United StatesLogistics and transport2023 – ongoingNorfolk Southern
Digital Train Inspection Portals
Norfolk Southern deployed AI train inspection with 38 deep-learning algorithms
Problem
Stationary inspections could miss defects that appeared while railcars were moving. Inspectors also needed views of components that were difficult to examine directly.
Approach
Georgia Tech Research Institute developed portal hardware, while Norfolk Southern's internal teams built the AI analysis. Trackside cameras captured passing railcars, and deep-learning models flagged possible defects. Specialists at the Network Operations Center reviewed the findings, with critical defects sent for immediate handling.
Results
Takeaway
The inspection system combined moving-train imagery with human review. Norfolk Southern did not publish numerical accuracy or false-positive rates.
Categories
- Size
- Large enterprise
- Industry
- Logistics and transport
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Computer vision
- Adoption
- Partner co-development
- Goal
- Quality and risk
- Evidence
- Company's own claim
Sources
Source summaries
Norfolk Southern described its first inspection portal in Leetonia, Ohio, and the human review process. It reported 38 deployed deep-learning algorithms, without numerical validation results.
Full details are in the original sources.