with automation and robotics are used. AI and ML systems help in improving welding operations’ safety, dependability, and compliance with industry requirements in different sectors [28, 29, 30, 31], as shown in Figures 7 and 8. There are following ways in which AI and ML provide consistent weld quality:

1. Real-time monitoring: real-time welding process monitoring is made possible by AI and ML. Any change from the standard initiates quick corrections. This ensures that regardless of changing circumstances, welds must adhere to the intended requirements

2. Quality control and defect detection: throughout the welding process, continuous collection of data through real-time sensor data, photos, and videos allows AI and ML systems to learn and improve over time. This feedback loop also allows for quick remedial action by identifying flaws like porosity, fractures, or partial penetration. Overall weld quality is improved by early identification and correction.

3. Data-driven insights and decision-making: AI and ML evaluate enormous volumes of historical welding data, to find trends and connections between welding parameters and weld quality. With the use of these technologies, data-driven decision support is possible, which helps in making wise decisions during intricate and dynamic welding situations (Figure 9).

4. Adaptive control: machine learning algorithms provide consistent weld quality in a variety of circumstances by adapting to different materials, joint geometry, and environmental factors. Robots can also be reprogrammed with ML algorithm to enhance the response, according to change in workpiece alignment and other settings.

5. Welding process optimization: welding parameters like voltage, current, wire feed speed, and gas flow can be optimized using ML models. These algorithms ensure the right parameters to be used at the right level for every welding process by adapting to differences in materials and ambient circumstances.

6. Predictive maintenance: by analyzing the sensor data of welding machines or components, AI can predict when welding equipment may require maintenance or calibration. It prevents quality issues due to equipment malfunction, prevents unexpected breakdowns, and lowers maintenance costs.

7. Welding simulation and training: ML techniques are utilized to generate realistic welding simulations for teaching purposes. These are especially useful for conducting tests on various welding settings in the absence of tangible prototypes and training welders.

8. Resource efficiency: AI and ML can help in optimizing and maximizing the utilization of resources, including energy and other consumables to make welding process more economical, sustainable, and environmentally friendly.

9. Enhanced worker safety: AI and ML help in improving the worker safety and reducing accidents and exposure to danger by tracking working circumstances and giving real-time feedback and alerts.

Figure 7.

Flowchart of process monitoring and control in welding [31].

Figure 8.

System architecture of TIG welding work cell [22].

Figure 9.

Data-driven manufacturing [32].

7. Benefits and challenges of AI and ML in welding

Benefits:

1. AI-driven welding systems detect defects in real-time, that helps in minimizing at the moment.

2. ML algorithms optimize welding parameters, scheduling, and resource usage that improve production time and reduce waste.

3. AI predicts equipment issues based on data and suggests when to provide maintenance, which helps in reducing downtime.

4. AI enhances safety by minimizing human exposure to hazards.

Challenges:

1. AI and ML heavily rely on quality data. Thus, ensuring accurate, comprehensive data collection and data preprocessing is a very important task.

2. Implementing AI and ML can be complex. So, training and expert support for continuous integration is essential.

3. Understanding AI decisions are complicated; thus, transparent AI models must be developed.

4. Reliable sensors and equipment are required for continuous data collection, which require large investment.

5. AI systems must adhere to regulations for compliance with industry standards.