In manufacturing, scrap and waste are constant challenges. Yet too many quality teams are missing their greatest opportunity to reduce both. See how better SPC tools bring possibilities to life.
Problem: Scrap and Waste Are Invisible—but Costly
We recently met with a group of quality managers from a top-tier tool manufacturer. This group of representatives—from sites around the globe—wanted to know how to reduce scrap and waste and improve quality control, productivity and profitability. We asked when they had last looked at their in-spec quality data.
They laughed!
“We never do that!” they all told us. Naturally, they believed that their attention should be on out-of-spec data. But then we showed them what their quality data could tell them when used in conjunction with a powerful quality intelligence platform like InfinityQS® Enact® or ProFicient™.
Proposed Solution: Use Better SPC Tools to Ask Better Questions
First, we asked the quality pros how they typically reviewed their real-time statistical process control (SPC) data, after it had been collected on the floor. In response, they gave us a spreadsheet with the raw data: column after column of individual data points they would have to scroll through. No wonder they weren’t finding solutions to their scrap and waste challenges!
Seeing the path to process and quality improvement requires stepping back a bit from the individual raw details. InfinityQS accomplishes this visibility into data by aggregating and rolling up data for all machines, products, lines, shifts, sites, and so on.
We then help manufacturers repurpose those data to see a bigger picture than is possible when reviewing the data for one product from one machine at one point in time.
When you can see a bigger picture of data, you can begin to ask vital questions:
- Which machine has the most problems?
- During which shift does the machine work the best?
- What is the person on that shift doing differently?
This is where the real value of SPC lies.
Result: Data Visibility Reveals Actionable Information
To show the company what was possible, we began with just four part numbers and several production lines’ data. Into the InfinityQS solution, we fed measurements for one of the products, a grinding blade, including—
- Thickness at several points along the blade
- Balance
- Weight
We looked at all the team could discern simply by querying those data from within the quality intelligence system:
- Just looking at raw data in a color-coded chart immediately provided visual feedback: red shading for data that fell outside of spec, yellow for measurements that were close to falling out of spec, and green for in-spec readings.
- Viewing the data in a standard control chart, the team saw that alarms popped up in real time whenever a measurement fell out of spec. By clicking an out-of-spec plot point, they could immediately see individual data values that helped them drill down to the problem.
- Checking the histogram view and box and whisker charts, the quality pros could easily see that thicknesses were running high, though not outside of specifications. Translation: Raw material was being wasted. By investigating the cause of these variations, the quality team might be able to stem this waste and save the company money in raw materials.The same is true of in-spec measurements that fall on the lower end of the spectrum; although not likely to prompt recalls, such products can cause customer complaints and dissatisfaction.
- By charting multiple key characteristics in a single view, the team could see the percentage of product that was above target, helping to further isolate the cause of the problem (e.g., a specific machine, shift, or production line).
- The team also looked at an event pareto chart to see aggregated defects and attribute data. This chart provides yet another means to zero in on potential problem areas—and big savings.
At the end of the demonstration, the quality pros were astounded at all the information they already had in hand, if they just used the right tools to manage their data.
The key wasn’t to collect more information—for them, it was simply to change the way they saw the data they were already collecting.