MSHA proximity detection: how underground coal miners are protected

MSHA's proximity detection rule, effective from March 2015, requires place changing continuous mining machines in underground coal mines to have systems that warn miners and stop the machine before it contacts them. The goal is to prevent pinning, crushing and striking deaths. Its performance requirements depend heavily on software and electronics.

What the rule requires

Operators must equip place changing continuous mining machines with proximity detection systems. These use electronic sensors on both the machines and the miners to detect motion or the position of one object relative to another. The system must give audible and visual warnings, and it must stop the machine before it contacts a miner.

Why MSHA acted

MSHA states that since 1984, 35 miners have died after being pinned, crushed or struck by continuous mining machines in underground coal mines. It projects the rule will prevent 49 injuries and 9 deaths over ten years. Those are MSHA's own projections.

Dates

The rule took effect on 16 March 2015, with compliance phased in over 8 to 36 months. A related MSHA reference lists the installation deadline as 16 March 2018. When the rule was issued, MSHA said that based on manufacturer information, 425 of roughly 863 machines in operation already had such systems. We did not verify later amendments, so check current status.

Where software failures become physical risk

A proximity system is a safety function. It reads signals from tags, decides if someone is too close and commands the machine to stop. Think through what each failure does:

• A missed detection leaves a miner unprotected without any visible sign.
• A false stop halts production, and frequent false stops give people a reason to defeat the system.
• A fault that leaves the machine moving is the dangerous one, so a failed sensor or controller should lead to a stop, not to silence.
• Interference from other equipment can degrade signals, so it belongs in the test plan.

That is the same logic as in functional safety: decide what the safe state is, make failure lead there, and test the failure paths.

Questions for a software or firmware team

1. What is the safe state, and how does the machine reach it on a power loss, sensor loss or processor reset?
2. How are updates delivered, authenticated and rolled back on equipment underground?
3. Who can change thresholds, and is each change recorded?
4. How do you test detection range and latency after a firmware change?
5. Are third party components listed and tracked for known flaws?

Question five is where an SBOM and dependency scanning help, for firmware you build. Safeguard covers software inventory and scanning, and does not provide mine safety equipment or monitoring.

Quick check

Sources

• Final rule: proximity detection systems for continuous mining machines, MSHA

Check the full rule text for exact requirements, exemptions and later amendments.

Last reviewed: 19 October 2026.