Can you explain the Crow-AMSAA graph on Novinium’s web site?
[dt_divider style=”thin” /]
I present the Novinium failure statistics in the chart nearby. Crow was a guy that worked for the “Army Material Systems Analysis Activity” or AMSAA. Crow developed the statistical model that now bears his name and that of his employer. Crow-AMSAA or “C-A” for short is widely recognized as a preferred model to predict the reliability of complex systems that experience multiple failure mechanisms.
The x-axis is the product of the feet of cable that have been treated by Novinium and the years that have elapsed since treatment. For example, if a 328 foot (100 meter) length of cable was treated three years ago, its contribution to the cumulative treated would be 984 feet*years. The x-axis is logarithmic. Plotted against the y-axis are the failures—56 in total. A least squares regression of the failures provides a slope, or beta, of 0.64. A beta less than 1 means the failure rate is decreasing. Process and chemistry improvements, together with the improving mastery of the Novinium’s craft workers, make Novinium technology more and more reliable. That’s not to say that when Novinium began injection operations over six years ago post-injection performance was unacceptable. Novinium started where the old technology, invented by Novinium founders, reached a reliability plateau. When I did this same C-A analysis nine months ago (November 2010) the beta was 0.72. So, not only is the failure rate decreasing, but the rate of decrease is accelerating! About 99.4% of all the cables, which Novinium have treated, remain in reliable service.