I find that the documenation for TransposeInterpolateRange doesn't tell me enough. The 'Transpose' part I understand. The word 'Interpolate' I understand when I can specify specific points in time such as in "Give me the temperature every hour on the hour". Then it would find the nearest measured points and find a curve between them and find the intersection of the curve and the top of the houre. But when that is applied to a "Range", I don't understand how PI is able to choose the point to report back as the result.

I ask this question because I saw a query written that returns an ontime of 100% for a full day, but I also saw that the day had not completed at the time that value was pulled.

@Joe: The definition of the

TransposeInterpolateRangefunction says that it retrieves interpolated data for a time range at aTimeStepfrequency.In other terms, this function allows to generate a sample of evenly space values in time for a time range for a number of attributes as defined per the element template. "PI" doesn't choose a point randomly put evaluate the value at a given timestamp, specified by your range and timestep, by an interpolation rule.

If your time range is extending in the future and the last value in the snapshot (PI Snapshot Subsystem) represents an "On" status for your asset as all the other values in that period, this interpolation (in this case, it would be more an extrapolation) would give a 100% ontime.