Skip to content
Industry reportsAll articles

OEE: What Overall Equipment Effectiveness Really Tells You

Overall equipment effectiveness (OEE) multiplies Availability, Performance, and Quality into one score. Here is the ISO 22400-2 math, TEEP vs OOE worked examples, Six Big Losses mapped to PackML PLC states, and how to build a real-time OEE stack.

Artem Zaitsev
Artem ZaitsevCTO · Idealogic
Author: Artem ZaitsevPublished: Updated: Reading time: 26 minHow we write →
On this page
Idealogic: OEE, overall equipment effectiveness explained

OEE (overall equipment effectiveness) is the percentage of planned production time in which a machine makes good parts at its ideal speed. The OEE formula is Availability × Performance × Quality, so 90% × 95% × 99% gives 84.6%, and 100% would mean no unplanned stops, no slow cycles, and no scrap. That's the OEE meaning in one line. The metric is a KPI defined in ISO 22400-2:2014 (a revision is in draft), and it comes from Seiichi Nakajima's Total Productive Maintenance (TPM) work at the Japan Institute of Plant Maintenance (JIPM).

So what is OEE in manufacturing practice? Usually one number on a plant wallboard, and an easy one to manipulate. A factory manager can raise reported OEE from 62% to 79% in an afternoon without fixing a single machine: reclassify die changeovers as "planned downtime", or pad the ideal cycle time in the ERP routing table. This guide covers the OEE calculation next to OOE and TEEP, with worked examples on a discrete and a continuous line. It also shows how the Six Big Losses map to ISA-TR88.00.02 PackML and SEMI E10 state models, how to detect micro-stops under two minutes at the edge, and how to build a real-time OEE software stack that runs from PLC registers to shift scorecards.

The short version

  • The ISO 22400-2 multiplication: OEE = Availability × Performance × Quality, which mathematically collapses to Good Count × Ideal Cycle Time ÷ Planned Production Time. If the factor product and the direct ratio disagree, your downtime or cycle-time logic has a bug.
  • Boundary defense via OOE and TEEP: OEE only judges active planned production time. Tracking OOE (including planned maintenance/changeovers) and TEEP (against all 24/7 calendar hours) stops teams from hiding changeover bloat outside the OEE denominator.
  • Automated Six Big Losses attribution: Mapping PLC registers to ISA-TR88.00.02 PackML states (Execute, Held, Suspended, Aborted) and the Starved and Blocked PackTags separates internal breakdowns from upstream starvation and downstream blockages without biased paper logs.
  • High-frequency micro-stop capture: Minor stops under 120 seconds and 10% spindle feed-rate dial-backs often cost more capacity than major breakdowns. You only catch them with sub-second edge polling over OPC UA or MTConnect.
  • Context-aware benchmarks: The classic 85% benchmark (90% x 95% x 99%) fits dedicated, long-run lines. A high-mix discrete job shop running six setups a shift can't reach the availability of a continuous chemical plant, and it shouldn't be judged against it.
  • Execution integration & 2026 TCO ($75/hr CEE vs $165/hr US): OEE belongs inside or alongside your manufacturing execution system (MES), feeding breakdown events to predictive maintenance software and net capacity to production planning software, as one layer of a wider smart factory setup. Engineering an owned OPC UA / MTConnect + Sparkplug B + TimescaleDB OEE stack costs $84,000 to $270,000 in CEE ($184,800 to $594,000 US), saving $573,800 to $911,600 over 36 months across 3 plants (90 machines) against Tulip, QAD Redzone, and MachineMetrics at the vendor rates we assume below.

Exact mathematical formulation: OEE vs OOE vs TEEP

OEE math starts with a time hierarchy. A common way to partition calendar time, one that maps onto the ISO 22400-2:2014 model and the six equipment states of the semiconductor standard SEMI E10, is four nested buckets. Fix them in software first and you avoid rounding drift and denominator manipulation later:

  1. Total Calendar Time (T_cal): All hours in the period (24 hours/day × 7 days = 168 hours/week, or 10,080 minutes).
  2. Scheduled Operations Time (T_sched): Calendar time minus unscheduled hours when the plant is closed, no crew is rostered, or there are no customer orders (No Demand).
  3. Planned Production Time (T_plan): Scheduled Operations Time minus Planned Downtime (scheduled preventive maintenance windows, statutory safety checks, and scheduled breaks where the line is intentionally shut down).
  4. Actual Run Time (T_run): Planned Production Time minus Unplanned Downtime (equipment breakdowns, tool failures, unplanned starvation, and setup/changeover time).

