A Borescope Replaces a Two-Person Inspection at a Truck Plant

Daimler's Mount Holly Truck Plant has deployed an AI inspection system from Loopr to verify the joint connecting the steering column to the truck frame — a place where a bolt must pass through two holes in the correct orientation and be properly torqued.

The previous process was a four-eye inspection: one inspector checking the assembly inline, a second inspecting it offline and signing off before the truck could be released. Loopr built a tablet-based system using a borescope to check joint fit, orientation and torque presence in a location operators could not easily see.

The justification is the interesting part. Plant general manager Joanna Cooper, who has eighteen years in manufacturing, tied the change to workforce turnover: average seniority at the plant had been cut in half, and with changing shift models she questioned whether all the eyes were actually working or only some were. That is a different argument from the usual defect-rate business case, and it is increasingly the real driver for vision AI on brownfield lines — inspection quality that depended on experienced people quietly degrades when the experience leaves.

The borescope detail matters too. This is a joint humans cannot see well, which reframes the system as extending inspection coverage rather than replacing labour. The four-eye process existed because the feature was hard to verify; two people looking at something neither can see properly is redundancy without accuracy.

Loopr founder and CEO Priyansha Bagaria was explicit that the path forward is a human in the loop. On a safety-relevant assembly in a vehicle, that framing is not a hedge — it is the condition under which quality engineering will accept a model's output at all.