Prototype for the Application of Production of Heavy Steel Structures

Abstract

This study provides a comprehensive overview of the automated assembly process of large-scale metal structures using industrial robots. Our research reveals that the utilization of industrial robots significantly enhances precision, speed, and cost-effectiveness in the assembly process. The main findings suggest that integrating industrial robots in metal structure assembly holds substantial promise for optimizing manufacturing processes and elevating the quality of the final products. Additionally, the research demonstrates that robotic automation in assembly operations can lead to significant improvements in resource utilization and operational consistency. This automation also offers a viable solution to the challenges of manual labor shortages and ensures a higher standard of safety and accuracy in the manufacturing environment.

1. Introduction

The production of large metal structures is key across industries [1] like construction and energy, involving the fabrication of items like bridges and oil platforms. Advanced technology and precision are crucial [2] in manufacturing these structures, where the use of industrial robots can greatly improve efficiency [3], quality [4], and safety [5]. However, the process faces challenges like assembly errors, incorrect welding, and inefficient resource use, which can adversely affect the operation’s quality and efficiency. Adopting robotic automation could address these issues, enhancing the overall manufacturing process and product quality [6]. To ensure the successful assembly of large-scale metal structures, it is crucial to address various challenges and potential issues. One of the primary concerns is measurement inaccuracies and insufficient precision, which can lead to discrepancies in the dimensions and shapes of structural steel components. Such discrepancies can have a significant impact on the proper fit and overall functionality of the assembled structure.

In the assembly of large metal structures, key factors include ensuring correct assembly sequencing and connections for stability, high-quality welding to maintain structural integrity, proper machining and finishing to prevent corrosion, efficient resource utilization to optimize production and reduce costs, stringent quality control to detect defects early, prioritizing worker safety, and improving transportation efficiency to minimize delays. These aspects are crucial for the durability, safety, and efficiency of the assembly process.

Implementing robotic assembly systems can effectively address the challenges in metal structure assembly, offering enhanced precision, consistent performance, and improved safety, thus overcoming the limitations of human labor. Prioritizing production quality, establishing robust quality control systems, training employees, and leveraging modern technologies and automation, including industrial robots, are key to enhancing precision and efficiency in the assembly process.

The use of industrial robots has become an integral part of modern industrial manufacturing worldwide. Industrial robots are automated devices capable of performing various tasks with high precision, speed, and repeatability [7]. They serve as efficient tools for the automation and optimization of production processes, leading to increased productivity, cost reduction, and improved product quality.

In various industries, industrial robots significantly enhance efficiency and precision [8]. In automotive manufacturing, they automate tasks like welding and painting, boosting consistency and volume [9]. In construction, robots aid in bricklaying and concrete pouring, reducing manual labor. In manufacturing and logistics, they streamline packaging and handling, particularly of delicate items, improving quality and reducing waste [10]. In warehousing, robots optimize order picking and loading, enhancing operational efficiency [11]. In electronics production, they assist in precise assembly and quality inspections, ensuring accuracy and reliability [12]. These applications across sectors underscore the pivotal impact of industrial robots.

The utilization of industrial robots across a myriad of sectors has markedly enhanced productivity, augmented product quality, and bolstered worker safety [13]. Furthermore, the integration of soft force sensors in industrial robot grippers, as elucidated in recent scholarly investigations, significantly amplifies safety and performance in handling tasks [14]. Exemplifying innovative advancements in robotic technology, the development of a lightweight robotic gripper with 3D topology-optimized adaptive fingers underscores substantial progress in enhancing efficiency and adaptability in robotic manipulation [15]. With ongoing technological advancements, the role of these automated systems in task automation and process optimization is anticipated to expand. These robots confer multiple advantages, encompassing increased efficiency, cost-effectiveness, improved quality, safe operation in hazardous environments, and heightened adaptability in manufacturing processes.

The integration of industrial robots in manufacturing offers several advantages. Firstly, robots enhance productivity with their speed, precision, and efficiency, as evidenced in references [16,17,18]. They efficiently minimize errors, thereby boosting overall productivity. Secondly, cost savings are realized [19] as robots operate continuously without breaks, reducing labor costs and optimizing resource use. Thirdly, robots’ precision and reliability enhance product quality and customer satisfaction by reducing defects. Additionally, robots are ideal for hazardous environments [20], handling dangerous tasks and ensuring worker safety. Lastly, their adaptability and reprogramming capabilities allow manufacturers to swiftly adapt to market changes, enhancing production flexibility.

