Using the Digital Twin of an Educational Robot Cell During Pandemic

The Challenge

During the 2020 pandemic caused by COVID-19, universities face the problem of how to teach laboratories without using the university installations. At ITAM, there is one specific lab that teaches students how to plan and program a production line using machine tools, robots and conveyors equipped with sensors and actuators controlled by PLC. During the pandemic, we built up a digital twin of the same robotic cell using SIMIO and other simulation tools to provide the experience of planning and improving the production line virtually.

Introduction

During undergraduate studies in engineering, laboratories constitute an important part of the educational process. During the course Robotic Cells at ITAM, students of mechatronic engineering and industrial engineering learn how to plan and implement a complete production line including the programming of PLC and the optimization of the production process.

Besides the conveyor, the robotic cell includes a 6 DoF industrial robot and two CNC machine tools (milling and turning machine tools). The process flow is implemented in SIMIO, and the G-Code to run the CNC machine is generated using PLM NX12.

During the 2020 worldwide pandemic, the university had to be closed and the lectures were taught virtually. This situation led to the challenge of how to provide an adequate lab experience. Since one part of the lab was already planned in SIMIO, it was obvious to integrate robot movements, trajectory planning, and manufacturing via CNC machines into the simulation model.

This case study discusses the digital twin of this specific educational robotic cell and explains the results and lessons learned.

The Solution

Creating the Digital Twin of a Robotic Cell

The robotic cell (RC) at ITAM’s laboratory contains four conveyors in a rectangular arrangement. Each corner has an elevator to move the pallets from one conveyor to the next. The elevator is equipped with a position sensor to detect the presence of a pallet.

At the end of each conveyor, an end stop and a sensor are installed to allow controlled passing to the next conveyor. In the center of the conveyors, the 6 DoF robot is mounted. Two CNC machine tools and a material storage are installed outside around the conveyors, each reachable by the robot.

This scenario is implemented in SIMIO to analyze and optimize the production process of chess figures. Under normal circumstances, students manufacture parts using CNC machines and program the robot to perform required trajectories and record process times. These times serve as input to the SIMIO model.

During the pandemic, the work relied heavily on simulation tools. The G-code for CNC machines was generated using PLM NX12, which estimates machining times through simulation. Although manufacturing was not possible, estimated operation times were used to feed the SIMIO model.

Robot movements were implemented using the robotic toolbox by Corke (2017). Trajectory times were calculated and stored for later robot programming. These times were interpreted as setup and teardown times when mounting workpieces in CNC machines.

The setup, processing, and teardown times were processed via Python into an Excel file, which was imported into SIMIO. The chess figures “TOWER” and “PAWN” were analyzed using the following combinations:

Table 1: Description of model alternatives

The Business Impact

The first variant led to waiting times since the milling process takes approximately three times longer than the corresponding turning process. Shifting some manufacturing operations from milling to turning reduced overall process time and balanced machine utilization.

Variants (2) to (4) include the production of a second chess figure (PAWN), which only requires turning operations and can be processed while the milling machine is running a TOWER operation.

Using multiple simulation tools, students learned how to align a production line using setup, processing, and teardown times during the pandemic. This approach may also reduce ramp-up time once physical lab access is restored.

Future work includes improving time integration using the .NET API provided by SIMIO and PLM NX, rather than relying on external Python scripts.

Proceedings of the 2020 Winter Simulation Conference K.-H. Bae, B. Feng, S. Kim, S. Lazarova-Molnar, Z. Zheng, T. Roeder, and R. Thiesing, eds.

Thomas Martin Rudolf
Luis Antonio Moncayo Martinez

Departamento Académico de Ingeniería Industrial y Operaciones
Instituto Tecnológico Autónomo de México – ITAM
Río Hondo No.1
Col. Progreso Tizapán, CDMX, 01080 MEXICO

References

Corke, P. 2017. Robotics, Vision and Control: Fundamental Algorithms in Matlab; 2nd ed., Springer International Publishing AG.