My God, the prescience.

What was the basis of Mathematica's cost projections? The analysts assumed that the shuttle fleet would stage at least 50 flights a year. With each vehicle having a 10-year life, that meant at least 500 flights over a 10-to-12-year period. (At one point in 1976, NASA was projecting 75 flights a year. The number has been dropping steadily since, and now stands at "around 40 to 50 flights a year," Lee says.) As with any volume merchandising, the more flights there are, the lower the cost of each individual flight. So it was important to project lots of flights. These are "safe life" numbers.

I had to look it up: 4.9 flights per year.

Some suspect the tile mounting is the least of Columbia's difficulties. "I don't think anybody appreciates the depths of the problems," Kapryan says. The tiles are the most important system NASA has ever designed as "safe life." That means there is no back-up for them. If they fail, the shuttle burns on reentry. If enough fall off, the shuttle may become unstable during landing, and thus un-pilotable. The worry runs deep enough that NASA investigated installing a crane assembly in Columbia so the crew could inspect and repair damaged tiles in space.

...

While Columbia was still in orbit, some engineers suspected damage, but NASA managers limited the investigation, under the rationale that the Columbia crew could not have fixed the problem.

The external fuel tank, for instance, is full of oxygen and hydrogen cooled to -400°F. to make the gases flow as liquids. Ice will form on the tank. When Columbia's tiles started popping off in a stiff breeze, it occurred to engineers that ice chunks from the tank would crash into the tiles during the sonic chaos of launch: Goodbye, Columbia. So insulation was added to the tank. But while thermal cladding solves the ice problem, it adds weight.

And it was this insulation that fell off, broke tiles, and killed Columbia.

Thanks for this.