Automated Assembly Line with 2D Cartesian Robot and Conveyor Belt
Corresponding email: david.pozo@udla.edu.ec
Published at : 28 May 2025
Volume : IJtech
Vol 16, No 3 (2025)
DOI : https://doi.org/10.14716/ijtech.v16i3.7287
Moya, V, Maldonado, M, Vásconez, JP, Abedrabbo, F, Pozo-Espin, D & Moya, M 2025, ‘Automated assembly line with 2D cartesian robot and conveyor belt’, International Journal of Technology, vol. 16, no. 3, pp. 914-932
| Viviana Moya | Faculty of Technical Sciences, International University of Ecuador UIDE, Quito, 170411, Ecuador |
| María Maldonado | Faculty of Technical Sciences, International University of Ecuador UIDE, Quito, 170411, Ecuador |
| Juan Pablo Vásconez | Energy Transformation Center, Engineering Faculty, Universidad Andres Bello, Santiago 7500971, Chile |
| Faruk Abedrabbo | Energy Transformation Center, Engineering Faculty, Universidad Andres Bello, Santiago 7500971, Chile |
| David Pozo-Espin | Ingeniería en Electrónica y Automatización, Facultad de Ingeniería y Ciencias Aplicadas, Universidad de las Américas, Quito, 170513, Pichincha, Ecuador |
| Marcelo Moya | Faculty of Technical Sciences, International University of Ecuador UIDE, Quito, 170411, Ecuador |
Industrial automation is constantly developing and considered critical to process optimization and efficiency. The integration of automated systems is traditionally focused on improving precision and productivity to significantly enhance assembly lines in producing general-purpose models in different materials such as polylactic acid (PLA) plastic. The models include toys, tools, complex mechanism components, and consumer or industrial products. Therefore, this study aimed to present the specifications for the design and testing of a low-cost teaching system for production lines using a 2D Cartesian robot and a conveyor belt. The purpose was to show how innovation in teaching industrial robotics could address the challenge of expensive systems. The proposed solution was to integrate a 2D Cartesian robot into a pneumatic gripper and a conveyor belt controlled by a Programmable Logic Controller (PLC) using a wireless connection and advanced electronic components to achieve efficient communication and operation. The results showed an accuracy of 100% in functional tests and the implications of the advances were to ensure considerable potential for reducing costs and improving learning in industrial and educational environments. The intention was to subsequently promote the implementation of new developments in the field of mechatronics.
Assembly line; Cartesian robot; Mechatronic system; Production line
Control and Structure Assembly The prototype was proposed to be integrated into an assembly line to produce polylactic acid (PLA) plastic models for general use, such as toys, tools, components of more complex mechanisms, and other consumer or industrial products. The design offers a solution to automate the repetitive task of assembling two PLA pieces, male and female, with subsequent classification based on color. The system receives, transports, and locks a male piece with its corresponding female counterpart. 2.1. System Architecture The proposed design consists of a conveyor belt, a 2D robotic arm with a pneumatic gripper attached, a control platform where the operator can interact with the machine, and two platforms on each side of the belt to store the female pieces waiting for assembly. The design picks and places circular, square, and octagonal male pieces of two colors into the female counterparts. The joint 2D robotic arm and the pneumatic gripper were selected due to the advantages of both configurations. For example, cartesian robots moved in straight lines rather than relying on rotational motion (Zhang Bin and Editors, 2017). This linear movement along the two axes offered precise control and predictable paths, making cartesian robots ideal for tasks that demanded high accuracy and straight-line travel, such as assembly and pick-and-place operations. Moreover, Cartesian robots are easy to assemble using commercial guides and bearings on the market and this leads to scalability feasible for a low-cost robot (Rubenstein et al., 2014). Pneumatic grippers offer several advantages in industrial automation which is the reason for the high popularity in handling tasks. Birglen and Schlicht (2018) validated that electronic grippers had better precision and force control in addition to the ease of having a sensor depending on the application. However, pneumatic grippers are lightweight and compact and this leads to cheaper cost and consideration as ideal for a low-cost machine. The application of an appropriate shape can allow the modification of grippers to hold different objects (Cui et al., 2021). Figure 1 shows a simplified machine operation diagram where the user places the male piece on the conveyor belt for subsequent transportation down to reach the color sensor to identify and store the color in memory. The male piece continues through the conveyor belt until it reaches the presence sensor when it is stopped and the robotic arm is activated. The information obtained from the presence sensor allows the gripper of the arm to move into position and graze the piece. The last stage is for the gripper to lift and transport the piece to the platform containing the female counterpart for the joining process to occur. The gripper releases the piece and moves to its resting position until another part has reached the presence sensor. Figure 1 General