April 1, 2023

A few years ago, I was tasked with a 10 day project to reduce cycle time on a respot cell. The cell featured 2 Fanuc 210F robots on 8 meter RTU's. The robots were equipped with tool changes and 2 SPR guns. At 341 seconds, this cell was very over cycle. I had offered many strategies that could be implemented in the 10 day time frame, with a risk to reward ratio. Some of the low risk items were payload estimation, path optimization, and motion enhancement. Medium to high risk items included process resequencing, zoning, and integrated rail setup. The customer had identified which items we would tackle in our timeframe and the project was a go.

In the first few days, the low risk items were taken care of. Payload optimization was successful. Although it did not benefit the goal of cycle time, it was a crucial item to protect the longevity of the robotic arm and prevent nuisance alarms.  Path optimization such as speed and approximation values were implemented to reduce moving time and smooth the flow of motion from point to point. Motion was enhanced by removing redundant positions and changing motion types to move efficiently. Although these items were not timed, there was a noticeable improvement during dayshift production.

A Fanuc Roboguide virtual simulation workcell was created during production hours to continue working on the process improvements while the cell was unavailable. This made the tedious and long challenge of different sequence scenarios possible and sped up the implementation time once a solution was proven out. Zoning and motion planning could also be done. This action was crucial in the success of the project overall given not only the short timeframe of the project, but the downtime where I along with the maintenance team were able to work on the cell and implement the cycletime reduction measures.

By the end of the second week, it was evident that the project was very successful. The process went from slow and staggered to smooth and fluid. The robots were almost always moving, except for minor zoning or toolchange process'. The cycle time went from 341 seconds to 285 seconds - a 56 second reduction. This meant that the customer could put out another 16.5 units in just an 8 hour shift.

There were many more strategies that could have been implemented in this project. Due to the time required to convert programs and a lack of experience at the time, the integrated rail was never implemented. Minor weld resequencing including balance between robots could have been done to get more time out of the process.  Devicenet quick connect function could have been setup for faster toolchanges. An extra toolchange could have been removed out of Robot 1. Most importantly, there were 2 other models that run where all of these items could have been implemented, but it was deemed not important due to the lower volume that those products run.

This was one of the most memorable projects in my career. For me, process efficiency is definitely the best part about any automation project. It is always fun pushing the limits of the designed product and implementing new strategies to get the lowest time possible. I hope more of these projects come my way in the near future.