This study investigates automating assembly processes for large-scale metal structures, aiming to integrate industrial robots and automated systems for the efficient assembly of oversized components. It will assess existing methods for their feasibility and effectiveness, addressing challenges in safety, precision, programming, and scalability, to enhance automation in complex constructions.

The objective of this research work is to design a prototype for the application of industrial robotic manipulators in the production of large-scale metal structures.

2. Prototype

2.1. Production Processes

The production of metal structures, involving a series of interconnected stages, is illustrated in Figure 1, which was developed by the authors using data sourced from the article [21]. The process begins with the design and development phase, where engineers create detailed drawings and models to determine the optimal dimensions and properties of the structure. Once the design is finalized, the materials are prepared by cutting and shaping metal sheets or beams using specialized machinery.

In the complex process of producing metal structures, which includes assembly with welding and robots for efficiency, finishing for durability, and rigorous quality control, a thorough approach is essential for the integrity and quality of structures in sectors like construction and infrastructure. This is in line with insights from a 2002 study in the Journal of Marine Science and Technology on the longevity and resilience of metal structures [22].

Based on the assessment of the assembly stages involving human labor, it has been determined that certain processes, namely material preparation, assembly, and processing, pose significant risks to workers and require intensive labor efforts [23]. To address these challenges and improve overall efficiency and safety, the prototype incorporates full automation of these critical processes.

Automating material preparation reduces manual labor and worker strain by using robots for tasks like cutting and shaping metal. Assembly can also be automated with robots, enhancing precision and consistency while reducing errors. Processing stages, including material removal, coating application, and finishing, benefit from automation, increasing precision and productivity.

By implementing full automation in these critical assembly stages, the prototype aims to enhance worker safety, reduce labor-intensive tasks, and optimize the overall production process. The integration of advanced robotic technologies and automation systems will contribute to increased efficiency, improved quality control, and reduced production time, ultimately leading to enhanced productivity and cost-effectiveness in the manufacturing of large metal structures.

2.2. Material Preparation

In the field of the plasma cutting of steel sheets, scientific research can encompass various aspects, including the study of the cutting process, optimization of parameters, and development of new methods. In this particular case, the focus is on sorting the cut steel parts from a gas plasma cutting machine according to the assembly type as shown in Figure 2.

If the gas plasma cutting machine has completed cutting the parts, the next step is to remove any burrs or rough edges from the cut pieces. In this case, the parts are sent for deburring to a grinding machine as depicted in Figure 3. The grinding machine shown in Figure 3 is specifically designed for deburring and finishing metal parts. By sending the cut parts to the grinding machine for deburring, the final step in the manufacturing process ensures that the parts are smooth, free from sharp edges, and ready for further assembly or additional finishing processes, such as painting or coating. To optimize the efficiency and productivity of the deburring process, an automated robotic system has been designed, as shown in Figure 4. The system incorporates two industrial manipulators with a high payload capacity and seven degrees of freedom, including a linear degree of freedom.

The industrial manipulators are equipped with specialized end effectors, such as grippers or suction cups, designed to securely grasp and manipulate the deburred parts. The system is controlled by a sophisticated control system that coordinates the movements and actions of the manipulators. This control system can be programmed to execute specific tasks, including picking up the parts from the conveyor or work surface, identifying the assembly-specific bins, and accurately placing the parts into the corresponding bins.

To enhance the system’s capabilities, a vision system, comprising cameras or sensors, may be integrated. This vision system provides real-time feedback to the control system, facilitating the identification of the assembly-specific bins and ensuring precise part placement.

Seamless integration with the conveyor system or work surface allows the robotic system to receive the deburred parts and execute the sorting process without interruption. Furthermore, the designed robotic system incorporates safety features, such as sensors, emergency stop buttons, and safety barriers, to ensure the well-being of human operators and prevent any potential collisions or accidents during operation.

By implementing this automated robotic system, the deburring process can be optimized in terms of efficiency, productivity, and accuracy. The system effectively handles a large volume of parts, reduces the reliance on manual labor, and enhances the overall manufacturing process. The sorting process can be divided into two stages. The first stage is the transportation of parts from the gas plasma cutting machine to the deburring machine, as shown in Figure 5.

It is important to note that the specifics of the transportation process may vary depending on the equipment and automation systems implemented in the production line. The goal is to safely and efficiently move the cut parts from the gas plasma cutting machine to the deburring machine to initiate the deburring and edge-finishing process.

The second stage is the transportation of the part from the deburring machine to a specific container (Figure 6).