Sequence of the Proposed Assembly Process 2.1.1. Motor requirements and sizing This subsection estimated the torques the cartesian robot motors required in the X- and Z-axes. Moreover, the Y-axis of the machine corresponded to the conveyor belt, and the torque was analyzed to determine the size of the motor moving the objects. The torque related to the X-axis movement was evaluated using the transmission mechanism of the pulley and belt. The process required determining the force produced at the beginning of the transmission system using the following Equation 1: Where, is the normal force to the X-axis 4.37, is the approximate travel speed 50, d is the pulley center distance 500 mm, g is the gravity 9.81 m/s2, and uk is the friction factor between guide wheels and the aluminum profile 0.31. The minimal force required to move the cartesian robot on the X-axis was 1.35 N. This value was used in addition to the radius of the pulley r set at 6 mm to estimate the torque Tx-axis as 0.008 Nm. The calculation was further used to select a NEMA 17 motor with a torque of 0.5 Nm considered sufficient to produce the horizontal movement. The torque for the movement in the vertical direction of the cartesian robot, Z-axis, was estimated using the force of the power screws. This study used a power screw with an 8 mm diameter (d), a pitch of 2 mm (p), and a feed rate of 8 mm per revolution (fr). The upward movement of the cartesian robot requires the power screw to surpass the weight of the beam with all its components (1.71 kg) and the weight of the pneumatic gripper (0.236 kg). Therefore, the minimum force required was determined using the following Equation (2): The total force to move the cartesian robot in the vertical direction was FZ-axis= 19.09 N which was divided into two because the proposed system had two actuators on the Z-axis, including one power screw per column of the robot. This led to the estimation of the torque for each power screw using Equation (3): Where, FZ-axis/2 is the lifting load 9.54 N, Dm is the average diameter = d – p/2 = 7 mm, dc is the nut diameter = 22 mm, f is the screw friction coefficient 0.19, and fc is the coefficient of friction of the nut 0.08. The results from Equation (3) show that the minimum torque required Tz-axis was 0.028 Nm. This showed that a NEMA 17 motor with a torque of 0.5 Nm satisfied the movement requirements in the Z-axis. Another motor was needed to move the pieces on the conveyor belt on the Y-axis. This led to the determination of the minimum force required to move the belt through three main aspects, including the need for the motor to move the weight of the belt, the rollers, and the pieces. The power required by the motor to move the conveyor belt depended on several aspects. First, the force required by the conveyor belt to carry was calculated using the weights of a female piece and the PVC belt as follows: Female pieces characteristics: Transport piece mass = 18.27 g. Transport weight = 0.01827kg . 9.81 m/s2 = 0.18 N. Belt characteristics: Belt weight= 2.5 kg/m2. Belt width = 38 mm. Belt length = 0.794 m. The characteristics of the belt were used to determine the total area which was subsequently used to calculate the total weight as 0.76 N. Moreover, the friction and stretching forces required to be overcome by the belt to produce the movement were calculated using Equations (4) and (5), respectively: Where, µ is the friction factor 0.25, m is the transport mass 0.093 kg, and g is the gravity 9.81 m/s2. Therefore, the value of the friction force was found to be fr = 0.22 N. The elongation force for the belt at 1% stretch is 8 N/mm according to the manufacturer. The belt length was 0.794 m and 1% of the length was 0.00794 m The total load on the system was determined by adding the values of the forces as follows: The travel speed v was 10 mm/s and this showed that the power required was The torque necessary to rotate at a speed of 10 mm/s with a radius of gyration of 9 mm of the drum was used to determine the angular speed as 110 rpm (11.51 rad/s). Therefore, the necessary torque was Ty-axis = 7.11 w/ (11.51 rad/s) = 0.618 Nm which led to the selection of a Pololu-type 37D motor with power specifications of 10 w and a torque of 0.98 Nm. 2.2. Control and Communication of System Components The control system of the prototype was managed by electronic components that received information from different sensors of the machine to execute tasks properly. The following subsections explain the electronic components used, their communication protocols, and the control system. Moreover, the algorithm followed by the machine during its operation is presented in Algorithm 1. Electronic Components Electronic components were used in the prototype to effectively control the mechanical parts during the execution of tasks. The system was managed by a PLC as the central core. Two drivers were connected to the PLC for the purpose of providing control to the 2D cartesian robotic arm equipped with the pneumatic gripper. The other components connected were the presence sensors and limit switches to provide boundaries to the movement of the conveyor belt and the robotic arm. Another critical component connected was the ESP-32 controller and the primary function was to regulate the DC motor of the conveyor belt and the color sensor. Meanwhile, the PLC was connected to the HMI screen in order to provide essential information about the status of the machine and the current task for the operator. The entire system is