OEE vs TEEP comes down to the denominator. OEE measures how much of Planned Production Time is truly productive. TEEP measures how much of All Time, every hour of a 168-hour week, is truly productive. TEEP equals OEE multiplied by Utilization, so it shows the calendar hours a plant could reclaim before buying equipment.
Picture a line running at 65% OEE across two shifts, five days a week. Demand goes up, the schedule looks full, and a request for a second line lands on the operations VP's desk. Before anyone argues about the vendor quote, there's a simpler question to ask: which hours did the capacity case divide by?
Same loss factors, different denominators
Both metrics are built from availability, performance and quality. The difference is the time base. Vorne's oee.com TEEP page puts it plainly: "OEE measures the percentage of Planned Production Time that is truly productive," while "TEEP measures the percentage of All Time that is truly productive."
The formulas follow from that:
-
OEE = Availability × Performance × Quality, measured against Planned Production Time.
-
Utilization = Planned Production Time / All Time.
-
TEEP = OEE × Utilization.
Vorne defines All Time as "24/7 time (it includes every minute of every day)," which is 168 hours a week. Because Utilization can't go above 100%, TEEP can never be higher than OEE. IBM's TEEP explainer describes the same split, saying TEEP "takes into account all the potential operating time, including planned and unplanned downtime" and "can be helpful in production capacity planning."
The OEE side is familiar to most plant managers. In Vorne's shift example, a 480-minute shift minus 60 minutes of breaks leaves 420 minutes of Planned Production Time. Subtract 47 minutes of downtime and Run Time is 373 minutes. With Availability at 88.81%, Performance at 86.11% and Quality at 97.80%, OEE comes to 74.79%. It's a reasonable score, and it says nothing about the 16 hours that shift wasn't scheduled.
Where schedule loss and empty shifts go
Vorne's OEE factors page calls time with "no intention of running production" Schedule Loss. Its examples are "plant shutdowns, breaks/lunches, or periods where there are no orders." Schedule Loss is part of TEEP, not OEE. Lean Production agrees: "Schedule Loss is not included in OEE calculations since there is no intention of running production."
That's the design, and for running a shift well it's the right one. A supervisor shouldn't be graded on hours nobody scheduled. For a capex case, though, it means unscheduled nights and weekends drop out of the OEE view entirely while TEEP counts every one of them.
Vorne's own TEEP example shows how wide the gap gets. A line at 65.00% OEE runs 80 planned hours (8 hours × 2 shifts × 5 days) out of 168. Utilization is 47.62% and TEEP is 30.95%. The line is performing respectably during its scheduled hours and sitting idle for more than half the week. Vorne says TEEP shows how much capacity sits in the "hidden factory" and that "reclaiming time from your hidden factory is a faster and less expensive alternative to purchasing new equipment."
Changeovers and planned maintenance don't land in the same bucket everywhere
This is where two people can measure the same line and come back with different numbers.
Vorne treats changeovers as planned stops. On its calculation page, "Planned Stops (e.g., Changeovers)" count as Stop Time and reduce Availability. The reasoning is that changeover time could otherwise be used for manufacturing, so it stays inside OEE where someone has to answer for it.
Plant Services takes a different line. In "How to use OEE and TEEP to calculate plant losses" (Joe Anderson, March 31, 2026), scheduled downtime includes "Changeovers, Engineering Planned Shutdown, Maintenance Planned Shutdown, Operations Planned Shutdown, or No Customer Orders." That time sits outside OEE. The calendar factor is called Loading, defined as Scheduled Time / Total Calendar Time, and TEEP = OEE × Loading.
Anderson's weekly example: 24 hours of planned maintenance plus 4 hours of changeovers make 28 hours of scheduled downtime, leaving 140 of 168 hours and Loading of 83%. Availability is 89%, Performance 56% and Quality 99%, for OEE of 49% and TEEP of 41%. The biggest losses in that plant weren't on the schedule at all: minor stops took 36 hours and speed losses 19 hours.
Move those 4 changeover hours from scheduled downtime into Availability and OEE drops while Loading rises. Nothing on the floor changed. A capex memo that compares one line's OEE under the Vorne model with another plant's OEE under the Plant Services model is comparing two different things. Write the time model down before anyone compares numbers: which hours are All Time, which are planned, and where changeovers and planned maintenance sit.
A worked example: third shift or second line
The figures below are hypothetical, built by our editors on Vorne's 80-hour, 65% OEE case. The arithmetic has been checked.
Assume an ideal rate of 3,600 units per hour (a 1.0-second cycle). At 80 planned hours and 65% OEE, good output is 80 × 3,600 × 0.65 = 187,200 units a week. Each planned hour yields 2,340 good units.