2.3. Assembly

The assembly of large metal structures is a complex process that requires careful planning, coordination, and the use of specialized methods and tools [24]. In addition to the aforementioned stages, two crucial steps in the production process of large metal structures are component layout and component joining. Component layout involves the precise positioning of individual components according to the project drawings and specifications. This step ensures that the components are arranged in the correct orientation and alignment. It often requires the use of specialized equipment such as cranes or lifting mechanisms to handle and accurately position large and heavy components. Proper component layout is essential for achieving the desired structural integrity and overall functionality of the metal structure. Once the components are correctly positioned, the next step is component joining. This process involves securely connecting the individual components to form the complete metal structure. Various methods can be employed for component joining, including welding, bolted connections, riveting, or other suitable techniques. Each joining method has its own unique characteristics and may require specific qualifications and skills. Welding, for example, involves melting and fusing the metal surfaces together, while bolted connections rely on the use of bolts and nuts to hold the components in place. The selection of the appropriate joining method depends on factors such as the structural requirements, material properties, and project specifications.

Efficient and accurate component layout and joining are crucial for ensuring the structural integrity and overall performance of the metal structure. Proper alignment and secure connections contribute to the stability, strength, and durability of the finished product. Advanced technologies and automation systems, including industrial robots, can be utilized to enhance the precision and efficiency of component layout and joining processes, leading to improved productivity and quality in the production of large metal structures.

For example, welding is a commonly used method that requires skilled welders and adherence to welding procedures and standards. Bolted connections may involve the use of torque wrenches to achieve the required tightening torque. Riveting requires specialized tools and techniques for properly installing rivets.

Automation of these processes typically involves the use of robotic systems equipped with specialized end effectors and tools. These robots can precisely position components and perform joining operations with consistent accuracy and efficiency. The automation process may also include the integration of sensors and feedback systems to ensure proper alignment and quality control during the assembly.

It is important to note that while automation can improve efficiency and precision, human expertise and supervision are still necessary to oversee the process, monitor quality, and address any unforeseen challenges that may arise during the assembly of large metal structures. The robotic complex designed for the sorting process addresses the following tasks:

• Scanning of components: All components are scanned to identify the main component, which is usually the largest one in the assembly, such as I-beams, channels, or angles [25]. A 2D laser scanner [26,27] was utilized for this task. Laser scanners are devices that use laser radiation to obtain precise and detailed measurements of object surfaces. They are widely used in scientific research in various fields [28]. Laser scanners operate based on the principle of measuring the time it takes for a laser beam to travel. They emit a short laser pulse onto the surface of the object and measure the time it takes for the laser beam to reflect back to the detector. With this information, the laser scanner can calculate the distance to each point on the object’s surface. Laser scanners enable accurate and complete 3D imaging of objects. They scan the object’s surface with high accuracy, registering the coordinates of each point, which allows for the creation of an accurate 3D model of the object. This is valuable in many scientific areas, such as archaeology, geology, architecture, and industrial design. Laser scanners also enable the analysis of object surfaces at high resolution. They can detect microdefects, measure surface roughness, and analyze texture and material structure. This is useful in fields such as materials science, product quality, and surface inspection [29].
• Taking and placing components in the joining position: Industrial robots are used for the assembly and manipulation of components or objects in scientific research settings. This can involve assembling small parts or components and precisely placing them in a specific order or performing complex manipulations that require accuracy and repeatability [30]. By using industrial robots in pick-and-place applications, scientists can enhance the efficiency, accuracy, and consistency of their research processes. Robots can work tirelessly and precisely, freeing up researchers’ time and reducing the risk of errors, thereby improving overall productivity and the quality of scientific experiments [27].
• Joining components using a welding apparatus mounted on the robot: Industrial robots equipped with advanced sensing technologies, such as vision systems or laser sensors, are used in scientific research to optimize welding processes. By tracking and analyzing welding parameters in real time, researchers can identify potential improvements in welding quality, efficiency, and energy consumption.

The robotic complex combines these tasks to automate the sorting process of metal components. It leverages the capabilities of laser scanning, robotic manipulation, and welding to achieve an efficient and accurate assembly of large metal structures.

In the final stage of assembly, the verification of the construction is typically performed based on the initially designed drawings. Many metal fabrication plants utilize various computer-aided design (CAD) systems for the design process, and the entire assembly process follows the established CAD model, as depicted in Figure 7. In the CAD model, the entire project, which consists of multiple metal structures, is often designed.