internet-connected through a router in order to enable remote access and management. Figure 2 shows the connection of all electronic parts used in the model while Figure 3 summarizes the electronic hardware and the main specifications. Algorithm 1 Functionality of the System 2.2.2. Communication The communication interfaces for each system component are shown in Figure 4a). It was observed that each part of the machine interacted with the remaining system during the execution of tasks as follows: The machine connects to the Internet via an Ethernet connection between the Router and the Siemens PLC. A second Ethernet cable connects the PLC with the HMI Screen to allow user interaction and monitoring. This communication assigns an IP address to each component and provides a secure and well-performing bus to interchange data and information about the execution of tasks. On the control side, the PLC communicates with the ESP-32 Controller via Wi-Fi. The method is highly efficient despite its potential unreliability for long-distance communication because the PLC is in close proximity to the EPS-32 controller. The proximity enables fast data exchange from the color sensor and the DC motor moving the conveyor belt without generating incorrect data. The limit and presence sensors are important in detecting the presence of objects or movement boundaries within different components and are directly interfaced with the PLC. These sensors transmit essential information, which the PLC processes and translates into commands. The commands are subsequently relayed to the motors of the robotic arm and the pneumatic gripper, both directly controlled by the PLC. Figure 2 Prototype's Electronic Diagram for Control, Actuators and Sensors Figure 3 Electronic components used 2.2.3. Conveyor Belt Control The proposed system presented in Figure 4b) allows the control of speed movement for the conveyor belt by the operator before the execution of tasks. The desired speed parameter is received as input and generates an output speed value using the data provided by the DC motor controller, its driver, and the motor. Meanwhile, error is provided as feedback to the system in order to decrease the difference between the actual and set speed. The Proportional–Integral–Derivative (PID) controller of the conveyor belt was tuned in steps by combining the Ziegler-Nichols method and the TIA PORTAL environment as well as the theoretical direction and empirical fine-tuning of the control system. The first aspect was to use the Ziegler-Nichols method to set up the baseline values of the parameters. The proportional gain (Ku=0.08) and proportional time (Tu=4) were determined by turning the controller into proportional-only control and increasing the gain up to the period oscillations were continuous. These values were applied to calculate the initial PID parameters, including Proportional Gain (Kp): 0.6×Ku, Integral Time (Ti): 0.5×Tu, and Derivative Time (Td): 0.125×Tu which were subsequently used to fine-tune under TIA PORTAL. The software enabled the setting of the control parameters such as the process, output, and PWM limit while the default values of the proportional gain time constant and derivative action time were assigned. The process further allowed fine-tuning using real-time responses to the system in order to determine a critical damping factor as well as the little overshoot and undershoot. The empirical tuning process was achieved by setting the controller values as follows: • Proportional gain: 0.05 • Integral action Time: 2.2 • Derivative action time: 0.1 These final tuning values provided a satisfactory and critically damped response which offered an acceptable degree between response and stability. The values were also assessed for further accuracy under different load levels to allow the parameters to work optimally. The control system developed in Figure 4 c) shows the control curve of the conveyor belt and the performance tests in terms of the response to the Cartesian robot. The graph shows the critically damped response of the system which is characterized by the absence of both overshoot and undershoot. The process variable (PV) closely tracks the setpoint (SP) to precise and efficient response of the system to varying operating conditions. Meanwhile, the control variable (CV) refers to the parameter adjusted to influence the behavior of the PV to ensure it remains as close as possible to the desired SP. The CV is the output signal from the PID controller and the critically damped behavior shows that the control parameters have been properly tuned. The trend further presents the stability and precision achieved in the implementation of the mechatronic system. The figure also shows the real-time control plot as visualized on the HMI screen. a) b) c) Figure 4 a) General communications connections, b) Conveyor belt close loop control, and c) Conveyor belt control, showing the established setpoint (SP), the process variable (PV), and the control variable (CV) Table 1 presents information on the characteristic parameters of the conveyor belt control curve measured empirically using a stopwatch. The settling time is the period necessary for the PV to reach and maintain its reference value (SP) within an acceptable range. The response or rise time represents how quickly the system reaches 90\% of the total variation from its initial position to stabilization. The absence of overshoot and undershoot is the precision of the response provided by the system. Meanwhile, the