The McKeesport Fire Department wants to know how likely it is that a truck pump will fail. The fire chief, George Pyro (no relation), thinks pump failure is a function of age (X1) and water hardness (X2 measured on a scale of 1 to 10).
The department statistician runs a regression on a dummy variable (coded 1 for failure and 0 for no failure) for 217 pumps. She finds the following:
 = .14 + .01X1 + 05X2
sb1 = .0002 sb2 = .025 sy|x = .04 R2 = .93 Adj. R2 = .92
Write a memo explaining what the regression means. If the average water hardness is 3 and the chief would like to replace any pump with a probability of failing of .80 or more, at what age should pumps be replaced?