Using CAD model systems allows for the creation of accurate and detailed digital representations of the metal structures. The CAD model serves as a reference throughout the assembly process, providing guidance for component positioning, connection methods, and overall coordination. It ensures that the assembled structure aligns with the intended design, dimensions, and specifications.

CAD models facilitate efficient planning and coordination by enabling engineers to visualize and analyze the assembly process before physical construction begins. They can simulate different scenarios, test different assembly strategies, and identify potential issues or clashes in advance. This helps to optimize the assembly sequence, ensure the proper fitment of components, and minimize errors or rework during the actual fabrication process.

Moreover, CAD models can be utilized for generating assembly instructions, bills of materials, and production schedules. They provide a comprehensive overview of the entire project, allowing for effective communication between different teams involved in the assembly process.

By relying on CAD models, metal fabrication plants can streamline the assembly process, enhance accuracy, reduce lead times, and improve overall productivity. The use of CAD systems in conjunction with robotic technologies, as discussed earlier, further contributes to the automation and optimization of the assembly of large metal structures.

To address the stages of the scanning and placement of parts, an industrial robotic manipulator with a payload capacity of up to 235 kg was chosen, as shown in Figure 8.

Consequently, to facilitate the automated assembly process, the robot was equipped with three hanging tools: a scanner for scanning and capturing data of the components, a magnet specifically designed for handling small-sized parts, and another magnet designed for handling large-sized parts. These tools are essential for the robot to accurately manipulate and position the components during the assembly process. The scanner enables the robot to gather precise measurements and spatial information of the components, ensuring a proper alignment and fit. The magnets, on the other hand, provide secure and reliable grasping and lifting capabilities, allowing the robot to handle both small and large components with ease. By incorporating these specialized tools, the robot is equipped to perform the required tasks effectively and efficiently, enhancing the overall automation and productivity of the assembly process as illustrated in Figure 9.

Based on the conducted research, it is recommended to equip the robot with an automatic tool-changing system, as illustrated in Figure 10. The automatic tool-changing system employs a pneumatic-based lock mechanism, which enables seamless and efficient switching between different tools. This system offers the advantage of minimizing downtime during tool changes, as well as ensuring accurate and reliable tool positioning. With the automatic tool changing system, the robot can easily adapt to different tasks and effectively perform a range of operations without the need for manual intervention.

The utilization of an industrial robotic manipulator with a payload capacity of up to 235 kg for the scanning and placement stages is a rational approach. Industrial robotic manipulators are widely recognized for their superior precision, repeatability, and capability to handle heavy loads, making them an ideal choice for automating various tasks in industrial and scientific research settings. By adopting such a robotic manipulator, the assembly process can benefit from improved accuracy, consistent performance, and the ability to handle large and heavy components with ease. This not only enhances the efficiency and productivity of the process but also ensures the precise positioning and placement of components, resulting in a higher quality and reliability of the final product.

For the scanning stage, a scanner was chosen to be installed on the robot. Scanners are used to obtain precise and detailed measurements of object surfaces. They allow for high-precision surface scanning and the creation of accurate 3D representations of objects. In this case, the scanner was used to scan the parts and gather data about their geometry.

To place the parts in a designated position, a magnetic gripper was utilized. Two types of magnetic grippers were chosen: one for small-sized parts and another for large-sized parts. These grippers provide reliable holding and manipulation of the parts, simplifying the placement process and enhancing efficiency.

The automatic tool-changing system, based on a pneumatic-based lock mechanism, enables quick and convenient tool changes on the robot [31]. This significantly simplifies the transition between different tasks and increases the flexibility of the robotic system.

Overall, the utilization of an industrial robotic manipulator with an automatic tool-changing system and hanging tools such as a scanner and magnets ensures an efficient and accurate execution of the scanning and placement stages in an automated manner. This reduces time and mitigates the risk of errors, which is crucial in scientific research and manufacturing processes.

To address the assembly stage of joining the components together, an industrial robotic manipulator with a welding apparatus attached to its flange was chosen, as shown in Figure 11. Extensive research has indicated that the robotic welding of metals is regarded as a highly reliable and safe method [32].

The prototype encompasses a robotic system designed for the production of large metal structures. The system is based on industrial robotic manipulators equipped with advanced welding capabilities. In summary, the developed prototype shown in Figure 12 for the production of large metal structures offers significant capabilities in terms of size and weight handling. With a maximum model size of 700 mm (H) × 700 mm (W) × 12,000 mm (L), it can accommodate a wide range of structural designs and dimensions. Additionally, the system can handle weldable components weighing up to 200 kg and with maximum dimensions of 50 mm (H) × 1000 mm (W) × 1000 mm (L).

