Quality assurance checks for auto parts are such a tedious task that only AI can perform them reliably, according to Daniel Davenport, senior director of automotive at Capgemini. He said that watching parts flow off the assembly line is a surefire way to tease a human inspector at missing faulty parts as they appear.
“Seals, welds, sockets, and lug nuts; these are things that are very difficult for a human to examine reliably,” Davenport said. “A 10-hour shift shows the advantage of computer vision.”
The problem was that specialized computer vision screening systems were expensive. According to Davenport, the creation of such systems has been slow, requiring up to four months to install cameras to gather information about defects in products.
Even worse, the desktop-based analysis platform kept its information captive, causing delays and reworks due to a lack of company-wide visibility of the quality data.
Working with industry expert John Ward at the University of South Carolina, Capgemini has developed a simpler optical inspection system that distributes its analysis using Verizon 5G. There are a few custom vision systems for manufacturing that support 5G, but the Ward team at USC found a ready-made solution: the Apple iPhone 13 running IBM’s Maximo visual inspection app.
The system aims to exploit the high speed and low latency of the 5G network to:
- Deliver high-resolution images in real time
- Speed up the inspection process
- Productivity improvement
- Address potential problems before they happen
Capgemini’s role is to provide advanced analytics that allows manufacturers to have a central view of the defects that Maximo has detected. “We joined in supporting some of the demos by getting data from a standalone product into the cloud and into our dashboard reporting system,” Davenport explained.
He said putting the iPhone in the grip of an industrial robot arm to inspect parts instead of building a traditional robotic vision system leads to cost savings of 10 to 15 times. This is only for the screening system, Davenport added, but the real cost benefits come ultimately, from lower rework and warranty costs. “OEMs can tell us their rework or warranty expense ratios, so they can see a real-time return on investment.”
He noted that the system could also provide visibility into increases in defects that occur at specific times, which could indicate the need for additional training for specific workers.
The team has already demonstrated the system at an unnamed manufacturing facility, according to Davenport. It’s easy to convince companies to give it a try because of the low cost of a single screening station and the ease with which the solution can be scaled up for increased usage.
South Carolina is home to assembly plants for Volvo and BMW, along with tire maker Michelin, so there are many opportunities for the University of South Carolina to work with nearby factories to enhance quality and save money for manufacturers in their hometown.
#Smartphonebased #manufacturing #inspection #reduces #quality #control #cost