From the article : Theoretically, you could build a structure that would never fail, but you'd have to put infinite money into it. For a critical bridge of thi

From the article : Theoretically, you could build a structure that would never fail, but you'd have to put infinite money into it. For a critical bridge of this type, we would consider an acceptable chance for failure to be 1 in 10,000 years. And at that point why not do a cost benefit analysis of a tunnel?