OEE calculation: the three core equations

How to calculate OEE in four steps: divide run time by planned production time (Availability), compare actual output with what the ideal cycle time allows (Performance), divide good count by total count (Quality), then multiply the three.

Given Ideal Cycle Time (C_ideal) (the minimum time required to produce one unit at maximum sustainable rated speed), Total Count Produced (N_total), and Good Count Produced (N_good = N_total - N_scrap - N_rework):

Availability (A) equals T_run divided by T_plan. Performance (P) equals total count times ideal cycle time, divided by T_run. Quality (Q) equals N_good divided by N_total. OEE is the product of the three, and the terms cancel down to N_good times C_ideal, divided by T_plan.

That algebraic cancellation makes a good unit test for an OEE codebase. T_run and N_total cancel out in the final product, leaving Valuable Operating Time (N_good × C_ideal) divided by Planned Production Time (T_plan). If a dashboard reports A = 88%, P = 94%, Q = 98% (OEE = 81.1%), yet the machine produced fewer good parts than 0.811 × (T_plan / C_ideal), the software is misclassifying run time or mixing SKU cycle times across shifts.

Why you must compute OOE and TEEP alongside OEE

Because OEE uses T_plan as its denominator, any stop labeled "Planned" vanishes from Availability. ISO 22400-2 defines OEE only. OOE and TEEP are widely used TPM conventions that change nothing but the time base. Compute all three in parallel so nobody can hide a 90-minute product changeover or a cleaning window inside Planned Downtime:

  • OEE (Overall Equipment Effectiveness): Denominator is Planned Production Time (T_plan). Judges execution when the line is scheduled to run.
  • OOE (Overall Operations Effectiveness): Denominator is Scheduled Operations Time (T_sched), including planned maintenance, cleaning, and scheduled changeovers during manned shifts.
  • TEEP (Total Effective Equipment Performance): Denominator is Total Calendar Time (T_cal) (24/7/365), revealing how much capacity sits in unscheduled shifts before buying new capital equipment.

Worked numerical comparison: high-mix discrete CNC cell vs continuous packaging line

Consider a 24-hour day (1,440 calendar minutes) across two industrial lines:

  • Line A (High-Mix Discrete CNC Cell): Scheduled for two 8-hour shifts (T_sched = 960 min; 480 min unscheduled night shift), with 60 min of planned preventive maintenance (T_plan = 900 min). Runs 4 SKUs requiring 4 changeovers totaling 140 min plus 55 min of unplanned hydraulic/tool-changer faults (T_run = 705 min). At C_ideal = 1.50 min/part (theoretical output = 470 parts), it produces 418 parts due to micro-stops and feed-rate dial-backs, and scraps 11 parts (N_good = 407).
  • Line B (High-Speed Rotary Bottling Line): Scheduled across three 8-hour shifts (T_sched = 1,440 min), with 90 min of planned CIP sanitation (T_plan = 1,350 min). Loses 45 min to labeler jams and 30 min to one changeover (T_run = 1,275 min). At 400 bottles/min (C_ideal = 0.0025 min), theoretical output is 510,000 bottles, actual output is 482,000 bottles, and vision rejects are 3,800 (N_good = 478,200).
Metric / Time ComponentFormulaLine A: High-Mix CNC CellLine B: Continuous Bottling Line
Calendar Time (T_cal)Total 24h clock1,440 min1,440 min
Scheduled Shift Time (T_sched)T_cal - Unscheduled Shifts960 min (2 shifts)1,440 min (3 shifts)
Planned Production Time (T_plan)T_sched - Planned PM / CIP900 min1,350 min
Actual Run Time (T_run)T_plan - (Breakdowns + Setups)705 min (140m setup + 55m fault)1,275 min (30m setup + 45m jam)
Availability (A_OEE)T_run ÷ T_plan78.3% (705 ÷ 900)94.4% (1,275 ÷ 1,350)
Performance (P)(N_total × C_ideal) ÷ T_run88.9% (627 min ÷ 705 min)94.5% (482,000 ÷ 510,000)
Quality (Q)N_good ÷ N_total97.4% (407 ÷ 418)99.2% (478,200 ÷ 482,000)
OEE (vs T_plan)A_OEE × P × Q67.8% (610.5 val min ÷ 900)88.6% (1,195.5 val min ÷ 1,350)
OOE (vs T_sched)(N_good × C_ideal) ÷ T_sched63.6% (610.5 ÷ 960)83.0% (1,195.5 ÷ 1,440)
TEEP (vs T_cal)(N_good × C_ideal) ÷ T_cal42.4% (610.5 ÷ 1,440)83.0% (1,195.5 ÷ 1,440)

Line A posts 67.8% OEE but only 42.4% TEEP. If sales demands 35% more machined housings next quarter, the plant doesn't need a second $650,000 CNC center. Adding a third shift and cutting setup time by 40% via SMED unlocks over 50% more output on the existing spindle.

