The most powerful way to improve maintenance of assets is to know the impact of maintenance tasks on business performance. This requires that each task is associated with a failure mode and effect. The association needs to be quantified so the maintenance decision maker knows the benefit of doing each task in terms of cost, risk and impact on operations.
So failure effects need to be defined in dollars per hour or event, the likely frequency can be expressed using parameters of the bathtub curve (Weibull), and the cost of performing tasks is given by cost of resources and durations. Now, the maintenance decision maker can evaluate the worth of doing planned maintenance, inspection, monitoring or running to failure. This provides the Reliability practitioner the means to talk to budget owners, planners, operations and risk managers. The best thing about using this approach is that now we can play "what if"?