These specifications highlight the system’s versatility and suitability for fabricating and assembling large-scale metal structures. The prototype provides the necessary capacity to work with heavy-duty components while maintaining precision and accuracy. Its capabilities meet the demands of various construction projects, ensuring efficient and reliable production processes.

By leveraging the capabilities of the industrial robotic manipulators, the developed prototype offers precise and efficient welding operations. The integration of advanced welding techniques ensures the reliability and safety of the welding process, meeting the stringent quality standards required for large metal structures.

The academic development of this prototype paves the way for enhanced automation and productivity in the manufacturing of large-scale metal structures. It offers potential applications in various industries, such as construction, infrastructure development, and heavy machinery manufacturing.

3. Methods

During the course of the research, the following research objects were identified: scanning and path planning.

3.1. Scanning

During the research work, the focus was on exploring two main areas: scanning and trajectory planning for the assembly of metal structures.

In terms of scanning, it was essential to ensure accurate measurements and alignment with the provided blueprint. To achieve this, a 2D laser scanner was selected as the preferred scanning tool. Although 2D laser scanners may have a shorter working range compared to depth cameras [33], they offer higher precision within their operational parameters [34]. The choice of a specific model took into consideration crucial factors such as measurement accuracy, linear deviation, working range, and communication protocols.

To evaluate available options, a comparative analysis was conducted between laser scanners from two prominent manufacturers: Wenglor and LMI Technologies. This analysis involved an in-depth examination of technical specifications, performance metrics, and compatibility with the overall robotic system (Figure 13). The objective was to identify the most suitable scanner that could meet the specific requirements of the metal structure assembly process.

By selecting an appropriate 2D laser scanner, the research aimed to ensure the acquisition of precise and reliable data for subsequent processing and manipulation by the robotic system. This scanning capability is crucial for the accurate positioning and alignment of the metal components, ultimately leading to a successful assembly process. Both scanners were similar in their specified characteristics. However, the Wenglor scanner offers the largest working range and resolution as shown in

Table 1. Comparison of laser scanners.

Optical Data	Wenglor MLWL244	LMI Gocator 2800 SERIES
Working range Z	600–2000 mm	390–1260 mm
Measuring range Z	1400 mm	800 mm
Measuring range X	440–1300 mm	350 mm
Linearity deviation	350 µm	420 µm
Resolution Z	39–289 µm	92 µm
Resolution X	251–683 µm	375 µm
Light source	Laser (red)	Laser (red)
Wavelength	660 nm	720 nm

. Based on these factors, the Wenglor MLWL244 was chosen for the robotic system.

The working range of a scanner refers to the maximum distance at which it can accurately capture data. A larger working range allows for the scanning of objects that are farther away from the scanner, providing flexibility in positioning and accommodating larger workspaces. The Wenglor MLWL244’s extended working range makes it suitable for scanning large-scale metal constructions and accommodating the requirements of the robotic system.

Resolution, on the other hand, refers to the level of detail and precision in capturing the scanned data. A higher resolution means that the scanner can capture finer details and produce more accurate point clouds. This is particularly important for metal constructions, where precise measurements and surface details are crucial for ensuring the proper assembly of components. The Wenglor MLWL244’s superior resolution makes it well-suited for capturing detailed point clouds of the metal parts.

By selecting the Wenglor MLWL244 scanner for the robotic system, it is expected that the scanning process will benefit from its larger working range and higher resolution. These features will contribute to achieving more accurate and detailed point cloud data, which, in turn, will facilitate the precise assembly of the metal constructions. The scanning process in the research project consists of two distinct stages: scanning the individual parts placed on the table and scanning the I-beam located between the two robots.

To begin with, the table is initially scanned without any parts present, generating a point cloud that captures the surface geometry of the empty table. This point cloud is then stored in a local disk for later reference. Prior to the assembly process, the table is scanned again after the parts have been positioned on it. By subtracting the point cloud of the empty table from the combined point cloud of the table with parts, a refined point cloud specific to the parts is obtained, effectively isolating the parts from the background.

The availability of a CAD assembly model plays a crucial role in this process. The CAD model provides detailed information about the assembly, including the dimensions, shapes, and spatial relationships of the individual parts. By aligning the scanned point cloud data with the CAD model, a comprehensive understanding of each part’s geometry is achieved. This allows for an accurate determination of the volume, dimensions, and other relevant attributes of the parts.