Mapping the Six Big Losses to PLC registers and ISA-TR88.00.02 (PackML) states

The Six Big Losses are the six standard categories of equipment loss, two each for Availability, Performance, and Quality. They come from JIPM's Total Productive Maintenance framework. The software job is to capture these six buckets from PLC state transitions and not from operator clipboards.

Mapping Nakajima's Six Big Losses and the TEEP schedule boundary directly to PLC state machines and inline quality sensors.

ISA-TR88.00.02 (PackML) standardizes a 17-state machine: Clearing, Stopped, Starting, Idle, Suspended, Execute, Stopping, Aborting, Aborted, Holding, Held, Unholding, Suspending, Unsuspending, Resetting, Completing, and Complete. Unit modes are separate from states. The base set is Production (1), Maintenance (2), and Manual (3), and a site can add its own, such as Changeover. The 2022 revision also adds Starved and Blocked PackTags. Connecting over OPC UA or MTConnect maps each loss to deterministic PLC signals:

  1. Equipment breakdowns (Availability Loss #1): Mode is Production (UnitModeCurrent == 1) and state drops from Execute (6) to Aborted (9) or Held (11) with a latching PLC fault integer (FaultCode > 0) for ≥ 120 seconds. The OEE service timestamps the drop to the millisecond, logs the PLC first-out fault code, and opens a reactive ticket in the CMMS software when technician dispatch is required.
  2. Setup and changeovers (Availability Loss #2): Operator scans the next work order in the MES or switches the PLC to a user-defined Changeover mode (for example UnitModeCurrent == 4, the value is site-specific). The timer runs from the last good part of Order A to the first verified good part of Order B at rated speed, comparing actual setup minutes against the production planning software changeover matrix.
  3. Idling and micro-stops (Performance Loss #1): State enters Suspended (5, external material starvation or downstream backup) or Held (11) for 3 to 119 seconds, or cycle inter-arrival time (Δ t = t_n - t_n-1) exceeds 1.5 × C_ideal. The 120-second cut-off is a convention used here and should be configurable. oee.com describes minor stops as typically a minute or two. No operator logs a 12-second stop on a tablet, so the edge agent attributes micro-stops automatically from the PLC starvation and blockage bits (InfeedStarved_Bit, OutfeedBlocked_Bit).
  4. Reduced speed cycles (Performance Loss #2): Machine stays continuously in Execute (6), but rolling median cycle time C_actual exceeds C_ideal or the MTConnect PATH_FEEDRATE data item with subType OVERRIDE (the controller's feedrate override) reads < 100%. That separates speed dial-backs from jam stops.
  5. Steady-state process defects (Quality Loss #1): Increments on RejectCounter_DINT from an inline checkweigher, vision camera, or test stand, or a part scanned as SCRAP/REWORK in quality management software. In strict ISO 22400-2 math, reworked parts fail First-Pass Yield (FPY) and penalize Quality on their first pass.
  6. Startup and warmup scrap (Quality Loss #2): Rejects accumulated during First-Article Qualification or the initial warmup shots while barrel heaters or web tensioners stabilize after a changeover.
Six Big Losses CategoryOEE FactorTypical WindowAutomated PLC / PackML Detection Logic
1. Equipment BreakdownsAvailability (A)≥ 120 sec stopState == Aborted/Held + FaultCode > 0 for ≥ 120s
2. Setup & ChangeoversAvailability (A)10 min to 4 hrLast good part of Order i to first good part of Order i+1
3. Idling & Micro-StopsPerformance (P)3 sec to 119 secState == Suspended/Held < 120s or cycle Δ t > 1.5 × C_ideal
4. Reduced Speed CyclesPerformance (P)Continuous driftRolling median C_actual > C_ideal during continuous Execute
5. Startup / Warmup ScrapQuality (Q)First N shotsRejectCount delta during First-Article or warmup window
6. In-Process DefectsQuality (Q)Steady-state runInline vision/gauge RejectCount delta or MES rework scan

What is a good OEE score? Why 85% is a trap without process context

A good OEE score depends on the process type. About 85% is the usual world-class mark for dedicated, long-run lines, and a high-mix plant can be well run far below that.