Demand rises to 260,000 units a week. At 65% OEE, that takes 260,000 / 2,340, or about 111.1 planned hours. The two-shift schedule has 80.
| Option | Planned hours/week | OEE required or assumed | Good output capacity | Utilization | TEEP |
|---|---|---|---|---|---|
| Current state | 80 | 65% | 187,200 | 47.6% | 31.0% |
| A: Fix OEE only | 80 | about 90.3% required | 260,000 | 47.6% | about 43.0% |
| B: Add weekday third shift | 120 | 65% | 280,800 | 71.4% | about 46.4% |
| C: Buy a second line | Depends on staffing | n/a | Depends on line | n/a | n/a |
Option A asks the existing two shifts to hit 260,000 / 288,000, or about 90.3% OEE. That's above the 85% figure that Vorne's world-class OEE page says "is often thought" to be world class. The figure is credited to Seiichi Nakajima's 1984 book Introduction to TPM, and the same page says "most manufacturing companies, even today, have OEE scores closer to 60%." Vorne also says it sees more companies below 45% than above 85%, and cautions that the benchmark comes from 1970s Japanese automotive practice. Planning a capacity increase on beating world class is a bet, not a plan.
Option B adds 40 weekday hours. At unchanged 65% OEE the line can make 120 × 2,340 = 280,800 units, with about 20,800 units of headroom. Utilization goes from 47.6% to 71.4% and TEEP from 31.0% to about 46.4%. There's slack in that number too: across 120 hours, demand is met at a blended OEE of about 60.2% (260,000 / 432,000), so a third shift that runs somewhat worse than days can still close the gap.
Option C is where an OEE-only reading leads. "65% OEE, schedule full" sounds like a plant out of room. The TEEP reading says something else: even after a third shift, 48 hours a week (168 minus 120) are still unused.
What reclaiming calendar time costs
A third shift isn't free, and the capex memo should price it as carefully as the new line. We aren't putting dollar figures on these because they depend on the plant, the labor market and the product. Treat them as line items for your own numbers:
-
Labor and supervision. Operators, a shift lead, and whatever maintenance, quality and material-handling coverage the line needs to run unattended by day staff. Include any shift premium your plant pays.
-
Displaced maintenance. If preventive work used to happen during idle nights, it now has to come out of planned time. Under a Plant Services-style model that shows up as lower Loading. Under a model that books it inside OEE, it shows up as lower Availability. Either way, budget the hours.
-
Night-shift performance. Third shifts can run at lower OEE than days. The worked example shows how much drop a plan can absorb; check your own margin the same way.
-
The practical ceiling. Vorne says a plant "typically achieves only 80% to 90% Utilization of total capacity." On a 168-hour week that's roughly 134 to 151 hours, not 168. Don't size the hidden factory as if the calendar were fully usable.
Pricing the comparison
The fair comparison is the annualized cost of added shifts against the second line's capex plus its own staffing, since a new line needs operators too. On the downtime side, ManufacturingMag's downtime cost calculator takes stopped hours per incident, saleable units per hour, contribution per unit, the share of output not recovered, incremental recovery cost per incident and incidents per month. It reports lost contribution separately from incremental recovery cost, and it uses contribution, which is not total sales revenue. That separation matters when deciding whether a lost hour is worth recovering on overtime or a weekend shift at all.
Use the same contribution figure across all options. A second-line case that values lost output at revenue while the shift case uses contribution will tilt the result before anyone reads the memo.
When a second line is the right answer
TEEP doesn't always argue against new equipment. It can also confirm that the calendar is used up. A second line deserves a hard look when:
-
Utilization is already close to the 80% to 90% range Vorne describes as typical for total capacity.
-
Maintenance or sanitation windows can't move into a shorter or different slot without risk to the equipment or the product.
-
The plant can't staff another shift reliably, whatever the wage.
In those cases the remaining lever on the existing line is OEE itself. Anderson's example is a reminder of where that often starts: minor stops and speed losses, which cost that plant 55 hours in a week, more than its 28 hours of scheduled downtime.
What the capex memo should report
-
OEE, Utilization (or Loading) and TEEP, side by side, for the line in question.
-
The schedule-loss categories behind Utilization: breaks, shutdowns, no-order periods, unstaffed shifts.
-
Where changeovers and planned maintenance sit in the time model, stated in writing.
-
The demand gap in hours at current OEE, not as a percentage. "We need 31 more planned hours a week" is easier to test than "we need 39% more capacity."
-
The cost of closing that gap with added hours, priced against the cost of the new line with its staffing.