Estimation of the maintenance cycle from small sample based on a local fault interval model

Maintenance cycle is an important factor of a repairable system, and it is closely related to the safety operation and the level of energy consumption of the system. Two methods are presented to estimate the maintenance cycle from small sample based on a local fault interval model, one is the multi-point statistics method, and the other is the difference of sigmoidals method. Both methods are tested by real sample data from a portal crane. It is found that both methods are able to estimate the maintenance cycle of a repairable system. Furthermore, these two methods have a smaller probability to invoke the false positive problem than the three-point statistics method, while the difference of sigmoidals method is more robust than the multi-point statistics method.