Executive presentations routinely cite 85% OEE (90% Availability x 95% Performance x 99% Quality, which is 84.6% before rounding) as "World Class." The numbers come from Nakajima's 1984 book Introduction to TPM, and as oee.com points out, they're rooted in 1970s Japanese automotive plants. The same source says most manufacturers score closer to 60%.

Apply an 85% target blindly to a high-mix job shop and the dysfunction is immediate: supervisors batch easy parts into giant runs to dodge setups, reject urgent small-lot customer orders, and inflate warehouse WIP while shipment dates slip.

Indicative OEE manufacturing benchmarks: typical plants sit near 60% (an oee.com observation, not a measured median), and reaching the 85% world-class bar takes compounded gains across Availability (90%), Performance (95%), and Quality (99%).

Because OEE multiplies three ratios, the compounding is harsh. If Availability, Performance, and Quality each sit at 80%, total OEE is 0.80^3 = 51.2%.

The ranges in the table below are indicative rules of thumb by process type, not a published dataset. Use them to set expectations, then benchmark against your own history.

Manufacturing ArchetypeTypical Baseline OEETop-Quartile OEEDominant Loss Driver & Structural Physics
Continuous Process (Refining, Paper, Float Glass, Bottling)78% to 86%92% to 96%Rare changeovers (campaigns run days); primary loss is speed drift or mechanical trips
High-Volume Dedicated Discrete (Auto Stamping, Tier-1 Cells)65% to 74%85% to 89%Dedicated tooling; primary loss is micro-stops across linked transfer stations
Batch Process & Packaging (Food, Pharma Tableting, Coatings)55% to 66%75% to 82%Allergen/CIP washdowns, regulatory line-clearance checks, and warmup scrap
High-Mix Job Shop (Aerospace CNC, Custom Sheet Metal, EMS)42% to 55%65% to 72%3 to 8 setups per shift, first-article CMM holds, and small lot sizes (10-100 units)

Bottleneck theory and line-level OEE: never average serial machines

A common mistake in multi-stage factories is computing OEE for five serial stations (Station 1 to Station 5) and averaging them to report "Line OEE."

In an unbuffered five-machine line, whenever Station 3 (the bottleneck ultrasonic welder) faults for 5 minutes, Stations 1 and 2 soon stop in a Blocked state and Stations 4 and 5 stop in a Starved state. Charging those 5 minutes against all five stations turns one 5-minute breakdown into 25 machine-minutes of loss. Conversely, raising OEE on Station 1 (a non-bottleneck with 30% excess capacity) from 75% to 90% yields zero extra finished units at pack-out while Station 3 stays the constraint. It just overflows the WIP buffer.

Following Goldratt's Theory of Constraints, and in line with NIST's work on ISO 22400 KPI structures, multi-machine lines need two rules:

  1. Measure Line OEE at the Pacemaker / Bottleneck Station: Compute line output at the constraint (or final pack-out) using the bottleneck's ideal cycle time C_bottleneck.
  2. Propagate Starvation and Blockage Causality: Use PackML's Starved and Blocked PackTags to trace stop cascades back to the upstream or downstream station that entered Held or Aborted first, directing buffer design and predictive maintenance software to the true constraint.
Tired of OEE spreadsheets that drop 15 points when audited?
We wire PLCs and CNCs over OPC UA and MTConnect into real-time OEE and MES pipelines that attribute micro-stops, starvation, and changeover losses automatically.
Explore our manufacturing software work

Architecting a real-time OEE software stack: from PLC tags to shift scorecards

When we engineer industrial telemetry platforms at Idealogic, we focus on three problems an off-the-shelf OEE dashboard widget misses: high-frequency edge pulse deduplication, stream-table context joins, and retroactive downtime splitting.

1. Edge pulse accumulators and cycle-delta histograms

Never count parts by polling a boolean PhotoEye_Bit from the cloud every two seconds; on a 180-pack/min line (333 ms/unit), HTTP polling misses 80% of parts and double-counts vibrating cartons. Instead, the PLC increments a hardware-debounced 32-bit unsigned accumulator (GoodPartCount_UDINT, ScrapPartCount_UDINT). The local edge gateway polls via OPC UA or MQTT Sparkplug B every 250 ms, computing delta_N = (Count_t - Count_{t-1}) mod 2^32 to survive rollovers and logging exact millisecond cycle deltas (delta_t_cycle) to separate speed drift from micro-stops.