peak time represents the instant at which the maximum value of the response is reached. These empirical values provide a quantitative assessment of control performance on the conveyor belt. The definition of the entire design of the machine and selection of the required components were followed by the cutting of the structural profiles and assemblage of the machine. The entire assembly process of the structure and connections of the components were developed in-house to form the line presented in Figure 5. The system incorporates physical control elements, including buttons for start, stop, and emergency, along with stop and start indicator lights and messages, to ensure effective management by the user. The following video https://youtu.be/9Jm9xLJmX5U shows a complete cycle to produce the ensemble of male and female parts in order to have a better understanding of the system sequence. The female and male pieces can be white or black and this shows the possibility of two positions in the male section, including one for black and another for white. Figure 6 a) shows that the first step of the process is to place the male pieces in their respective station. Moreover, the entrance of the conveyor belt was unidirectional, and nominal speed was used for the transport of female pieces as presented in Figures 6 b) and c). The color sensor QTR-3A, which is an infrared reflectance sensor array, detects the color of the female piece entering the conveyor belt. This is necessary to ensure the Cartesian robot selects a male piece of the same color as shown in Figure 6 d). The Cartesian robot subsequently transported and placed the male piece on top of the female located at the end of the conveyor belt, as presented in Figure 6 e). The male-female assembly was also transported to the exit station as shown in Figure 6 f). Table 1 Specification for Time Domain Figure 5 CAD and main components of the prototype In terms of mechanical design, a Cartesian robot was implemented to stack the male and female pieces, and the final effect was achieved using a pneumatic gripper for precise manipulation. The Cartesian robot used NEMA 17 motors and limit switches to operate. Moreover, the inclusion of capacitive position sensors on the conveyor belt assisted in detecting when a piece arrived at the color detection module and the moment it reached the end. The module had a graphical user interface (HMI) that provided informative messages about the process, control of variables, and visualization of the manipulation. Furthermore, the variability in the color of female pieces was considered and addressed in the design to ensure accurate detection. The module also satisfied maximum dimensions of 750x350x700 millimeters, leading to the integration into the pneumatic maintenance unit and equipped with quick couplings to connect to the compressed air system. This was necessary to ensure efficient operation and easy maintenance. Figure 6 Operating process: (a) Placing white and black male pieces on the station, (b) placing a female piece at the entrance of the conveyor belt, (c) female piece reaches the end of the conveyor belt, (d) gripper picks up the male piece of the same color to be inserted into the female piece, (e) robot places the male piece on top of the female, and (f) the robot takes the set to the exit station. System Validation and Results Experiments were conducted in controlled laboratory conditions to verify the functionality of the prototype towards ensuring consistent and reliable performance. The focus was on the capacity of the system to classify two colors, black and white, as well as three distinct plastic forms, including circular, square, and octagonal shapes. This was in addition to the corresponding female counterparts at three speeds of the conveyor belt, including 40 RPM, 80 RPM, and 120 RPM. The detailed list of the parameters and conditions used during the tests is stated as follows. Ten tests were conducted for each shape and speed to have a total of 180 runs. The tests were conducted in a stable environment with the ambient temperatures controlled at approximately 22°C to minimize the impact of environmental variability on the sensors and mechanical components. The workspace was uniformly illuminated to ensure fluctuating light levels on the prototype sensors did not affect color detection. The calibration of sensors as well as mechanical and electronic components were monitored and maintained to ensure the only variables used were shape, color, and speed. The results were uniform throughout multiple experiments for each condition and this showed minimal to negligible fluctuation in completion times and sensor activations. Moreover, all discrepancies, including sensor inactivation or missing classifications, were documented. The results presented in Table 2 showed that the system attained a success rate of 100% for circular shapes in all test iterations. The trend showed the stability and reliability of the system at each evaluated speed. Table 2 Validation Test Results The variance in completion time for the tests performed was determined through the application of boxplots as shown in Figure 7. Each boxplot corresponds to the completion time of each shape for a specific speed and color. Moreover, the shape of the pieces was a common factor for all speeds tested. Circular pieces required less completion time than the other two variants, without considering the color, due to the mechanical design of the gripper. The circular shape ensured an