Statistical Tools for Ultrasonic Thickness Data Analysis, Part 1: Piping Thickness Analysis

Introduction

Ultrasonic thickness monitoring programs represent one of the most intensive inspection activities in refining and petrochemical facilities. Despite numerous improvements in ultrasonic testing equipment and inspection techniques, however, there has been little advancement in analyzing this valuable data since the early 1980’s.

In this multi-part series, we will review statistical tools for analyzing ultrasonic thickness data. Most of the examples will focus on piping systems, which often account for the largest volume of data and demand the most inspection resources. The statistical techniques described, however, are easily extended to thickness data from any type of fixed equipment. In Part 1, we’ll review some basic concepts and use probability plots to estimate the minimum thickness in a piping system. These concepts will be expanded in Part 2, utilizing probability plots to obtain piping system remaining life estimates. Other fixed equipment applications, such as estimating exchanger bundle failure frequencies, predicting maximum pit depths and projecting tank bottom remaining life will be discussed in Part 3.

Carefully acquired and properly analyzed thickness data can yield many benefits. Long- range reliability, inspection, and replacement forecasts are possible when accurate remaining life projections are made. Better forecasting translates into reduced process risk as well as lower inspection and replacement costs.