Skip to main content

Industry

Consumer Goods

Result 1

99.7% Accuracy in detecting all defects 4% Improvement in defect detection compared to manual inspection $25K cost savings anually

Deployments

The images captured from the cameras were stitched, sent to Jidoka’s AI/ML engine for identifying possible defects and the result made available within 0.8 seconds The output generated by the system is sent to an automatic rejection system which removes the defective bottles from the production line An alert mechanism is triggered to indicate continuous occurrence of any particular type of defect

Result 2


Manual defect detection for ~150,000 bottles per day at 120bpm leading to high false positives Jidoka leveraged Kompass software with Deep Learning AI for quality inspection of printing on 360 degrees of the bottles Infrastructure with 2 cameras to t