Table 1.

PHM challenges.

Challenge
Diagnostics Large sensitivity to signal noise, dependence on environmental and operating conditions, and lack of fault detection Maintenance diagnosis is mainly a specialized process Standard methods, guidelines, and trained personnel to fully verify and validate diagnostics are lacking
Prognostics Failures are intermittent and hence difficult to predict with typical prediction methods Lack of standards for the evaluation of prognostic algorithms Methods for continuous real-time estimation of the RUL need development Tools for PHM designers to know how PHM systems impact the total logistic system are lacking Limits to the accuracy and precision of prognostics due to uncertainties Multiple failures may complicate the prediction of RUL
Components and PHM Architecture Difficulties exist for consistent data, communication, and security across a plant floor Improvement in the relationship between controls engineering and maintenance engineering is needed Lack of integration of PLC information and PHM capabilities Deficient guidance to deal with abnormal events in complex systems and processes Prognostics capability for electronics are lacking
Business Level Maintenance is usually regarded as a net cost and not as a net benefit; it is difficult to quantify the cost savings due to PHM
Human Factors Resistance due to culture, norms, expertise, and customer and supplier relationships Creation of user-friendly PHM applications Incorporation of subjective information from experienced workers/industry with faults