It seems that the U.K. MoD's Principal R&M Engineer, Mr. Ian Knowles, keeps getting questions about suitable replacements for Mil-Hdbk-217. In response to the many recommendation requests he receives, Ian has compiled a set of alternative "Reliability Prediction" methods.

RELIABILITY PREDICTION USING ALTERNATIVE METHODS

1.  INTRODUCTION

1.1.   Alternative methods may be used for predicting reliability, when approved by the Project Manager. This section outlines these methods and the conditions under which they may be applied.

2.   METHODS

2.1.   The alternative methods covered by this section are:

a.   Crystal Ball
b.   Astrological
c.   Wet finger
d.   Miscellaneous techniques (Tarot cards, ouija boards, tea leaves, etc.).
e.   Witchcraft.

These will be discussed in turn.

2.2.   Crystal Ball. Crystal balls may be used, if operated by qualified personnel. Qualified personnel include gypsies and other operators approved to AQAP-1, and reliability engineers in AQAP-I approved companies. Crystal balls must be calibrated against master standards in accordance with AQAP-6.

2.3.   Astrology. Astrological prediction methods may be used, but only as applied by AQAP-1 approved astrological scientists. Use of astrological data from daily newspapers, etc., is not acceptable, as these do not necessarily meet approval criteria and are not traceable.

For astrological predictions, the dates of birth of the following must be provided:

a.   Project managers (MOD and supplier).
b.   Senior reliability engineer.
c.   Chief designer.
d.   Secretaries of above, if they are evaluated as likely to affect attitudes, motivation and performance.
e.   All the people who will make any part of the system. (These data are only required when prediction accuracy of better than 0.001% is required).

2.4.   Wet Finger. This is a simple and popular method, particularly among project managers.. Since little expertise is required, it is a low-cost method. If features some of the advantages of the witchcraft methods (see below), but the secondary effects are less unpredictable, particularly if the personnel making the predictions are replaced or forget what the prediction was.

2.5.   Miscellaneous Methods. Other methods may also be used, eg. Tarot cards, ouija boards, tea leaves, etc.), bug only be qualified staff or consultants approved as above. In each case appropriate data must be provided as required by the operator.

2.6.   Witchcraft. Witchcraft may be used. However, this method of prediction is a reliability forcing process rather than a prediction. Great care must be exercised to ensure that only achievable, conservative MTBF values are forced in this way, since death, serious injury or madness can result if personnel observe failures or running times which lead to MTBFs different to the set values.

There are no Defence or Military standards for witches, so operator selection is difficult. It is advisable to employ one or more witches on the project, preferably as part of the reliability team, if this method is to be used. However, care must be exercised to ensure that they confine their powers to the reliability requirement.

Security clearance can sometimes present further difficulties, particularly for non-NATO practitioners such as witch doctors, obeah men, voodoo priests, etc.

NOTE:   Secretaries should not be asked to perform these rites, even when appearances indicate likely competence.

3.   PREDICTION ACCURACY.

3.1.   The alternative methods described generally provide much more accurate, reliability predictions than the techniques described by Mil-Hdbk-217, Bellcore, HDRD-5, CNET and progency.