easier grasp of the piece while the corners in the other two led to some difficulty in picking which caused longer completion times. The results for the 80 RPM presented in Figures 7c) and d) showed that a faster conveyor belt led to lower completion times as expected. For this case, white pieces required less time to complete than black and the candles showed that most of the tests were conducted at times considered lower than average. The plots in Figures 7 e) and f) show the results for the fastest speed of 120 RPM and the completion time decreased but was less than the previous test. The circular shape achieved the shortest time of all tests by finishing the assembly in 24.14 seconds. The white square pieces also produced errors on five tests because one of the presence sensors did not activate and the task was not completed. Only 17 out of the 180 tests were not completed due to the inactivation of the second presence sensor. This problem manifested mainly in the square and octagonal shapes while the circular shapes did not experience any difficulty. Figure 7 a) Results for 40 RPM speed - Black Pieces, b) Results for 40 RPM speed - White Pieces, c) Results for 80 RPM speed - Black Pieces, d) Results for 80 RPM speed - White Pieces, e) Results for 120 RPM speed - Black Pieces, and f) Results for 120 RPM speed - White Pieces The module achieved a remarkable 100% accuracy rate in operational tests. The conveyor belt showed a severely damped control curve which led to excellent precision and stability in assembly line operations. The ability of the module to assemble an average of two pieces per minute was an indication of its high efficiency which led to optimism about the capacity of the system to satisfy production requirements. An in-depth analysis was conducted to compare the shape and color with the performance of the machine. Figure 8 provides information on how shape and color affect the system and Figure 8 a) shows how white pieces tend to achieve the task in less time than black pieces. This could be due to the process used to tune the sensors to easily detect lighter colors which led to a faster reaction from the machine. Figure 8 b) shows that a circular shape achieves better completion time than the others. This is due to the ease of grabbing and placing into the female counterparts by the gripper because of the round form. There was no concern about plane sides or inadequate match which in turn could have caused a delay in the overall operation. Circular shapes fit in any position while the slight variation in position of square and octagonal pieces is capable of causing problems. a) b) Figure 8 a) Average Completion Time for White and Black Pieces. b) Average Completion Time for Circular, Square, and Octagonal Pieces Machine Cost and Comparison Table 3 presents the approximate costs associated with the production of the machine, including the materials, mechanical and electronic components, manufacturing, assembly, and design hours. The total expenditure for the machine was USD 3546 and the profit expected from the sales was USD 650 which led to the setting of the final price at USD 4196. The comparison with similar machines in the industry which typically range from $5000 to USD 10000 (Semi-Automatic Screw Machine/Custom-Made Machinery/ Automatic Assembly Machine, n.d.) shows the solution proposed offers a significant cost advantage. The specific cost-benefit advantage is approximately 16.08% to 58.04% which represents substantial savings of more than 50% of the cost, thereby showing the value and competitiveness of the machine in the market. Table 3 Table of costs
7.94 mm. Therefore, the stretching force was calculated using the following Equation (5):
= 6.464 w. Moreover, the consideration of 10% losses showed that the real power P was 7.11 w.
2. System Assembly and Initial Functional Tests
In conclusion, the proposed automated assembly machine demonstrated adequate performance due to the benefits of PLC control and ESP32-based communication networks. The PLC system provided precise operational management, ensuring efficient synchronization between the Cartesian robot and conveyor belt, thus enhancing productivity and responsiveness to production variability. Furthermore, the integration of the ESP32 improved system functionality by enabling real-time data exchange and remote monitoring, allowing engineers with immediate access to process insights and data. The testing phase confirmed a remarkable 100% accuracy rate in piece picking and classification. However, increased motor speeds introduced variability in completion times, notably influenced by piece color and shape. At lower speeds (40 RPM), black pieces exhibited lower time dispersion compared to white pieces, although this distinction diminished at higher speeds (80 and 120 RPM), suggesting that sensor identification time could contribute to observed variances. Additionally, the Human-Machine Interface (HMI) significantly enhanced operational usability by providing intuitive, real-time data visualization, fostering informed decision-making, and promoting innovation in automated mechatronic systems. The comprehensive cost analysis underscored the economic viability of the proposed solution, presenting savings between 16.08% and 58.04% compared to commercial alternatives. Future studies should address testing under industrial conditions and extensive use to validate performance and broaden the applicability of this automation system across diverse industrial contexts.