2. Time-series storage and ERP/MES context joins

Raw PLC states have no meaning until joined with Shift Calendars (active shift vs break vs unscheduled hours) and Work Order Routing Context (active SKU and C_ideal from the MES or manufacturing ERP). We land state intervals and 1-minute counter rollups in TimescaleDB (from Tiger Data) or InfluxDB, and continuous aggregates serve shift, line, and SKU scorecards in under 100ms. Our ClickHouse vs TimescaleDB comparison covers the storage trade-offs.

3. Retroactive operator downtime splitting

When a line stops from 10:12 to 10:34 (22 min), a technician often arrives at 10:40 to split the event: 8 minutes waiting for a forklift (Material Handling) and 14 minutes replacing a shear-pin (Mechanical Breakdown). Our ledger models downtime as mutable intervals ([10:12, 10:20) and [10:20, 10:34)) backed by an append-only audit trail that recalculates shift OEE asynchronously on edit.

When plants switch from shift-end Excel sheets to direct PLC telemetry over OPC UA, their reported OEE almost always drops by 12 to 18 percentage points on day one. Plant managers panic until they realize nothing on the floor got worse; the measurement simply stopped lying about micro-stops and changeover overruns. You cannot recover hidden capacity until your software refuses to hide it.

Artem Zaitsev · CTO, Idealogic

In our AI industrial reporting and analytics system, we built high-throughput ingestion pipelines that consolidate raw operational telemetry and shift records into verifiable performance analytics without manual spreadsheet stitching.

Four ways OEE gets gamed (and how software prevents them)

Even with automated PLC connections, OEE implementations fail when the governance rules allow subtle manipulation. Lock down four vectors in the data model:

  1. Padding Ideal Cycle Time (C_ideal): If a stamping press sustainably runs at 60 strokes per minute (1.0 s/part) but the ERP routing holds a padded finance standard of 42 strokes per minute (1.43 s/part), the machine can lose 25% of its shift to micro-stops and still report 107% Performance. Software fix: Separate the Finance Standard Cost Cycle Time from the Engineering OEE Ideal Cycle Time, setting C_ideal automatically to the 95th-percentile sustained speed (p5 cycle duration) demonstrated by that machine-SKU pair over the last 90 days.
  2. Hiding changeover overruns in Planned Downtime: If a mold changeover is budgeted for 30 minutes and takes 75 minutes, deducting all 75 minutes from Planned Production Time (T_plan) erases a 45-minute Availability loss from the scoreboard. Software fix: Keep all product changeovers inside T_plan as an Availability loss (standard JIPM TPM), or cap planned setup deductions strictly at the standard SMED allowance and charge every minute past the timer to Unplanned Setup Overrun.
  3. The 200-reason-code dropdown trap: Giving operators a four-level nested tree of 250 downtime reason codes guarantees they will tap the top item on the list (Mechanical - Other) every time just to clear the prompt and restart the line. Software fix: Capture equipment fault codes automatically from the PLC first-out register and limit human HMI selection to 8 to 12 high-contrast buttons tailored to that specific work center.
  4. Chasing OEE at the expense of Schedule Adherence: Running 50,000 units of one easy SKU inflates OEE by avoiding setups while starving manufacturing inventory software of the dozen other SKUs required for Friday customer shipments. What to do in software: pair OEE on every scorecard with Schedule Adherence, First-Pass Yield (FPY), and SMRP-standard MTBF (Mean Time Between Failures) and MTTR (Mean Time To Repair).

OEE software build vs buy: standalone SaaS vs integrated MES telemetry

Standalone OEE tracking software gets a pilot running fast. OEE monitoring built into your MES data layer takes longer to stand up and costs less to scale. The table shows where the two differ.

