This question needs some clarification. Do you mean that "most of the time 4x is the first data point" or did you mean that "most of the time 4x is the data point that is the first one to be less than 10% lower than the next longer point"? This second interpretation is now outmoded. If you use a saturation point solver program then this gives you a unique value for peening intensity. For your example of the unique point on the curve at 2.55 then you have to decide which is the nearest PRACTICABLE time. Using integral passes then 3 passes is obviously the nearest to 2.55. You would have to persuade auditors that to use 3 passes was not practicable.