The authors express acknowledgment to Universidad Internacional del Ecuador (UIDE) for providing the resources and support necessary for the completion of this project. The study was also supported by ANID (National Research and Development Agency of Chile) under Fondecyt Iniciación 2024 Grant 11240105.
| Filename | Description |
|---|---|
| R1-ME-7287-20250526143753.docx | --- |
Ahmad, S, Ali, S & Tabasha, R 2020, ‘The design and implementation of a fuzzy gain-scheduled PID controller for the Festo MPS PA compact workstation liquid level control’, Engineering Science and Technology, an International Journal, vol. 23, no. 2, pp. 307–315. https://doi.org/10.1016/j.jestch.2019.05.014
Arjun, P & Mirnalinee, TT 2016, ‘Machine parts recognition and defect detection in automated assembly systems using computer vision techniques’, Rev. Téc. Ing. Univ. Zulia, vol. 39, no. 1, pp. 71–80
Ashari, H, Yusoff, YM, Zamani, SN & Talib, AN 2018, ‘A study of the effect of market orientation on malaysian automotive industry supply chain performance’, Journal of Technology, pp. 1651-1657, https://doi.org/10.14716/ijtech.v9i8.2749
Asrol, M 2024, ‘Industry 4.0 adoption in supply chain operations: A systematic literature review. International Journal of Technology, vol. 15, no. 3, pp. 544–560, https://doi.org/10.14716/ijtech.v15i3.5958
Belda, K & Jirsa, J 2021, ‘Control principles of autonomous mobile robots used in cyber-physical factories’, 2021 23rd International Conference on Process Control (PC), pp. 96–101, https://doi.org/10.1109/PC52310.2021.9447468
Bendowska, K & Zawadzki, P 2023, ‘Development and verification of a simulation model of an automated assembly line’, Applied Sciences, vol. 13, no. 18, p. 10142, https://doi.org/10.3390/app131810142
Birglen, L & Schlicht, T 2018, ‘A statistical review of industrial robotic grippers’, Robotics and Computer-Integrated Manufacturing, vol. 49, pp. 88–97, https://doi.org/10.1016/j.rcim.2017.05.007
Burggräf, P, Wagner, J, Dannapfel, M, Fluchs, S, Müller, K & Koke, B 2019, ‘Automation decisions in flow-line assembly systems based on a cost-benefit analysis’, Procedia CIRP, vol. 81, pp. 529–534
Carlson, JS, Spensieri, D, Wärmefjord, K, Segeborn, J & Söderberg, R 2014, ‘Minimizing dimensional variation and robot traveling time in welding stations’, Procedia Cirp, vol. 23, pp. 77–82
Chakraborti, T, Isahagian, V, Khalaf, R, Khazaeni, Y, Muthusamy, V, Rizk, Y & Unuvar, M 2020, ‘From robotic process automation to intelligent process automation: –emerging trends–‘, Business Process Management: Blockchain and Robotic Process Automation Forum: BPM 2020 Blockchain and RPA Forum, Seville, Spain, September 13–18, 2020, Proceedings 18, pp. 215–228, https://doi.org/10.1007/978-3-030-58779-6_15
Chen, H, Zhang, G, Wang, J, Eakins, W & Fuhlbrigge, T 2008, ‘Flexible assembly automation using industrial robots’, 2008 IEEE International Conference on Technologies for Practical Robot Applications, pp. 46–51, https://doi.org/10.1109/TEPRA.2008.4686671
Chutima, P 2023, ‘Assembly line balancing with cobots: An extensive review and critiques’, International Journal of Industrial Engineering Computations, vol. 14, no. 4, pp. 785–804, https://doi.org/10.5267/j.ijiec.2023.7.001
Costa, R, Sousa, V, Silva, FJG, Campilho, R, Sá, JC, Pinto, A & Pereira, J 2022, ‘Increasing the sustainability of manufacturing processes in plastic injection: recovering out-of-service robots to eliminate manual assembly operations’, Sustainability (Switzerland), vol. 14, no. 19, https://doi.org/10.3390/su141912300
Cross, N 2021, Engineering design methods: strategies for product design, John Wiley & Sons
