Autonomous robots carry products between neighboring companies

TRECO manufactures postal equipment, such as advanced community mail boxes, from sheet metal in Aalestrup, Denmark. It cooperates with another SME, J-Coating, which provides coating services next door. When TRECO has finished welding and grinding the seams of its mail boxes, J-Coating applies powder coating to them. Later TRECO continues with final mechanical and electrical assembly.

The bulky and heavy boxes are transported between workstations and storage areas and between the storage areas of each facility. It is, however, impossible to enter the storage area of J-Coating while the coating process is running. This is why boxes are transported between facilities only by night.

Boxes are moved some 12 km each day, half of which is done by night. This is done with specially designed carts holding eight boxes each and weighing close to 300 kg.

“It is difficult to find staff willing to sacrifice the best part of an evening for a tedious and repetitive job hauling heavy boxes. Even during the day, we would like to free our specialized workers from hauling boxes, to increase employee satisfaction and productivity,” says Lars V. Jørgensen, CEO of TRECO. They decided to test automation in the L4MS experiment.


Workers oversee the process at home

The Danish Foundation Autonomous set to work on a solution that would automate the transport of boxes between TRECO and J-Coating. They replaced the manual work with an AGV (Automated Guided Vehicle).

In the solution, workers use their tablets to summon the robot and to request it to start transporting. By night, they oversee the process on their tablets at home.

“During the installation, interviews with employees inspired us to additional solution tweaks. For example, our robot now collaborates with the operator at the grinder station when loading the cart. The robot, or actually the cobot, moves forward while the operator uses his overhead crane. Besides improving ergonomics, this saves two minutes per cart, which makes 50 minutes per day,” says Henrik L. Egestad, the Project Director at Autonomous.

The solution is based on the OPIL platform and it includes modifications made by Autonomous. As they were not allowed to use the MiR robot’s own computer, they developed their own versions of RAN and SAN to enable communication with the robot. They also cleaned up the HMI provided by OPILs, so that it now holds no other information than Get and Go buttons.


Mutual understanding with the help of 3D simulation

For 3D simulation, Autonomous used the Visual Components software.

“It gave us a lot of insight into the project needs. Not only did it bring a mutual understanding of challenges and potentials, but it also gave us instant feedback and facts on proposed changes,” Henrik Egestad says.

Experts at Autonomous designed the outline of the solution in four weeks, but the more detailed design and simulation took longer, as the logistics set up was quite complicated with separated facilities involved. The final installations and tests were executed in about three weeks.

The cost of implementation amounted to about 120.000 EUR, and the cost of the robot to some 55.000 EUR, including necessary peripherals, such as a hook and an automatic charging station. Autonomous does not include software development in these calculations, as a considerable share of the work was not dedicated to this project only.

Once fully implemented, the solution is estimated to

  • save man hours by at least 20%
  • increase production capacity by 10%.


Robots producing production data

Another goal of the experiment was to convert robot actions into real-time production data, which would help companies to share data and optimize production. The framework was designed, but not implemented.

“We believe, however, that the overall principle of converting robot actions into valuable production data can be applied in many other companies,” Henrik Egestad says. According to him, the automation solution itself won’t be applicable as such in other companies, as the use case including two companies is unique.


OPIL modules and other modules used in the TRECO/J-Coating solution by Autonomous:

  • The core of the OPIL platform
  • OPIL Human Machine Interface with modifications
  • OPIL RAN and SAN with modifications
  • 3D simulation tool Visual Components to create a digital twin


Partners in the L4MS experiment:

  • Factories: TRECO A/S and J-Coating
  • Solution provider: Foundation Autonomous


Information source: L4MS


TRECO A/S and J-Coating

Manufactures postal equipment


Solution provider: Foundation Autonomous