The Growing Importance of Understanding NDE Applications and Results

Risk Based Inspection

A typical risk based inspection (RBI) analysis should include past inspection results, tempered by confidence in those results. For example, API’s (American Petroleum Institute) RBI methodology and software when calculating the likelihood of failure side of the risk equation asks for past inspection histories. This includes dates of past inspections for potential damage mechanisms, the effectiveness of those inspection techniques to find the anticipated damage and amounts of coverage. Via this logic, the program constructs a factor to represent the probable damage population scatter band and multiplies this times the entered corrosion rates, cracking susceptibilities or bulk damage rates. Corrosion rates are derived either from inspection results, expert opinion (typically used in the absence of data or absence of reliable inspection data) or calculated from damage modules, based on available industry data, experience and technology. In the absence of past inspection histories the program calculates a larger factor to multiply times the anticipated damage rate, producing a larger damage factor to account for the unknown.

In API RBI software, for example, a conservative ductile overload calculation is used, in conjunction with the calculated corrosion rate or cracking severity. This can provide much more flexibility and realism when compared to using the corrosion allowance as the upper acceptable limit, while still providing a sufficient amount of conservatism.

Similar logic is used by the API technology in cracking scenarios with the added tempering of time dependency, i.e. when was the last time the vessel was inspected for this type of cracking, as well as the effectiveness of the inspection technique to find that type of cracking. This time dependency is especially important due to the lack of agreed upon, reliable crack growth rate models and in the output recommended inspection plan timing.