Cui, Y, Liu, X-J, Dong, X, Zhou, J & Zhao, H 2021, ‘Enhancing the universality of a pneumatic gripper via continuously adjustable initial grasp postures’, IEEE TRANSACTIONS ON ROBOTICS, vol. 37, no. 5, https://doi.org/10.1109/TRO.2021
De Carlo, F, Arleo, MA & Tucci, M 2014, ‘OEE evaluation of a paced assembly line through different calculation and simulation methods: a case study in the pharmaceutical environment’, International Journal of Engineering Business Management, 6(Godište 2014), pp. 6–27
Dimény, I & Koltai, T 2023, ‘Comparison of MILP and CP models for balancing partially automated assembly lines’, Central European Journal of Operations Research, https://doi.org/10.1007/s10100-023-00885-x
Eriawan, R, Alfan, DA & Susi, K 2024, ‘Development of efficient and scalable production process of analytical grade sodium chloride at laboratory scale’, International Journal of Technology, pp. 743-752, https://doi.org/10.14716/ijtech.v15i3.5606
Garduno-Aparicio, M, Rodriguez-Resendiz, J, Macias-Bobadilla, G & Thenozhi, S 2018, ‘A multidisciplinary industrial robot approach for teaching mechatronics-related courses’, IEEE Transactions on Education, vol. 61, no. 1, pp. 55–62, https://doi.org/10.1109/TE.2017.2741446
Giberti, H, Abbattista, T, Carnevale, M, Giagu, L & Cristini, F 2022, ‘A methodology for flexible implementation of collaborative robots in smart manufacturing systems’, Robotics, vol. 11, no. 1, p. 9, https://doi.org/10.3390/robotics11010009
He, T, Li, D & Yoon, SW 2017, ‘A heuristic algorithm to balance workloads of high-speed SMT machines in a PCB assembly line’, Procedia Manufacturing, vol. 11, pp. 1790–1797
Hu, SJ, Ko, J, Weyand, L, Elmaraghy, HA, Lien, TK, Koren, Y, Bley, H, Chryssolouris, G, Nasr, N & Shpitalni, M 2011, ‘Assembly system design and operations for product variety’, CIRP Annals, vol. 60, no. 2, pp. 715–733, https://doi.org/10.1016/J.CIRP.2011.05.004
Hultman, E 2022, ‘Introducing robotized stator cable winding to rotating electric machines’, Machines, vol. 10, no. 8, https://doi.org/10.3390/machines10080695
Jupir, J, Aziz, KA & Hassan, H 2023, ‘Determinants of successful collaborative project management: insights from malaysian construction industry’, International Journal of Technology, vol. 14, no. 6, pp. 1344–1353, https://doi.org/10.14716/ijtech.v14i6.6651
Kudryavtseva, T & Skhvediani, A 2020, ‘Effectiveness assessment of investments in robotic biological plant protection’, International Journal of Technology, vol. 11, no. 8, pp. 1589–1597, https://doi.org/10.14716/ijtech.v11i8.4528
Lee, J-D, Li, W-C, Shen, J-H & Chuang, C-W 2018, ‘Multi-robotic arms automated production line’, 2018 4th International Conference on Control, Automation and Robotics (ICCAR), pp. 26–30, https://doi.org/10.1109/ICCAR.2018.8384639
Li, S, Zhu, F & Wu, Q 2024, ‘A robust texture-less gray-scale surface matching method applied to a liquid crystal display tv diffuser plate assembly system’, Applied Sciences (Switzerland), vol. 14, no. 5, https://doi.org/10.3390/app14052019
Ma’ruf, A, Nasution, AAR & Leuveano, RAC 2024, ‘Machine learning approach for early assembly design cost estimation: a case from make-to-order manufacturing industry’, International Journal of Technology, vol. 15, no. 4, pp. 1037–1047, https://doi.org/10.14716/ijtech.v15i4.5675
Moreira, B, Gouveia, RM, Silva, FJG & Campilho, R 2017, ‘A novel concept of production and assembly processes integration’, Procedia Manufacturing, vol. 11, pp. 1385–1395, https://doi.org/10.1016/j.promfg.2017.07.268
Moya, V, Quito, A, Pilco, A, Vásconez, JP & Vargas, C 2024, ‘Crop detection and maturity classification using a YOLOv5-based image analysis’, Emerging Science Journal, vol. 8, no. 2, pp. 496–512, https://doi.org/10.28991/ESJ-2024-08-02-08
