Live data flowing in 7 days. No IT department required. See exactly what it looks like for your machines.
Calculate Your Downtime CostPractical guides, benchmarks, and shop floor strategies — no fluff, no spam, unsubscribe anytime.
The TPM Six Big Losses framework applied to CNC machining. Equipment failure, setup time, micro-stops, reduced speed, defects, and startup losses — with data on where most shops lose the most.
Reactive vs preventive vs predictive maintenance for CNC shops. What signals to monitor, how vibration analysis works, real-world savings for a 5-machine shop, and common pitfalls to avoid.
Published OEE benchmarks are misleading because they mix industries. Here are CNC-specific benchmarks by machine type — turning, milling, multi-axis, Swiss — with realistic percentile ranges.
Every shop owner knows downtime is expensive. But when you ask “how expensive?” the answer is usually a shrug and a guess. The real number is almost always worse than the guess — because the visible costs are only the beginning.
These are the numbers most shop owners know, or at least can estimate:
For a single machine running at $150/hr fully loaded with operator, one hour of unplanned downtime has a direct cost of $150. That part is straightforward. What follows is not.
The indirect costs of unplanned downtime typically run 1.5x to 3x the direct cost. Here is where the money actually goes:
When a machine goes down, the jobs scheduled on it slip. If that job has a hard delivery date — and in aerospace, automotive, and medical, they all do — you are looking at expedite charges, split shipments, or contractual penalties. One missed delivery to a Tier 1 automotive supplier can cost you the next RFQ.
The work does not disappear when the machine goes down. It shifts to other machines, other shifts, or weekends. Overtime labor at 1.5x is expensive, and weekend emergency runs are even worse. A 4-hour machine failure on Tuesday can cost you 8 hours of overtime on Saturday to recover the schedule.
This is the cost that never shows up on a spreadsheet but kills shops slowly. A buyer at a major OEM remembers who delivers on time and who does not. You do not lose the customer all at once — they just stop sending you the good jobs. The ones with volume. The ones with repeat orders. You never even know you lost them.
A CNC shop is a flow system. When Machine A goes down, the parts it was feeding to the next operation stack up. Machine B runs out of work. Operators get shuffled. The day's production plan is shot. The disruption ripples forward through the entire week's schedule.
Let us calculate the annual downtime cost for a typical 5-machine CNC job shop running one shift.
| Cost Component | Calculation | Annual Cost |
|---|---|---|
| Direct machine cost | $150/hr x 3 hrs/wk avg x 5 machines x 50 wks | $112,500 |
| Operator idle time | $35/hr x 3 hrs/wk x 5 machines x 50 wks | $26,250 |
| Overtime recovery | ~40% of downtime recovered at 1.5x rate | $31,500 |
| Expedite / logistics | Rush shipping, split orders (~$500/mo) | $6,000 |
| Repair / parts | Average $800/machine/yr for unplanned | $4,000 |
| Total visible cost | $180,250 | |
| Total with hidden costs (1.5x) | Lost customers, schedule disruption | $270,375 |
Three hours of unplanned downtime per machine per week is conservative. Industry surveys put the average closer to 4-5 hours for shops without monitoring. At 5 hours per week, the total crosses $400,000 annually for a 5-machine shop.
Across the shops we work with, the same five causes account for roughly 80% of unplanned downtime:
| Rank | Cause | % of Downtime | Detectable by Monitoring? |
|---|---|---|---|
| 1 | Tool failure / wear | 25-30% | Yes — vibration + spindle load trends |
| 2 | Material / stock shortage | 15-20% | Partially — idle time detection triggers alert |
| 3 | Programming / setup errors | 15-18% | Partially — abnormal cycle time patterns |
| 4 | Mechanical failure (bearings, spindle) | 12-15% | Yes — vibration is the primary leading indicator |
| 5 | Operator availability | 10-12% | Yes — machine idle with no job change |
The top two causes — tool failure and mechanical failure — are precisely the categories where vibration and load monitoring provide the most warning. A failing bearing does not go from healthy to seized in an instant. It degrades over days or weeks, and the vibration signature changes measurably before anyone hears or feels the difference.
The goal is not to eliminate all downtime. That is impossible. Tooling wears out. Machines need maintenance. The goal is to convert unplanned downtime into planned downtime — because planned downtime costs a fraction of what unplanned downtime costs.
A planned spindle bearing replacement takes 4 hours and costs $2,000 in parts and labor. An unplanned spindle seizure takes 3 days (waiting for parts, emergency service call, realignment) and costs $15,000 or more — plus every job that was on that machine now needs rescheduling.
Here is a real example from our lab: a vibration sensor on a Haas VF-2 detected a velocity increase from 2.1 mm/s to 4.2 mm/s over 72 hours. That is a jump from ISO 10816 Zone A (good) to Zone B (acceptable), trending toward Zone C (investigate). The alert fired at 3.5 mm/s. Maintenance pulled the machine during a scheduled break, found a worn bearing in the spindle assembly, and replaced it that evening.
Total cost: $1,800 in parts and 3 hours of planned downtime. Without monitoring, that bearing would have failed within a week, taking the spindle with it. Estimated avoided cost: $15,000 in repairs plus 3 days of unplanned downtime ($3,600 in lost production). Net savings from that single catch: roughly $16,800.
What is downtime costing your shop?
Our benchmark calculator estimates your annual downtime cost based on your machine count, shift pattern, and hourly rate. It takes 2 minutes and requires no login.
Calculate your downtime cost →Machine monitoring at $599/month for 5 machines costs $7,188 per year. If monitoring prevents a single catastrophic failure (common — most shops catch one in the first 90 days), you have paid for nearly 2 years of monitoring with one save.
But the bigger ROI comes from the aggregate: reducing unplanned downtime by even 20% across your shop. For our 5-machine example at $180,250 in annual downtime cost, a 20% reduction is $36,050 saved — a 7.5x return on the monitoring investment.
That is why we offer a 90-day ROI guarantee. If the monitoring does not pay for itself within the first 90 days, you do not pay. We have never had a shop take us up on it — because the math works every single time.
Unplanned CNC downtime costs a typical 5-machine job shop between $180,000 and $400,000 per year. The visible costs — machine time, operator wages — are less than half of it. The rest hides in overtime, missed deliveries, and customers who quietly stop calling.
Monitoring does not eliminate downtime. It converts unplanned downtime into planned downtime, catches failures before they become catastrophic, and gives you the data to attack the root causes. At $80 per machine per month, the breakeven is measured in days, not months.