Applying logistic regression to maintenance data to establish inspection intervals

As budgets are decreasing, it is imperative to select maintenance inspection interval lengths that minimize costs without risk of compromising safety or operational effectiveness issues. However, the data required are not always available or conducive to standard analytic techniques. This paper discusses the application of logistic regression to existing maintenance inspection data to establish inspection intervals. Logistic regression response variables are binary or go/no-go, variables which do not lend themselves to analysis with traditional methods. This paper presents the methodology along with pertinent results