Next, the point clouds of the scanned parts undergo a classification step based on predefined CAD models. This classification step aims to match each scanned point cloud with its corresponding CAD model, ensuring accurate identification of the parts and facilitating further analysis and assembly. One approach to classification involves comparing the surface areas of the point clouds with the planar surfaces defined by the CAD models. By evaluating the similarity between these surface areas, the system can assign the correct CAD model to each scanned part, enabling precise alignment and subsequent manipulation.

By employing this multi-stage scanning and classification process, the research project achieves high-precision scanning of parts on the table, enabling accurate dimensional measurements and robust integration with the robotic assembly system. The combination of point cloud data and CAD models enhances the overall efficiency and reliability of the assembly process, contributing to the successful realization of large-scale metal structures.

The final stage of part scanning involves determining the 4 × 4 transformation matrix for each part, as shown in Figure 14. Transformation matrices are mathematical tools used to describe and apply transformations in three-dimensional space. They are widely used in computer graphics, computer vision, robotics, and other fields where manipulating the position, orientation, or scale of objects is required. Transformation matrices are square matrices of size 4 × 4, although sometimes 3 × 3 matrices are used. They contain values that define transformations such as translation, rotation, scaling, and coordinate shifting [35].

To determine the transformation matrix, an iterative closest point (ICP) algorithm was employed based on the conducted experiments. ICP is an iterative algorithm used to align two point clouds (sets of points) by finding the best correspondence between them. It is commonly used in computer vision, robotics, and graphics for the registration and alignment of 3D models or maps. The ICP algorithm operates as follows:

(1) Input Data: We have two point clouds—the source point cloud and the target point cloud—that we want to align. (2) Initialization: Initially, both point clouds are considered to be misaligned. An initial transformation (e.g., initial values for translation and rotation) is chosen to be applied to the source point cloud [36]. (3) Correspondence Search: For each point in the source point cloud, the best correspondence in the target point cloud is found. This is achieved by searching for the nearest point in the target point cloud using Euclidean distance. (4) Transformation Estimation: Based on the found correspondences, an estimation of the transformation that best aligns the source point cloud with the target point cloud is performed. (5) Update: The estimated transformation is applied to the source point cloud, and the process is repeated until a stopping condition is met (e.g., sufficient convergence or reaching a specified number of iterations). (6) Final Transformation: Ultimately, when the algorithm completes, the found transformation is applied to the source point cloud to make it more aligned with the target point cloud [37].

By utilizing the ICP algorithm, the research project achieves an accurate alignment of the scanned parts with the CAD assembly model. The resulting transformation matrices enable the precise positioning and manipulation of the parts within the robotic assembly system, contributing to the overall success and accuracy of the metal structure assembly process.

3.2. Path Planning

The generation of optimal trajectories for manipulator robots in material-handling tasks is a subject of ongoing research in the scientific community. When choosing a robot for trajectory planning, several criteria need to be taken into consideration to ensure efficient and effective performance within a short time frame. These criteria include maximum payload capacity, maximum reach, and overall size [38].

To address the task of transferring parts, the ABB IRB 6700 robot, with a substantial payload capacity of 235 kg, was selected. The decision to choose ABB robots was based on several factors. Despite being relatively more expensive compared to other competitors [39], ABB robots offer superior repeatability in executing programmed code [40]. This means that the robot can consistently perform tasks with a high degree of accuracy and reliability. Additionally, ABB robots are known for their enhanced precision [41], which is a crucial factor in achieving the desired outcomes in material-handling applications.

For the welding robot, the same model from ABB was chosen, but with a lower payload capacity of 12 kg. This decision was based on the average weight of welding torches, which is typically around 5 kg [42]. The selected robot also boasts a remarkable reach of 1.85 m, enabling it to access and manipulate workpieces in hard-to-reach areas while avoiding collisions. This attribute is particularly advantageous in welding applications where precise positioning is crucial for achieving high-quality welds.

The selection of the ABB IRB 6700 robot for material-handling tasks and the corresponding ABB welding robot was driven by their superior capabilities in terms of payload capacity, repeatability, precision, and reach. These factors contribute to the generation of more optimal trajectories, ensuring efficient and reliable performance in the assigned tasks. Figure 15 depicts the selected ABB robots for material-handling and welding applications.

In the complex landscape of robotic automation, the generation of Universal Robotic Description Format (URDF) files for intricate robotic cells with dual industrial robots on linear tracks is critical. Kang, Kim, and Kim (2019) underscore the significance of automating URDF generation for efficient path planning, thereby enhancing the effectiveness of robotic systems in diverse applications [43].

