Ask a plant head what electricity cost last month and you will get a number to the rupee. Ask what breakdowns cost and you get a shrug — "a lot". That asymmetry decides budgets: the measurable bill gets managed, the unmeasured one gets absorbed.
The 3-line formula
- Lost contribution = rated output/hour × contribution margin per unit × downtime hours
- Committed cost of idle resources = operators + machine overheads still running while output is zero
- Recovery cost = overtime, expedited freight, rejected startup batches
A CNC rated at 40 pcs/hour with ₹150 contribution per piece loses ₹6,000/hour in contribution alone. A 3-hour bearing failure is an ₹18,000 event before overtime — and a plant averaging one such event per machine per month across 25 machines is quietly absorbing ₹4.5 lakh monthly.
Why the gut feel is always low
Three reasons. First, downtime is recalled, not recorded — memory rounds a 4-hour outage down to "a couple of hours". Second, waiting time hides: the machine was down at 9:15 but the repair "started" when the fitter arrived at 10:40, and nobody counts the gap. Third, small stops never get reported at all, and small stops are usually the biggest bucket.
What to do this month
- Fix the two numbers per critical machine: rated output/hour and downtime cost/hour. One meeting with production and accounts settles both.
- Record breakdowns at the moment they happen, with the real reported time — not in the evening from memory.
- Split the clock: reported → response → repair. The split tells you whether to fix the process or the machine.
Do just this and next month you will have a Pareto: which five machines ate 80% of the money. That list — not a new machine budget — is where maintenance transformation starts.
AssetAI does the recording part automatically: operators report by scanning the machine QR, every timestamp is captured as people work, and the loss shows up in rupees per machine without anyone maintaining a register.