Evaluation DimensionStandalone Cloud OEE Tracker SaaSCustom Edge-to-Historian OEE & MES Layer (Idealogic Build)
Machine ConnectivityBasic digital I/O pulse puck or simple Modbus counterNative OPC UA, MTConnect, Sparkplug B, and brownfield PLC drivers (S7, EtherNet/IP)
SKU & Cycle-Time ContextManual SKU selection on an iPad at the stationAutomated sync with manufacturing ERP and MES active work-order routings
Line Causality (PackML)Treats each machine as an isolated islandPropagates Starved and Blocked PackTags across serial lines to isolate bottleneck root causes
Licensing EconomicsRecurring per-machine or per-station fees (Tulip lists $100 to $250 per interface/month)100% IP ownership; $0 per-machine recurring license fees across all plants

MachineMetrics vs Tulip vs QAD Redzone & Vorne XL: 2026 36-Month TCO & Custom OEE Cost

When a discrete or batch-process manufacturer expands beyond a single 10-machine pilot toward a 3-plant, 90-machine enterprise footprint, per-machine and per-station SaaS licensing can push plant telemetry toward seven-figure spend over three years. Operations leaders typically compare four commercial paths against an owned IEC 62541 OPC UA / ANSI/MTC1.4 MTConnect + MQTT Sparkplug B (ISO/IEC 20237) edge collector, TimescaleDB historian, ISO 22400-2 KPI engine, and operator tablet Andon UI wired directly into MES work-order routings and predictive maintenance software:

Evaluation DimensionMachineMetricsTulipQAD Redzone (vs Vorne XL)Owned OPC UA / MTConnect + Sparkplug B + TimescaleDB Engine
Primary Shop-Floor FitCNC and discrete machine monitoring through a plug-in edge deviceComposable no-code frontline operator work-instruction & station appsPlant-wide huddle/coaching methodology + iPad OEE (Vorne XL = standalone LED hardware scoreboard)Unified multi-plant OT/IT historian, PackML line causality & tablet/LED Andon UI
Protocol & Line CausalityStrong MTConnect / FANUC FOCAS; isolated per-machine focus on serial linesFlexible station triggers; high-rate 250 ms sub-second micro-stop math needs custom glueManual/sensor hybrid line tracking (Vorne XL uses dry-contact pulses per isolated machine)Native IEC 62541 OPC UA, ANSI/MTC1.4 MTConnect, ISO/IEC 20237 Sparkplug B, S7 & EtherNet/IP
ERP / MES Routing ContextWebhook/API exports; dynamic SKU C_ideal overrides need middlewareRequires enterprise connector tier + citizen-app maintenance across plantsSiloed plant database outside core ERP/APS (Vorne XL requires manual barcode/web sync)Real-time stream-table joins against ERP/MES routings, shift calendars & PackML Starved/Blocked tags
Commercial Pricing ModelQuote-based, no public rate (modeled here at $320/machine/mo) + edge hardware & onboarding fees$100 to $250/interface/mo list price, Enterprise quote-based + connector add-onsQuote-based (modeled here at $6,500/plant/mo); Vorne XL: one-time hardware, $0 SaaS, budgeted here at $4k to $5.5k/machine$0 per-machine & $0 per-station SaaS tax; 100% owned IP across all plants

Pricing notes: Tulip's plans page lists Essentials at $100 and Professional at $250 per interface per month, with Enterprise quote-based. MachineMetrics and QAD Redzone don't publish rates, and we found no public list price for Vorne XL hardware. So the MachineMetrics, Redzone, and Vorne figures in this section are our budgetary assumptions. Get real quotes before you rely on them.

  • When commercial OEE tools or hardware scoreboards win: Choose MachineMetrics to instrument 10 to 20 CNC spindles in two weeks with zero OT engineering staff, Tulip when operator-guided assembly apps matter more than high-frequency PLC state historians, QAD Redzone when buying an external Lean shop-floor coaching transformation, or Vorne XL (budgeted at $360,000 to $495,000 one-time hardware across 90 machines) when simple standalone LED pace boards satisfy single-station packaging lines without central TimescaleDB historian analytics.
  • Where per-machine SaaS breaks across 3+ plants: At 90 machines across three plants, modeled subscriptions of $18,000 to $28,800/mo add up to $648,000 to $1,036,800 in fees over 36 months, and the bill grows with every brownfield conveyor, press, or inspection cell you connect to your Unified Namespace.