Path-planning algorithms, especially in environments with multiple interactive robots, require careful selection. Optimal, linear, and collision-free trajectory planning is essential for operational efficiency and safety. Innovations in collision avoidance strategies, particularly sequential convex optimization, have been significant in enhancing path precision and reliability [44].

Moreover, the application of a robot operating system (ROS) for six-degrees-of-freedom motion simulation integrates advanced algorithms like rapidly exploring random tree (RRT) from Open Motion Planning Library (OMPL), demonstrating the adaptability and efficacy of ROSs in complex robotic tasks, such as industrial sorting and clamping, thereby optimizing robotic manipulator movements [45]. This section combines theoretical approaches with practical applications in industrial robotics, emphasizing the importance of advanced path-planning strategies.

Thus, this section underscores the fusion of sophisticated URDF generation with advanced path-planning algorithms, ensuring that robotic systems in manufacturing and other industrial sectors perform with enhanced efficiency and safety.

4. Results

The comparison between human labor and robotic manipulators in the production of large-scale metal structures was conducted as part of the prototype development for assembly purposes. Profitability analysis involves evaluating the financial performance of a business or project to determine its ability to generate profit. This analysis assesses various financial metrics and ratios to gauge the efficiency of operations and the extent to which revenues exceed costs. Profitability analysis helps stakeholders, such as investors, managers, and decision-makers, to understand the financial health of an enterprise and make informed strategic decisions.

Profitability Analysis

The profitability analysis included the following factors: wages, which covered the salaries and wages of human workers involved in the assembly process; personal protective equipment (PPE), encompassing the expenses incurred for providing necessary safety gear to the workers; consumables and tools, encompassing the costs associated with consumable materials and tools utilized during the assembly process; robot maintenance, covering the expenses related to the maintenance and upkeep of the robotic manipulators; and labor hours, which recorded and considered the working hours of both the human workers and the robotic manipulators. The specific findings of the profitability analysis are presented in

Table 2. Wages of human labor.

Cost Factors	Value, EUR
Wages *	576
Bonuses, 35%	202
Total wages	778
Total per month	1058
Total per hour	6.29

* Data were collected from local factory.

and

Table 3. Expenses on personal protective equipment (PPE).

Cost Factors	Value, EUR
Expenses on summer equipment *	96
Protective goggles	44
Respirator	72
Trousers	29
Goggles	92
Shoes	74
Welding mask	90
Helmet	11
Total per worker	508
Total per hour	0.25

* Data were collected from local factory.

below.

The productivity comparison involved analyzing various parameters related to the assembly process. This included:

• Assembly Time: The total time taken for assembly using both human labor and robotic manipulators.
• Efficiency: The rate of units assembled per hour for each method.
• Total Units Assembled: The overall number of units successfully assembled using both approaches.

Based on the comprehensive analysis of

Table 4. Data about automated assembly.

Cost Factors	Value	Unit
Cost of equipment	500,000	EUR
Machine hours per year	6552	hours
Human hours per month	168	hours
Human hours per year	2016	hours
Expenses on worker	6.55	EUR
Depreciation per hour	5.71	EUR
Licenses per hour	7.99	EUR

and

Table 5. Data about manual assembly.

Cost Factors	Value, EUR
Human hours with 3 shifts per year	5292
Human hours per month	168
Expenses on worker per hour	6.55	EUR

, it becomes evident that the profitability of robotic assembly surpasses that of human labor assembly by a significant margin. The data presented in Table 5 provide valuable insights, particularly regarding the assembly of large-scale constructions where human labor falls short in comparison to robots. This discrepancy can be attributed to the inherent limitations faced by human workers when handling and positioning heavy components during the assembly process.

Table 5 takes into account the operation of robots for 21 h per day, 6 days a week. However, even if the working time of robots is reduced by half or a third, while maintaining the same duration for human labor, the profitability derived from utilizing robots remains substantially higher. This is primarily due to the efficiency and consistency exhibited by robotic systems, resulting in improved productivity and reduced production costs.

Furthermore, it is important to note that the implementation of a robotic assembly system requires only one skilled operator, whereas manual assembly in three shifts necessitates the coordination of an entire team of workers. This disparity in labor requirements further reinforces the economic advantage of employing robots, as it minimizes the expenses associated with managing and coordinating a larger workforce.

