An unplanned kiln stop rarely costs what the downtime log says. Before a finance team signs off on reliability AI, it needs a plain view of where the money goes: clinker not made, fuel burned to reheat, hours spent ramping back to rate, and stress on the lining. This page builds that figure phase by phase from assumed inputs you replace with your own, then turns it into payback, NPV and a one-page board summary. To run the numbers with your kiln data, book a short session.
Put a Defensible Price on Every Unplanned Kiln Stop
iFactory AI Analytics links your stop log, kiln rate and cost inputs into one ROI model. Move any assumption and the annual value moves with it, so the CFO can test each number instead of taking it on trust.
Five Phases Inside One Unplanned Kiln Stop
Most stop logs record the minutes between trip and feed-on. The cost runs longer than that. A kiln that trips has to be held, reheated before feed goes back in, brought up to rate through hours of unstable running, and then lives with whatever the temperature swing did to the coating and bricks. Each phase has its own cost driver, so price them separately. Our support team can map these phases to the tags in your DCS.
Refractory specialists describe thermal cycling from stops and restarts as harder on the lining than steady high-temperature running. Short stops add up across a campaign. Put a per-stop allowance in the model rather than ignoring it, and agree the figure with your refractory supplier.
One 12-Hour Stop Priced Phase by Phase
Here is a single illustrative stop on a 4,800 tpd kiln, which makes 200 tonnes of clinker an hour. Every input is assumed and labelled. The result gives the hourly figure that the hero slider uses, so the whole page runs on one chain of arithmetic. Swap in your own figures in a working session.
Use contribution margin only if the plant is sold out and every tonne lost is a sale lost. If clinker stock covers the gap, value it at the cost of buying or making it later. If neither applies, the two lost-clinker lines, hold and ramp, may be close to zero, and the case rests on fuel, lining and repair.
Kiln Size Sets the Scale of Every Number
The same 12-hour stop means very different tonnage on different lines. The EU reference document for cement BAT describes a typical kiln as around 3,000 tonnes of clinker a day, and suppliers have built single kiln systems of 12,000 tonnes a day. Start from your own rated output, not a benchmark. Ask our support team how to pull actual hourly rates from your historian.
Mill Stops Cost Little or a Lot, Depending on the Silos
A cement mill stop is priced very differently from a kiln stop. When clinker stock and cement silo space are healthy and the mill has spare hours in the week, lost grinding can be caught up later, often at the cost of extra power and overtime. When silos are low and the mill is already the bottleneck, every hour down is cement not shipped. Check which case applies before you count it, and see how the model handles both in a live walkthrough.
- 10-hour stop on a 150 t/h mill, assumed
- Lost sales: none, the tonnes are ground later
- Cost: repair plus catch-up running, $8,000 assumed
- Count it under repair cost, not lost margin
- Same 10-hour stop: 1,500 t of cement not ground
- At $30 margin per tonne, assumed: $45,000
- Plus repair: $8,000. Stop total: $53,000
- Hourly lost sales: 150 t × $30 = $4,500
A full clinker yard lets the cement mills keep grinding through a kiln stop, so shipments may not suffer. In that case the kiln stop cost is mostly fuel, ramp-up and lining, not lost cement sales.
Let Finance Question Every Input
iFactory AI Analytics keeps each assumption visible and linked to its source: the stop log, the kiln rate, your margin and your fuel price. When the CFO challenges a number, change it and the whole model follows.
Six Savings Lines, Graded by the Evidence Behind Them
Reliability AI can save money in several places, but the evidence for each is not equally firm. Grade every line before it goes into the case. Hard lines come straight from your own records. Firm lines need one agreed assumption. Soft lines are real but hard to prove, so keep them as notes rather than totals. Our support team can help you grade them.
How the Annual Figure Moves With the Avoidable Share
The share of stop hours a reliability program can avoid is the most uncertain input, so show the board how much it matters. With the example's $12,500 per stop hour and 96 stop hours a year, the annual figure scales directly with that share. Review your own stop causes with our team in a stop-log review.
Payback and NPV From the Same Inputs
Payback tells the board how quickly the money comes back. Net present value tells it what the program is worth today after the time value of money. Both use the same inputs. Here, the annual benefit adds the kiln figure to the mill figure: 30 unbuffered mill hours a year at $4,500, with 30% avoidable. Investment and running costs are assumed placeholders until you have a quote, which our team can provide.
Use your company's own discount rate and evaluation period. If the case only works at the high end of the avoidable share, say so on the board page rather than hiding it.
A One-Page Summary the Board Can Read in Two Minutes
Boards approve cases they can follow. Keep the summary to one page, show where every number comes from and state the biggest uncertainty plainly. The fields below follow the worked example. Fill them from your own model, or have iFactory produce the page from your data in a short demo.
A Fixed-Scope Server That Firms Up the Cost Line
A business case is only as firm as its investment line. iFactory ships a pre-configured NVIDIA AI server, racked and ready, with analytics and ROI models pre-loaded and your kiln and mill tags mapped during setup. You rack it, plug in power and Ethernet, and the AI is live. Data stays on site, and the scope is fixed up front. Ask our support team for a turnkey quote.
- Pre-configured NVIDIA AI server, racked and ready
- Links to DCS, historian and stop log
- Failure models for kiln drive, fans and mills
- ROI model with every assumption editable
- Board summary built from live plant data
- Operator and engineer training
- 24×7 remote monitoring of models and hardware
Server installed, data links made, 12 months of stop history loaded and coded.
Models trained on your assets and run alongside the team. ROI inputs agreed with finance.
Alerts switched on, staff trained, first quarterly stop review scheduled.
Cement Plant AI ROI Calculator Software With the Plant Data Behind It
iFactory combines reliability analytics with the ROI model, so the value case is rebuilt from real stop data every quarter instead of written once and forgotten. See it in a 30-minute demo.
What Cement CFOs and Plant Managers Ask About AI ROI
Take a Board-Ready Kiln Case to Your Next Review
Bring your stop log and cost inputs. iFactory prices each kiln stop phase by phase, separates buffered from unbuffered mill time and gives you payback, NPV and a one-page summary built on your own data.
