Guidelines for the statistical analysis of ageing test data - Part 3: Minimum specimen numbers at different test conditions with given experimental data
Introduction: The objective of this document is to clarify how a statistical analysis can be done even with a small number of samples. When the scatter of data is sufficiently small, a statistical analysis should be possible. Generally, a statistical analysis applies to a larger number of samples similar to ordinary cases. On the other hand, this document may be useful in clarifying how the scatter of a small number of data points can be used to estimate "lifetime" when the number of specimens is limited (e.g. around five) and there are few ageing conditions (e.g. two or three conditions). Therefore, the aim is to document, for a small group of specimens with limited scatter of data, whether or not it is possible to estimate characteristics such as lifetime within a certain statistical error. If such a simulation were available, it would be very helpful for users. Manufacturers, for example, may prefer a new simplified, accelerated test method as an alternative to their current traditional test methods. A new test method would be easier and less expensive, especially if the specimens were small in size (dimension), although it may not be as suitable for large specimen sizes or expensive materials or devices (such as motors). It would be very welcome if a statistical treatment method of this type, applying to a smaller number of specimens, were defined. Though inappropriate cases are anticipated, this document deals with cases where such statistical treatment is appropriate.
Scope: This part of IEC 60493 clarifies how a statistical analysis can be done with a small number of samples. This document will be useful when the accelerated test method is difficult to carry out, for example in cases where the dimensions of test specimens (including test devices) are very large in scale or the cost of test specimens is high. Testing is facilitated by enabling users to reduce the number of test specimens.
Contents:
- FOREWORD
- INTRODUCTION
- Scope