Ore Car Fill Detection AI
Measure how full every ore car, skip, and haul truck is at the loading point.
Ore Car Fill Detection AI for the Toughest Underground Rail Haulage, Shaft, and Open Pit Operations
Deploy Anywhere, Run Everywhere
Run ore car fill detection on loading chute, dump station, and shaft cameras on the edge, on-prem, in your VPC, or via API, in dust, water, and low light.
One Platform, Full Adoption
Tools every mine team can adopt, from production engineers to mine planners, no separate ML team required.
Secure, Compliant, and Audit-Ready
Data stays safe with SOC 2 Type II compliance, encrypted data, HIPAA compliance, and an uptime SLA.
- Fill Level Per Ore Car & Skip
- Underloads & Overloads at the Chute
- Carryback After Dumping
- Spillage on Track & Around Loading Points
- Car ID & Train Consist Reads
- Tonnage Estimates Per Train & Shift
“Roboflow has been instrumental in accelerating our deployment of innovative AI solutions.”
Travis Turnbull
Vice President & CIO, Pella Corporation
Talk to a Vision AI engineer who's shipped in mining.
Bring us your toughest ore car fill detection problem and we'll map a working solution.
Ask us about:
- Solution architecture that fits MSHA 30 CFR Part 56 and 57, ISO 45001, and your haulage and hoisting controls
- A live demo on your own loading chute, dump, or shaft footage
- Deployment options: edge underground, on-prem, air-gapped, VPC, or on dust- and water-rated cameras at loading points
- ROI modeling against lost tonnage from underfilled cars, carryback, spillage cleanup, and hoist capacity
Over 16,000 organizations build with Roboflow.
- Rivian
- Pella
- Chobani
- USG Corporation
- BNSF Railway
- American Woodmark
Vision AI is transforming businesses
Customers across the board are solving complex challenges and driving meaningful impact.
- Automotive customer
$10 million
Saved by automatically detecting defects on the production line
- Logistics & freight company
90%
Less time spent manually tracking shipping inventory
- Building materials supplier
60%
Lower customer return rate with improved product quality
Fill Every Ore Car Right, from the Loading Chute to the Dump Station
Bring real-time intelligence to every ore car and skip with Roboflow Vision AI for ore car fill detection. Built for underground and open pit mines where underfilled cars waste haulage capacity, overfilled cars spill on the track, and carryback after dumping quietly steals tonnage every trip.
Fill at the loading point:
- Measure the fill level of each ore car, skip, or haul truck bed as it is loaded, with a model trained on your own cars, ore, and lighting.
- Alert the chute operator when a car is underloaded or about to overflow.
- Estimate volume per car from the fill surface, and convert it to tonnage with your ore density.
Carryback and spillage:
- Detect carryback left in cars and skips after dumping, so sticky ore is cleaned before it builds up.
- Flag spillage on the track and around loading and dump points.
- Watch the dump station for cars that did not fully empty.
Train and shift records:
- Read car numbers and tie fill level to each car in the train consist, next to railcar identification.
- Report fill rate and tonnage per train, loading point, and shift.
- Send fill data to the PLC and production reporting so haulage and hoisting run at capacity.
Bring intelligence to every ore car today. Stop underfilled cars, carryback, and spillage from becoming lost tonnage and cleanup shifts.
Frequently asked questions
What is ore car fill detection with Vision AI?
Ore car fill detection with Vision AI uses cameras at loading chutes, dump stations, and shafts and a trained computer vision model to measure how full each ore car, skip, or haul truck is, alert on underloads and overloads, detect carryback after dumping, and report fill and tonnage per car, train, and shift. It pairs with stockpile volume measurement and rock fragmentation analysis across computer vision in mining.
Can Vision AI estimate how full an ore car is from a camera?
Yes. A segmentation model outlines the ore surface inside the car, and a camera calibrated to the car's known dimensions converts the surface height and shape into a fill percentage and volume.
Does ore car fill detection support MSHA and ISO 45001 programs?
Ore car fill detection supports haulage and loading safety programs under MSHA 30 CFR Part 56 and 57 (US metal and nonmetal mine safety standards, including haulage) and ISO 45001 (the occupational health and safety management system standard) by reducing overloads and spillage and keeping a record of every loading event.
Roboflow serves as the measurement engine, and your production and safety teams own the fill targets, loading procedures, and control responses.
Can it integrate with our loading controls and production reporting?
Yes. Roboflow writes fill level and alerts to a PLC tag over OPC UA, Modbus TCP, or MQTT, so chute gates and loading controls can respond, and sends fill and tonnage records over REST or to a database for SCADA, historians such as AVEVA PI, and production reporting and fleet systems such as Modular Mining and Deswik.