Below is the worked 36-month OEE TCO matrix for a 3-plant, 90-machine manufacturer. Vendor rates are the budgetary assumptions above, and tablet mounts and plant Ethernet cabling are excluded equally across all columns:

36-Month TCO Line Item (3 plants, 90 CNC/PLC machines & operator stations)MachineMetrics (90 machines)Tulip (90 stations)QAD Redzone (3 plants)Owned OPC UA / MTConnect + Sparkplug B + TimescaleDB Stack
1. Edge hardware, rollout or custom build CapEx$95,000 (edge gateways, I/O kits & 90-machine onboarding)$110,000 (edge I/O devices, enterprise setup & app rollout)$135,000 ($45k/plant sensor kits & coaching rollout x 3)$168,000 (9 DIN-rail IPCs [$18k] + 2,000 hrs CEE @ $75/hr [$150k]; $348k US)
2. 36-mo per-machine / station / plant SaaS fees$1,036,800 (90 machines @ $320/mo assumed = $28,800/mo x 36)$648,000 (90 stations @ $200/mo = $18,000/mo x 36)$702,000 (3 plants @ $6,500/plant/mo assumed = $19,500/mo x 36)$0 (100% owned IP; $0 per-machine or per-station license across unlimited lines)
3. 36-mo historian cloud, ERP sync & OT/IT ops$90,000 ($2,500/mo ERP/CMMS webhook & tag ops x 36)$126,000 ($3,500/mo app governance & OPC UA sync x 36)$90,000 ($2,500/mo ERP work-order & BI export ops x 36)$142,200 ($950/mo TimescaleDB infra + 40 hrs/mo CEE ops @ $75/hr [$3,000/mo] = $3,950/mo x 36)
Total 36-Month All-In Multi-Plant OEE TCO$1,221,800$884,000$927,000$310,200 (100% of $150k software + $18k IPC hardware included)

Under those assumptions, over 36 months across 90 machines, owning a custom OPC UA, MTConnect, MQTT Sparkplug B, and TimescaleDB OEE platform built in CEE saves $573,800 against Tulip, $616,800 against QAD Redzone, and $911,600 against MachineMetrics, while letting you connect machine #91 through #300 for the cost of a DIN-rail gateway alone.

Here are Idealogic's 2026 owned OEE & shop-floor telemetry build tiers comparing senior CEE manufacturing software and custom software development squads ($75/hr) against domestic US agencies ($165/hr, 2.2x). For context, Eurostat puts the 2025 EU average hourly labour cost at EUR 34.9, with Poland at EUR 19.3 and Romania at EUR 13.6. That is an employer cost and not an agency billing rate, but it explains the gap:

2026 Owned OEE & Edge Telemetry TierIndustrial Protocol, Historian & ISO 22400-2 DeliverablesTimelineCEE Build (@ $75/hr)US Agency (@ $165/hr)
Tier 1: Single-Plant OPC UA / MTConnect OEE & Tablet Andon MVPDockerized IEC 62541 OPC UA & ANSI/MTC1.4 MTConnect edge collector (250 ms debounce), TimescaleDB continuous aggregates, ISO 22400-2 OEE engine with OOE and TEEP, retroactive downtime splitting & operator tablet/TV Andon UI (1,120 to 1,560 hrs)12 to 16 weeks$84,000 to $117,000$184,800 to $257,400
Tier 2: Multi-Plant Sparkplug B Unified Namespace & PackML CausalityISO/IEC 20237 MQTT Sparkplug B broker hierarchy with store-and-forward offline buffering, ISA-TR88.00.02 PackML Starved/Blocked bottleneck propagation, automated ERP/MES SKU C_ideal sync & shift Pareto scorecards (1,760 to 2,400 hrs)16 to 22 weeks$132,000 to $180,000$290,400 to $396,000
Tier 3: Brownfield PLC Suite, Digital SMED, Inline SPC & Closed-Loop CMMSLegacy Siemens S7, Allen-Bradley EtherNet/IP & Modbus TCP adapters, digital SMED changeover timers, inline vision/gauge C_pk SPC drift alerts, and automated CMMS work-order + APS capacity feedback (2,600 to 3,600 hrs)24 to 32 weeks$195,000 to $270,000$429,000 to $594,000
Scale OEE across 50+ machines without per-machine subscription fees
Share your PLC/CNC mix, plant count, and ERP/MES landscape. Our manufacturing architects will return a 36-month TCO comparison and OPC UA / Sparkplug B edge architecture.
Get your multi-plant OEE build plan

Driving continuous improvement from OEE telemetry

Once PLC-driven OEE tracking has run for four weeks, the loss Pareto tells process engineering where to act:

  • Digital SMED for setups: Compare timestamped changeover phases across shifts and display live internal vs external setup countdowns on overhead Andon boards. Crews see where the minutes go, and setup losses can drop substantially.
  • Micro-stop Pareto hunting: Rank minor stops by cumulative weekly minutes lost and not by single-event length. A feeder track that pauses for 18 seconds forty times a shift costs 12 minutes per shift, roughly 3 to 4 hours of bottleneck capacity a week, and a simple guide-rail adjustment often fixes it.
  • Inline SPC quality gates: Tie dimensional gauges and vision cameras to quality management software and PLM software tolerance tables so tool wear is caught via C_pk drift before scrap fills the bin.