Natesan, N, Jegadeeshwaran, R, Sakthivel, G & Arunkumar, V 2023, ‘Design and analysis of robot and fixture pedestal for OTC FD-V8 MAG welding robot’, Journal of Physics: Conference Series, 2601, https://doi.org/10.1088/1742-6596/2601/1/012008
Neitmann, M, Rothe, T, Kappel, E & Hühne, C 2022, ‘Assembly of compliant structures with autonomous industrial mobile manipulators (AIMM) using an end effector with active deformation compensation for the assembly of flaps’, Machines, vol. 10, no. 5, p. 291, https://doi.org/10.3390/machines10050291
Novandra, RP & Putri, AA 2022, ‘Improvement design of packaged juice production process using business process reengineering’, International Journal of Technology, pp. 1851-1861, https://doi.org/10.14716/ijtech.v15i6.5630
Osejo Sornoza, AG & Berrú Pontón, YM 2020, Diseño e implementación de un módulo didáctico para el control y aprendizaje de manejo de dos brazos robóticos a escala de RV2AJ Y SCARA, Universidad Politécnica Saleasiana, Ingeniería Electrónica
Paucar Tenorio, JA & Quinllay Ramos, GL 2014, Diseño, dimensionamiento e implementación de una estación didáctica de presionado con músculo neumático, Escuela Superior Politécnica del Chimborazo, Facultad de Informática y Electrónica
Pires, M, Couto, P, Santos, A & Filipe, V 2022, ‘Obstacle detection for autonomous guided vehicles through point cloud clustering using depth data’, Machines, vol. 10, no. 5, p. 332, https://doi.org/10.3390/machines10050332
Rashid, MFF, Hutabarat, W & Tiwari, A 2012, ‘A review on assembly sequence planning and assembly line balancing optimisation using soft computing approaches’, The International Journal of Advanced Manufacturing Technology, vol. 59, pp. 335–349, https://doi.org/10.1007/s00170-011-3499-8
Robeller, C, Weinand, Y, Helm, V, Thoma, A, Gramazio, F & Kohler, M 2017, ‘Robotic integral attachment, Fabricate 17, Rethinking Design and Construction, vol. 3, pp. 92–97
Rubenstein, M, Ahler, C, Hoff, N, Cabrera, A & Nagpal, R 2014, ‘Kilobot: A low cost robot with scalable operations designed for collective behaviors’, Robotics and Autonomous Systems, vol. 62, no. 7, pp. 966–975, https://doi.org/10.1016/j.robot.2013.08.006
Ryalat, M, Franco, E, Elmoaqet, H, Almtireen, N & Al-Refai, G 2024, ‘the integration of advanced mechatronic systems into industry 4.0 for smart manufacturing’, Sustainability (Switzerland), vol. 16, no. 19, https://doi.org/10.3390/su16198504
Semi-Automatic Screw Machine/Custom-Made Machinery/ Automatic Assembly Machine, (n.d.), https://shorturl.at/9IU6P
Tigse Soto, SB 2017, Diseño y construcción de un módulo didáctico con el plc s7-1200 para la simulación de variación de presión de un fluido, Ecuador: La Mana: Universidad Técnica de Cotopaxi; Facultad de Ciencias de la Ingeniería y Aplicadas
Wang, X & Li, J 2013, ‘Detecting communities by the core-vertex and intimate degree in complex networks’, Physica A: Statistical Mechanics and Its Applications, vol. 392, no. 10, pp. 2555–2563, https://doi.org/10.1016/j.physa.2013.01.039
Xie, X 2023, ‘Industrial robot assembly line design using machine vision’, Journal of Robotics, https://doi.org/10.1155/2023/4409033
Yousif, I, Burns, L, El Kalach, F & Harik, R 2024, ‘Leveraging computer vision towards high-efficiency autonomous industrial facilities’, Journal of Intelligent Manufacturing, https://doi.org/10.1007/s10845-024-02396-1
Zhang Bin, D & Editors, W, (n.d.), Lecture Notes in Mechanical Engineering Mechatronics and Robotics Engineering for Advanced and Intelligent Manufacturing, http://www.springer.com/series/11236