We were very happy to receive expert assistance in the context of the NWE Interreg Project Machining4.0. We now have more accurate information to consider when we make decisions regarding important production processes involving human-robot collaboration.

Alejandro Muñoz Espiago, Director of Technology at Steripack

In Industry 4.0, collaborative robots may play an important role in production execution, where they perform various tasks next to a human operator.  A collaborative robot arm mounted on top of a mobile platform is called a collaborative mobile manipulator. A collaborative manipulator can perform various tasks, such as pick and place, sorting, restocking inventory, and many other tasks that require the robot to be static and move around on the shop floor. These additional features, when compared to a traditional robotic arm, can help SMEs to improve their production efficiency as well as to save costs as these collaborative manipulators come with integrated safety features and functionalities. Therefore, a collaborative mobile manipulator can be used in the manufacturing process to meet a set of diverse requirements.

Problem Statement

Packaging is one of the essential processes that can be seen in most manufacturing industries. However, most of the packaging is a semi-automated or fully manual process involving human operators. This leads to an increased cycle time. Therefore, this project aims to investigate the potential to automate the assembly of box type packages and thereby estimate the cycle time of more automated processes. One of the main criteria for this project was to reduce cycle time and increase the box packaging output by at least 150%. In order to achieve this, various configurations of the process were tested using an advanced simulation environment as well as a real mobile robotic platform. Furthermore, a cost analysis for the proposed solution using a collaborative manipulator was analysed to estimate the feasibility from the SME’s point of view.

Task Description

For this experiment, a human-robot collaboration scenario (HRC) involving a box package similar to the one shown in Figure 1 was chosen.  The box was made of cardboard and was stacked in a dispenser vertically. The task was broken down into various sub task. These sub-tasks involve picking the box up from the initial dispenser to completing the operation. As this is an HRC scenario, the sub-tasks involve both human and the robot working together. The box packaging task was considered as an assembly line balancing problem, and different configurations were tested. At present, this operation is carried out manually, with a number of operators performing the task.

From the stack, a box was taken out and then unfolded, as seen in Figure 1. After this process, few items such as a pamphlet, a small case and a wooden stick were placed inside the box. Finally, the box was closed and stored away in a container. The process was carried out manually by an operator, and later the same was tested out using a collaborative mobile manipulator.

 

Special Requirements:

While testing with the collaborative mobile robot, special fixtures shown in Figure 2A were designed to aid the robot in unfolding the box shown in Figure 2B. Figure 2C shows a linear actuator used to unfold the box in order to reduce the cycle time.

 

 

Result

Two sets of experiments were carried out. Firstly, a test using the collaborative arm was executed. Then, process simulation environments were used to optimise the process in order to achieve the desired results.

Physical Experiment

During the initial test with the robot, the outcome of the experiment was not as expected. The time is taken to finish the task per box was significantly higher than the manual operation by 170%. The increase in the time was mainly due to the:

• Process sequence
• Allocation of resources to the sub-tasks
• Robot speed
• Untrained operator carrying out the manual task

Simulation Experiment 

In order to optimise the process, a series of modifications were made to the sub-task, such that the required cycle time was achieved. The changes were mainly:

• Combining two subsequent tasks to reduce the cycle time and reducing the robot wait times
• Including a second collaborative robotic arm
• Using a linear actuator to the special fixture as shown in Figure 2C

Based on the experiments carried out in the simulation environment, it was observed that the process time was reduced to the desired outcome. Nevertheless, this involves two collaborative robotic arms and the same number of human operators in the production line. By this, the required cycle time is achieved, thereby increasing the output capacity by 150%. However, again, this is purely based on the simulation.

Remarks

The return on investment for the two collaborative manipulators, the gripper and the special fixture, is approximately 15 months, subjected to the market demand, which is assumed to be 100%. When the demand drops to 75%, it takes 30 months to recover the return on investment. 

Conclusion

In conclusion, the initial experiments with the real robot showed a significant increase in the cycle time compared with the current manual assembly process. With further testing and optimisation in the simulation environment, two collaborative manipulator configurations with the same number of human operators involved in the task were able to bring down the cycle time and increase the output by 150%. This optimised layout tested in the simulation environment is yet to be tested with the real robot and the layout configuration.

This report provides an initial summary of the potential impact of involving collaborative manipulators for packaging application. Though there are various aspects to consider, one of the main things to look for is the potential bottleneck in the process that involves a lack of training for the human operator to perform the task. Secondly, the special fixture design can be modified such that the time can be reduced further.

From an economic point of view, the SME might have to invest a significant amount of money in procuring the collaborative robotic arms, grippers and the special fixture. Furthermore, it also requires modification of the layout configuration to accommodate the new devices. But, on the other hand, unlike the traditional industrial robots in a cell, the collaborative manipulators can be moved around easily within the production line and can be integrated into a different process with ease.

[1]    “Personalised window Rectangular Toothpaste boxes with eye-catching colors and printed information entice customers.” https://www.emenacpackaging.com/customized/toothpaste-boxes/.