Quality Scan: In-Line Process-Inspection Economics
By Bruce A. Johnson
Director Business Development
Why is it that when you buy a bottle of beer from one of the major North American beer producers you can just about guarantee that it:
1.) Tastes the same each time
2.) Looks the same each time
3.) Is packaged the same each time
4.) Costs almost nothing to manufacture.
The correct answer is automated in-line inspection. When you walk into a modern brewery, nearly every variable is identified, monitored, trended--and, if it makes economic sense, controlled or eliminated. When possible, automated process feedback loops control a vast majority of the equipment--again with the idea of minimizing process variation. They want the product repeatable, repeatable, repeatable.
So making parts is different than making beer? Well, what's the end goal? To make the best-quality finished product for the lowest possible price in the fastest time-repeatably. More simply put: to make money. We monitor the process to remove variables that consume time or cost money.
The biggest difference between the manufacture of beer and component fabrication or major assembly is typically speed and spatial analysis. In the case of beer, statistics are rarely required to decide what to inspect--brewers simply check nearly every major point that could induce process or product variation. Statistics are used while continuously monitoring the product and process, looking for the best opportunities to lower cost or improve product consistency. The major difference is that beer companies live primarily in a 2-D world for inspection.
Over the past 3 - 5 years, manufacturing has been driving toward developing multidimensional in-line inspection tools. Within the past 2 - 3 years, multiple systems have been installed using 3-D technology combined with six and seven-axis robotic mobility. Imagine a part coming off a stamping press. An in-line 3-D inspection laser scans the part and validates all critical features. It's then accepted or rejected, and each feature is statistically monitored with data stored to allow trending of problems.
Take this one step further. Imagine a part comes off a machining center, is picked up by a robot, and presented to an in-line 3-D laser-inspection system. One of the critical features is still in specification, but starting to go out of control due to tool wear. The robot moves the part to the next station, with feature data plotted in control charts. Feedback is given to the machining center, a tool is changed, and the next part comes off without a hitch. Parts are all matched to dimensional IDs through optical character recognition (OCR), and sent to the plant network for warranty archiving and review by quality personnel.
So how does this happen? One means is robotic laser measurement systems employing 3-D laser cameras mounted to six, seven, and even eight-axis robotics to give high-quality data. With the costs of robotics dropping and computer processing speeds rising, these systems are not the expensive ones people used to dread. Many times a fully automated robotic 3-D in-line inspection station can be installed and running 24 hr/day without supervision for the price of an off-line CMM inspection station. Such systems are being used today in a variety of machining and assembly applications.
The real question is: If you could automatically check the critical features of every part in-line, what could you hope to gain? Some economic benefits enjoyed by others (including brewers) are:
- Reduced rework
- Increased production throughput
- Reduced labor
- Improved part quality
- Improved assembly quality
- Lower assembly costs
- Improved assembly delivery time
- Increased customer satisfaction
- Increased sales opportunities
- Reduced part FOB costs
- Archive data for warranty/liability protection
- Reduced overtime
It's hard to determine the optimum time to begin investing in in-line inspection systems. If the metal machining and fabrication field is to survive in North America, however, it must continue to move toward automation. Tomorrow's in-line inspection systems will inevitably replace the eyes and ears of operators to decide what is good and, via closed feedback loops, continue to adjust and improve other systems.
This article was first published in the April 20004 edition of Manufacturing Engineering magazine.