Gordon Seidoh Worley comments on Optimum number of single points of failure