If you are responsible for keeping a facility running, the failure you want least is the one that arrives with no warning. Most of those failures did give warnings. There was just nobody set up to read them. Facilities don't skip this because it is difficult. They skip it because a working piece of equipment never asks for anybody's attention.
The four pieces of a condition-based program
Rank the equipment by consequence
What it is. Work out which components would hurt most if they failed, based on what they feed and how long a replacement takes to arrive.
The trade-off. Half a day of work that sets up everything after it. Monitoring every asset at the same intensity is how these programs get expensive and then get cancelled.
Measure the condition indicators
What it is. Infrared thermography on connections and terminations, dissolved gas analysis on oil-filled transformers, insulation resistance and power factor testing, breaker operations counts, and equipment age against rated service life.
The trade-off. None of it shows up on a walk through the electrical room, which is the point of doing it. It turns a general sense that things are fine into a number you can trend.
Plan against lead time, not against failure
What it is. Using the condition data to schedule replacement while there's still time to order.
The trade-off. Specialty equipment can take six months to a year to arrive. Knowing a transformer is degrading a year ahead is the difference between a changeout you schedule and an outage of unknown length.
Accept the inspection cost
What it is. Some testing can be done energized. Some can't, and doing it right means an outage window and lost production hours.
The trade-off. This is the real obstacle and it decides whether any of the rest gets done. Preventive spending is hard to justify against a failure that hasn't happened yet. The comparison that helps is with the unplanned version of the same outage, which costs more and comes on a schedule you don't choose.
The shift that makes it work
The whole difference is in treating aging electrical infrastructure as something you monitor rather than something you run until it quits.
Defined condition checks, a clear picture of which equipment matters most, and a replacement plan that allows for lead times is what separates the facility that swaps a transformer on its own schedule from the one that loses a line for four months waiting on a build slot.
The equipment will fail eventually whatever you do. What you get to choose is whether it happens on a date you picked.