What changed when they automated
What changed when a Virginia fabricator put a robot on grinding 1,000 trays a day
Lawrence Brothers of Bluefield, Virginia grinds 1,000 metal material handling trays a day, a job that drove turnover and sent eight to ten trays a day back for rework. A FANUC arm with scanning software that generates each tray's grind path now does it. The vendor reports production up 38 percent and rework down 95 percent.
The short answer
A heavy fabricator moved a tiresome, injury prone grinding step onto an industrial robot whose software scans each incoming tray and plans the grind path, so the part type can change every cycle. The vendor publishes a 38 percent production increase, a 95 percent drop in rework and 2,000 dollars a week saved on grinding discs. The account below is the maker's own case story, so read every number as the vendor published it, not as an audited result.
What the job looked like before
Every one of about 1,000 trays a day needed grinding. The work drove high turnover in a region with a limited workforce, and eight to ten trays a day went back for rework. The company has more than 100 employees and 80,000 square feet per the vendor; a state manufacturing extension partnership case study describes about 60 workers at the plant, so published headcount varies by source and date.
What they put in
A FANUC M-710iC/70 six axis robot, 70 kilogram payload and 2,050 millimeter reach, deployed by GrayMatter Robotics, a FANUC authorized integrator, running GrayMatter's scanning and path planning software. Displaced grinders moved into robot operating roles. Abrasive sequence, force control hardware, dust extraction and fixturing were not published.
What changed, in their numbers
Production up 38 percent. Rework down 95 percent, stated as a drop from eight to ten trays per day to eight to ten per month. Material cost savings of 2,000 dollars a week on grinding discs. The cell reached target rates in its first week. Baseline output of 1,000 trays a day. All published by the vendor. Payback, cell cost and cycle time were not published.
What this does not tell you
It does not publish the cell cost or payback. It does not say what the abrasive sequence or the extraction system looks like, which are large line items in any grinding cell. And it is a high volume, single family application; the scanning approach is built for variation within a family, not for a job shop's random mix.
What to verify in your own plant
Your rework rate on the finishing step, because that is the number this account moved most. Your abrasive spend, which a robot can raise or lower depending on how it is tuned. Dust and spark extraction. And whether your parts vary within a family, which suits scanning, or across unrelated families, which does not.
Sources: https://www.fanucamerica.com/case-studies/how-a-steel-fabricator-boosted-output-upskilled-workers and https://genedge.org/impact/genedge-success-stories/lawrence-brothers-case-study/
If you have a grinding or finishing step with a rework number you can name, describe it to Joe, the concierge on this site. It is free, and it will ask about variation first.
Related on this site
- Sanding and finishing
- Weld seam grinding
- Deburring wasn't automatable in a small shop until force control grew up
- I never thought we could afford a sanding robot. The dust and the overtime said otherwise
- The casebook of published deployments
Variation was the problem, so the fix was a robot that scans each part and plans its own path. The grinders became robot operators, which the account says was the plan.
Put it to the test on your own job
Tell Joe what you are running and he will apply this to your numbers, including telling you when a robot is the wrong answer.
More notes