Embed Notice

At the start, much of what Chris did in his new job felt like bricolage.

He took data gathered by others and work done by others and repurposed it to his narrow problem.

His immediate goal was to create for the pillars inside the tunnels of longwall mines the equivalent of what engineers call a safety rating.

A safety rating is the load-bearing capacity of whatever is holding the load,
divided by the load.

(If it’s less than one, don’t look up.)

Bieniawski had created a formula for calculating the load-bearing capacity of coal pillars,
but to use it you needed to know the load that needed bearing.

Calculating this was tricky.

It changed as coal was removed from the mine in ways that were not obvious,
and that varied from mine to mine.

The rock that collapsed harmlessly behind the mining machine did not have the same ability to support the mountain above it as the previously intact seam of coal.

Crumbled cake offered less support to whatever was above it than intact cake.

The weight of the mountain needed to travel someplace.

One place it went was onto the remaining coal pillars.

The more coal you removed, the greater the so-called abutment load
— not the load that was vertically over the pillar, but the load that moved, horizontally, onto it.

Chris spent several years measuring the way the load on the pillars changed as coal was mined.

His aim was to reduce his findings to a set of equations that could be used by mine designers.

Given the length of the mining wall, the depth of the mine and the height of the roof, etc.,
the load should be roughly X.

X was the numerator of his safety factor, which,
to avoid the impression that the entire mine was rendered safe by it,
he renamed the “stability factor.”

He then back-tested the number against case histories to see whether coal mine roofs had indeed collapsed when the stability factor was less than his model thought it needed to be.

He was turning pillar stability into a science.

“All I’m doing is taking trial and error and looking at the data more scientifically,” he said.

By academic statistician standards, his work was more than a bit loose.

“I’ll never have a database that is large enough
— or collected in the random way that you’d need to do precise statistical analysis,” he said.

“I’ll never be able to say ‘there’s a 95 percent chance the roof will hold up.’

You’ll never know the exact probabilities.

I’m using statistics to make better engineering judgments.”