← Cases
South KoreaManufacturing2022 – ongoing

Doosan Bobcat

Cogniac visual inspection of tractor parts kits

Doosan Bobcat reported parts-kit errors fell to 1 in 20,000 after computer-vision inspection

WorkingCompany's own claimVendor ↗

Problem

Incorrect or incomplete parts kits interrupted tractor assembly. Kits were prepared in Minnesota and shipped to North Dakota.

Approach

Doosan Bobcat used Cogniac vision pipelines to monitor kits for different tractor configurations. Cogniac supplied a computer-vision platform backed by NVIDIA GPUs.

Results

1 in 20,000kit error frequency after implementation, versus one-third beforehand, according to Doosan BobcatCompany's own claim[1]

Takeaway

Inspection targeted kit completeness before missing parts could disrupt downstream assembly.

Categories

Innovation type
Process innovation
AI technology
Computer vision
Evidence
Company's own claim

Sources

Source summaries

[1]NVIDIA7 Jan 2022

NVIDIA described Cogniac inspecting Doosan Bobcat parts kits before assembly. It attributed the reduction from one-third incorrect or incomplete kits to 1 in 20,000 to Doosan Bobcat.

Full details are in the original sources.

Added 1 Nov 2026 · Updated 1 Nov 2026
Open case page ↗