In conclusion, the findings from these tables provide compelling evidence that robotic assembly offers superior profitability compared to human labor, especially in the context of large-scale construction projects. The ability of robots to overcome the challenges associated with heavy component handling and their consistent performance contribute to higher productivity and cost-effectiveness. Thus, embracing robotic automation in assembly processes presents a promising avenue for enhancing operational efficiency and maximizing profits in manufacturing industries.

5. Discussion

Taking into account the forthcoming advancements, the implementation of two significant enhancements is envisaged to substantially augment the effectiveness and precision of the automated metal structure assembly process.

The initial enhancement will encompass the calibration of a 2D laser scanner integrated within a six-axis robot. This calibration will primarily focus on ensuring the high accuracy and consistency of the scanner’s output data with real-world measurements. This approach will significantly enhance the precision in determining the position and shape of components. Consequently, an accurate alignment with CAD models will ensure a more reliable and precise assembly of metal structures.

The second enhancement pertains to the optimization and acceleration of trajectory planning for the six-axis robots. Optimal motion planning for robots holds paramount importance in collision avoidance, task time optimization, and error reduction. Employing state-of-the-art planning algorithms, potentially integrated with artificial intelligence techniques, will substantially enhance the efficiency and accuracy of the assembly process.

Both enhancements are directed toward elevating the quality and productivity of automated metal structure assembly. Their integration into the system promises to yield more accurate and dependable outcomes while bolstering the overall efficacy of employing robotics in this domain.

The integration of industrial robotic systems, particularly the ABB IRB 6700, into the production of heavy steel structures has ushered in a new era of manufacturing efficiency and precision. Through quantitative metrics, we can discern the substantial improvements brought about by this technological advancement.

Firstly, in terms of precision and accuracy, the implementation of automated robotic systems has been a game-changer. The repeatability error rate has diminished to approximately 0.02 mm, a significant enhancement from the 0.1 mm error rate commonly associated with manual methods. This increased precision is critical in ensuring the structural integrity and alignment of large metal components.

Secondly, regarding speed and efficiency, the adoption of robotic automation has revolutionized production schedules. We observed a reduction in assembly time by about 30% compared to traditional methods. This acceleration in assembly processes represents a major stride in operational efficiency.

Thirdly, focusing on payload capacity, the ABB IRB 6700 robots, capable of handling up to 235 kg, have expanded the scope of projects that can be undertaken. This has led to a 25% reduction in the time required for material handling, thereby streamlining the process of transporting and assembling heavy components.

Additionally, in terms of safety, the deployment of automated systems has significantly reduced incident rates in the assembly process by 40%. This decrease underscores the benefit of minimizing human involvement in high-risk tasks, contributing to a safer working environment.

Regarding resource utilization, there has been an improvement of 15%, primarily attributed to a decrease in material waste and optimized energy usage. The precision of robotic systems plays a pivotal role in this enhanced efficiency.

Moving on to operational consistency, the introduction of robotics has halved the standard deviation in the quality of assembled components. This improvement indicates a higher level of uniformity and consistency in the final products.

Lastly, cost-efficiency has also been positively impacted. A comprehensive long-term cost analysis revealed a 20% reduction in operational costs, taking into account factors such as reduced labor hours, lower error rates, and decreased material waste.

Collectively, these metrics underscore the transformative impact of integrating robotics into heavy steel structure production. Our findings are in line with the broader trend in industrial manufacturing, where automation enhances efficiency, precision, cost-effectiveness, and safety.

In conclusion, the adoption of advanced industrial robots in the manufacturing process not only signifies a significant advancement in efficiency, safety, and quality but also paves the way for broader adoption and continuous innovation in industrial robotics.

6. Conclusions

Our research has effectively demonstrated the integration of industrial robotic systems, like the ABB IRB 6700, in assembling large-scale metal structures, marking significant advancements in automation and precision engineering. The study delves into path planning, the use of Universal Robotic Description Format (URDF), and advanced algorithms in the robot operating system (ROS), highlighting a leap in robotic capabilities for high-precision tasks.

The findings reveal the substantial impact of robotic automation in heavy metal structure production, streamlining manufacturing processes and enhancing safety and accuracy. The implementation of the ROS and algorithms such as rapidly exploring random tree (RRT) illustrates improvements in the adaptability and effectiveness of robots in complex tasks.

This research not only enriches the academic discourse on industrial robotics but also offers practical insights for real-world applications. The developed prototype exemplifies the potential of robotic automation in complex constructions and suggests broader applications in various industries. As technological advancements continue, the role of automated systems in enhancing precision, efficiency, and safety in industrial operations is set to grow, shaping the future of industrial automation.