The Next Generation of AI-based Visual Inspection in Pharma

In partnership with leading machine builders, Dabrico and Genesis Packaging Technologies, Boon Logic introduces a new approach to automated visual inspection built for the toughest inspection use cases.

The most powerful partnership in

Pharma vision

With over 100 combined years in Pharma, Dabrico, Genesis Packaging Technologies, and Boon Logic have partnered to tackle the toughest visual inspection challenges. Our partnership combines the material handling, service, and support expertise of Dabrico and Genesis Packaging Technology with Boon Logic’s cutting-edge software for unsupervised ML-based visual inspection, AVIS. Whether retrofitting an existing SAVI or deploying a new line for upstream or downstream inspection, we have you covered.

Built around cutting-edge AI

Pharma vision reimagined

ML-based automated visual inspection

DAI-50

Built by Dabrico powered by AVIS

  • Compact 36″x36″ automated visual inspection machine
  • Three camera, AI-based inspection at up to 50 vials per minute
  • Train a recipe for new products in one day
  • No AI or computer vision expertise needed
  • Trains from compliant units (no defect sets needed for training)
  • Integrated material handling system and ejection

Advisum-AI

Built by Genesis Packaging Technologies powered by AVIS

  • Mounts into operator inspection cabin turning SAVI into AVI
  • Three camera, AI-based inspection at up to 50 vials per minute
  • Train a recipe for new products in one day
  • No AI or computer vision expertise needed
  • Trains from compliant units (no defect sets needed for training)
  • Currently available for the Seidenader V90 (ask about other SAVI machines)
  • -integrates with SAVI conveyor control and ejection system

Built for the toughest

Inspection use cases

Powder and lyophilized product

Distinguishes between compliant and non-compliant product formation to ensure a high rate of detection and low false eject rate.

Tubular and molded glass

Learns normal variations within tubular glass and even in more challenging molded glass to minimize false ejects.

Liquids prone to bubbles

Learns normal formation of bubbles from compliant product and detects unusual variants.

Translucent plastics

Learns normal variations in seams and sidewall thickness to maximize detection while minimizing false ejects.

Detection results

20 ml with Powder Product

Product type

20 ml molded glass with powder-fill product

Challenge

Significant compliant variation exists in powder deposition, variations in thickness of the molded glass, and small bubbles embedded in the glass render traditional automated inspection methods impractical.

Solution

AVIS learns all of these normal variations and automates the visual inspection process enabling the customer to scale production without hiring more human inspectors.

Results

AVIS detected 98% of all defects with a false eject rate of 2.7%. Particles below 600 microns were learned as compliant powder deposition whereas anything above 800 microns was detected as a defect as this was not present in the training set.

Examples of detected defects

Examples of detected defects

Schedule a demonstration