Cookies Management
We use cookies to provide you with the best site experience.
Cookies Management
Cookie Settings
Cookies necessary for the correct operation of the site are always enabled.
Other cookies are configurable.
Essential cookies
Always On. These cookies are essential so that you can use the website and use its functions. They cannot be turned off. They're set in response to requests made by you, such as setting your privacy preferences, logging in or filling in forms.
Analytics cookies
These cookies collect information to help us understand how our Websites are being used or how effective our marketing campaigns are, or to help us customise our Websites for you. See a list of the analytics cookies we use here.
Advertising cookies
These cookies provide advertising companies with information about your online activity to help them deliver more relevant online advertising to you or to limit how many times you see an ad. This information may be shared with other advertising companies. See a list of the advertising cookies we use here.

A Proven AI Algorithm to Reduce AOI False Calls and Manual Verification by Up to 60%

October 21, 2022
Factory intelligence analytics historical analysis
As our very own Mitch DeCaire recently pointed out in an interview: “I know a lot of people are talking about artificial intelligence and I always like to ask the question - “OK, what have you done in a real factory? Give me an example that I can understand.” We naturally agree with his sentiment and would like to share a concrete example of how an AI algorithm can be used to reduce AOI false calls and manual verification by up to 60% during the SMT process.

Use Case: AOI False Calls Reduction

False Calls are, and have always been, a major concern for every electronics manufacturer. Based on discussions with current customers, the estimated ratio of AOI false calls ranges anywhere between 30% to 80% depending on the AOI program. Since these false calls can’t be depended upon, manufacturers normally have an operator standing behind the machine to verify all the defects the AOI delivers – which translates to higher labor costs and a significant risk to decreasing the line’s cycle time. Needless to say, not an ideal scenario.

This is where AI, or artificial intelligence, comes in to make a stunning difference. Using one of the IIoT.Edge’s machine learning algorithms, the application is able to review every defect flagged by the AOI and detect false calls with a high degree of confidence. The algorithm then automatically screens out the high confidence false calls, and only sends the remaining defects to the verification operator. This in turn, dramatically reduces the amount of potential false calls the Operator needs to screen.

Less time spent verifying false calls means less wasted time and labor. We’ve seen this most recently with a major EMS customer in Europe who reported a reduction of up to 60% in manual verification due to the use of this algorithm. Additionally, by reducing manual verification, we’ve also seen an indirect impact on overall throughput, which is something all electronics manufacturers can appreciate!

Figure 1: Screeshot of IIoT.Edge's dashboard

Screenshot of IIoT.Edge – False Calls Reduction

Now that you know how you could benefit from IIoT.Edge, here’s a sneak peek of its dashboard. In this screenshot, you can see in the top left that “real defects” (shown in red) are the ones sent to manual verification. All the others (colored in green) have been identified as false calls with a high enough confidence degree (set by the user) and will not be re-verified by an Operator. Even though this is only an example, we can clearly see that the operator would receive only three defect reviews, instead of eight without the use of the algorithm.

If you look at the bottom left of the dashboard, you’ll see that as the time goes by, the algorithm gets better and better at identifying false calls. While the base algorithm comes in already pre-trained, it will improve in accuracy by working with your specific products, until it is finely tuned to your specific environment. Who knows, maybe you’ll even experience more than a 60% decrease and report even better results!

Sending real-time factory data to IIoT.Edge

IIoT.Edge is one of the most advanced solutions in the industry. One of the reasons why is that it includes the most complete SMT-centric machine connectivity solution, Co-NECT, to send domain-specific data directly from the machine to the application in real-time.

Curious to understand how else this technology can transform the way your line runs? Here’s a figure to give you an overview of our solution. As you can see in the top of the graphic, IIoT.Edge can also be used for predictive maintenance, which is something we’ve already covered in a previous article.

Figure 1: IIoT.Edge diagram


In conclusion, while many suppliers may start talking about AI, we continue to outperform the competition by offering pragmatic and usable solutions for the electronics manufacturing industry. Interested in learning more about our technologies? Contact one of our experts today!
You might also like