Analysing recovery from pandemics by Learning Theory: the case of CoVid-19

Romney B. Duffey; Enrico Zio

It is clear in Equation (2) that for practical applications a suitable measure of the experience/knowledge or risk exposure accumulated with respect to the initial one, (ε-ε0), is needed and that the original or starting one, ε0, is a suitable arbitrary or convenient reference dependent on the problem at hand. In other words, the measure for the accumulated experience/knowledge or risk exposure is technology/system specific. Then, for data inter-comparisons, it is useful to render non-dimensional the quantities of interest, which results in the Universal Learning Curve (ULC) for the non-dimensional error rate E*= (R(ε)-Rm)/(R0-Rm) as a function of the non-dimensional experience/knowledge or risk exposure ε* = ε/ εT, where εT is the maximum accumulated experience or risk exposure thanks to which the error is recovered (the problem is considered under control), its rate having reached the lowest or minimum achievable value Rm. From (2), * =