Wednesday, May 23, 2018
Home/Current Issue >  Industry Articles > 

Advanced Modelling Inspection Technique Achieves Success
The modelling approach provides significant advantages in speed compared to algorithm-based approaches.

Automatic optical inspection (AOI) is an established solution for the reliable inspection of printed circuit boards (PCBs) in the electronics manufacturing industry. AOI systems have developed considerably since their introduction in the mid-1990s, and now appear on most surface-mount-technology (SMT) production lines worldwide. The majority of AOI systems use standard vision analysis technology in the form of multiple controlling algorithms. Although there are many variations of this approach, most are "programmed" and "tuned" in the same way. Modelling technology, however, is completely different in that it does not rely on a standard algorithmic approach but calculates process variations in real-time on real production data by analyzing pixel by pixel the image of an actual production PCB. First conceived in the mid-1990s and extensively developed since, Modelling is based on Principal Component Analysis (PCA) technology and has many advantages for companies looking for a fast, versatile system that can be deployed to production quickly and with minimal ongoing production tuning.

A key metric when selecting an AOI solution is the total cost of ownership once that system is deployed in production. Programming time and production tuning time are major contributors to the on-going costs of ownership and should be measured and understood well in advance of integrating such a system into production. In low-volume/high-mix applications, the setup time is even more critical since an AOI program must be ready to provide reliable inspection functionality before the production run is complete if it is to have any real value. For production batch sizes below 20X PCBs, this can be very difficult to achieve on most algorithm AOI systems.
This plot shows false fail PPM by component.

Most algorithm technology systems are set up with the user having to anticipate the possible range of defects that could occur in production. The set-up process includes selecting combinations of algorithms and setting their parameters in addition to parameters required for controlling image acquisition. This process can be very time-consuming with careful attention required to ensure everything is set up accurately.

Modelling Technology Benefits
Statistical appearance modelling technology can be set up easily based on an image of the first production PCB. The system learns "real-world variations" based on operator interaction with the reported results of inspection tasks. This results in an accurate statistical description of normal variations in a product. Using this description during inspection makes possible accurate reporting of what is acceptable and what is not acceptable to a user, based on individual process or quality requirements. Clear advantages of this approach are that a user need not anticipate potential defects or process issues since the system will "flag" anything that is outside of the "normal production range." In addition, because the system is programmed with real production variations, it is very sensitive to subtle changes, enabling very reliable defect detection. Recent developments in this technology include autonomous prediction of process variations, enabling AOI systems to be set up from a single PCB with production-ready performance. Setup times can be as short as 15 minutes from data input to first PCB inspection, making it extremely attractive for new product introduction (NPI) and first off verification.

Automotive Production Evaluations
Production evaluations recently performed in the automotive industry help to demonstrate the effectiveness of this statistical appearance modelling technology for electronic production. A PCB that was a 6-up panel measuring 300mm in length and 200mm wide with total component count of 1380 and 73 individual component types was tested. The inspection setup included tasks to reliably detect all component body, position, text, value, and solder joint-related defects and the total individual inspection task count was 8,307.

The production evaluations revealed a significant performance advantage of the statistical appearance modelling approach compared to algorithm-based technology for this application, with as much as 3X faster setup time for the modelling approach. Of course, fast setup time is only an advantage if the system can reliably inspect PCBs afterwards with a low false failure rate and provide immediate value to the user. When counting false failures between the two approaches, after programming from 1, 10, and 30 panels, the modelling system was approximately three times lower in failure rate than the algorithm-based approach.

AOI systems have developed considerably over the past 20 years; with the constant advances in computational technology, there is no doubt that this pace of development will continue. Modelling technology is a key area of image analysis that is benefiting from these advances and is already an attractive alternative to traditional algorithm technology when applied to SMT inspection. With the ever increasing demand for faster set ups and improved inspection performance on a wide range of applications, statistical modelling technology is a very interesting and valuable solution especially where set up times and cost of ownership are critical to success.


Contact: CyberOptics Corporation, 5900 Golden Hills Dr., Minneapolis, MN 55416 763-542-5000 Web:

search login