Fab 40: Impulse Manufacturing tackles downtime with data
After reducing downtime, throughput skyrockets
You can't improve something you don't measure. Impulse Manufacturing wants to cut down the time between jobs—specifically, between the last good part of the previous job to the first good part of the next job.
The employees at Impulse Manufacturing live by this lean manufacturing adage: When you consider the overall manufacturing cycle, the time work-in-process spends between operations really counts. If a machine is down longer than it should be between setups, something’s awry.
Of course, you can’t improve something that you don’t measure, and in recent years the Dawsonville, Ga., fabricator—No. 14 on this year’s Fab 40— has stepped up its measuring effort. In 2009, at the worst of the recession, the company installed arc-on-time devices by its welding cells. The data showed when a welder’s arc was on during a shift, when it wasn’t on, and for how long. If the system showed a welder had a prolonged amount of downtime between jobs, supervisors could ask questions and offer help. Was the tooling unavailable? Was the setup difficult, or the orientation challenging?
Karl Baysden, director of sales and marketing, added that getting employee buy-in wasn’t all that difficult. Managers didn’t approach it from a judgmental or accusatory standpoint. “It was about identifying areas of improvement and working as a team to get it down. Culturally, it wasn’t an issue for us. At first we were just collecting data. And then when we shared the data, it was more eye-opening. It opened a lot of discussion, and people embraced it.” The company has a place where employees can make suggestions for improvement, and the monitoring systems “really sparked a lot of communication.”
Over the past two years the company has expanded this downtime measurement to include most equipment on the floor. Beside each machine is a module from Wintriss Controls called ShopFloorConnect (previously known as LETS). If a machine isn’t running after a certain time period, the machine won’t start unless the operator selects a reason for the downtime from the drop-down menu before starting the job. It may be because material is not available, a machine is unexpectedly down for maintenance, a schedule lapse, or any number of other reasons.
In each department the tracking system has a different set of criteria to check. In the laser area, the system checks for beam-off time. In welding, the company can track the time between jobs, or how many minutes of arc time over a given time frame.
Impulse has expanded this tracking to include not just setup time, but also the total time between jobs. Say an operator clocks in on a job and sets up a machine. The tracking system shows it took 20 minutes to set up tooling and produce the first good part. That’s fine, but what about the time before the technician clocked in on the job?
That’s why, for some jobs, as soon as a technician clocks out of the previous job, the clock starts ticking. If it takes longer than a specified time for him to clock in on the next job, the system will shut the machine down until the operator inputs the reason into the system. This essentially tracks time from the last good part of the previous job to the first good part of the next job.
“We’ve implemented this functionality on literally every machine in the building,” said Baysden, from spot welding to press brakes and hardware insertion.
The system also monitors manufacturing effectiveness. Say that it was estimated that a certain press brake operator could produce 100 parts per hour. “So within that hour, after the operator is set up on the machine, maybe he gets all 100, and we think we’re doing great. But if we start monitoring for downtime within that hour, or slow periods during that hour, we can get even more efficient.”
For example, how could the technician’s job be made easier, so he doesn’t have to slow down? Can the setup be improved upon to make the process more reliable? Could tools be placed closer to the point of use?
As Baysden explained, the monitoring doesn’t uncover solutions to problems, but it allows people to identify potential areas for improvement. With that data, employees can start asking the right questions. The data paints an accurate picture of shop floor productivity, which in turn provides the foundation for improvement efforts.
For instance, the data showed that operators spent a lot of time signing out tools and supplies, something as simple as an operator hunting for or requesting specific work gloves needed in certain departments took an enormous amount of time. Now, vending machines supply the gloves needed in those areas. An operator simply checks them out, then goes back to the job at hand. In turn, the vending machines tie directly into the ERP system, so purchasing can see when certain supplies are running low and order more.
“Our new tracking systems will tell us if we should be more efficient than what we have quoted in our system,” Baysden said. “Over the years we’ve identified all these little issues that we didn’t know were there. For instance, we’re finding that for some jobs, we’re able to weld 250 to 300 percent more product in the same time frame than we did in 2009. In fact, we’re seeing increased throughput in every single department.”
The FABRICATOR is North America's leading magazine for the metal forming and fabricating industry. The magazine delivers the news, technical articles, and case histories that enable fabricators to do their jobs more efficiently. The FABRICATOR has served the industry since 1971.