CV and AI-based Parts Recognition for Automotive Inventory

Business Objective

A US heavy duty aftermarket network manages millions of parts across hundreds of categories and manufacturers. Staff rely on manual catalog lookups and institutional knowledge, making counter productivity dependent on veteran expertise.

The network needed a faster, more reliable way to identify unfamiliar parts at scale.

The Solution

OLSYS designed and deployed an AI-powered visual recognition platform that replaces manual catalog searches with a short smartphone video or picture. Once it is captured and uploaded by counter staff, it eliminates continuous barcode lookups, part-number reliance, or dependence on veteran SME expertise.

Cloud-based ML model instantly categorizes the part, seamlessly routing high-confidence matches while filtering out uncertain results to guarantee accuracy.

The platform then fuses advanced computer vision with basic physical specs. Once staff inputs standard dimensions and a qualifying attribute, the system cross-references them against strict tolerances – returning exact part numbers, fitment data, and confidence scores in seconds. The method is patent pending.

Results 

• High-Precision Visual Identification:
  Staff are quickly identifying unfamiliar parts, reducing reliance on expertise and helping prevent costly returns.
• 35+ Hours Saved per Location:
  Teams are reclaiming valuable capacity for customer interactions while reducing manual catalog work. 
• Elevated Customer Satisfaction:

  Customers are spending less time waiting as staff retrieve accurate part information within seconds.

• Network-Wide Workforce Resilience:

  Institutional knowledge is being shared across locations, helping new team members support customers confidently from day one.