Maybe you need OEE software across brownfield machinery. Maybe it's a complete smart factory architecture that combines custom software development and AI software development. Our manufacturing software development team builds both, and the OEE number comes straight from PLC data.

Want an OEE number the shop floor actually believes?
We connect legacy and modern PLCs over OPC UA, MTConnect, and Sparkplug B to build real-time OEE, downtime, and loss-attribution systems with no per-machine license fees.
Talk to our manufacturing engineering team

Occasional field notes on building software, no spam

Protected by Cloudflare Turnstile · Privacy · Terms

Artem Zaitsev
Artem ZaitsevCTO · Idealogic Development OÜ

CTO and Co-founder at Idealogic. Directing engineering, systems architecture, security, and full-stack delivery across web, mobile, and blockchain.

Frequently asked questions

  • Overall Equipment Effectiveness (OEE) is a manufacturing KPI defined in ISO 22400 Part 2. It measures the percentage of planned production time that turns into defect-free output at ideal speed. The OEE formula multiplies three ratios: Availability (Actual Run Time divided by Planned Production Time), Performance (Actual Output multiplied by Ideal Cycle Time, divided by Run Time), and Quality (Good Count divided by Total Produced Count).

  • They differ only in the time base used as the denominator. OEE divides valuable operating time by Planned Production Time, so unscheduled shifts and planned maintenance are left out. Overall Operations Effectiveness (OOE) divides by Total Scheduled Shift Time, which includes planned maintenance and changeovers inside the shift. Total Effective Equipment Performance (TEEP) divides by all calendar time (24 hours a day, 365 days a year) and shows unscheduled capacity. ISO 22400-2 defines OEE. OOE and TEEP are TPM conventions.

  • Formalized by Seiichi Nakajima at the Japan Institute of Plant Maintenance (JIPM), the Six Big Losses group equipment waste into three pairs matching the OEE factors. Unplanned breakdowns and setup or changeover adjustments reduce Availability. Idling and minor stops (micro-stops, counted here as stops under two minutes, though the threshold is a site convention) and reduced speed cycles reduce Performance. Startup and warmup rejects after recipe changes and steady-state in-process defects reduce Quality.

  • It depends on the process. The widely cited 85% world-class OEE benchmark (90% Availability x 95% Performance x 99% Quality, about 84.6%) comes from Seiichi Nakajima's TPM work in 1970s Japanese automotive plants with long runs. Top-quartile continuous process lines in chemicals, paper, or beverage bottling can reach 90% or more because changeovers are rare. oee.com reports that most manufacturers sit closer to 60%. A high-mix job shop doing several setups per shift can be well run at 42% to 55%, an indicative range and not a published statistic.

  • Because manual logs overstate OEE. TeepTrak's State of OEE 2026 benchmark puts the gap at roughly 8 to 15 points. Operators don't record 15-second sensor jams, memory smooths a rough shift into round 15-minute blocks at the end of the day, and supervisors reclassify slow changeovers as planned downtime. PLC telemetry records every state change and cycle timestamp, so the lower number is the real baseline.

  • ISA-TR88.00.02 (PackML) standardizes seventeen machine states inside the PLC, including Execute, Held, Suspended, and Aborted. The 2022 revision also defines Starved and Blocked PackTags. OEE software reads the active state, those two tags, and the first-out fault register over OPC UA. That's enough to tell an internal breakdown from upstream material starvation or a downstream conveyor blockage, so the operator is asked for a reason code only when human context is needed.

  • No, a legitimate OEE can't exceed 100%. Availability and Quality are bounded at 100% by definition, so any reading over 100% comes from the Performance factor exceeding 1.0. This happens when the Ideal Cycle Time stored in the ERP routing is padded or slower than the machine's true fastest sustainable speed, or when batch counter bounce records phantom part pulses. A Performance score above 100% indicates a master-data calibration error.

  • Standalone cloud OEE trackers work well for quick visibility across 5 to 15 simple machines that only need cycle counters and an operator tablet. Building OEE into a unified manufacturing data platform or custom MES wins when you need recipe-dependent ideal cycle times from the ERP, automated PackML state correlation across multi-machine lines, zero per-machine SaaS fees across dozens of assets, and closed-loop triggers to maintenance and scheduling.

  • MachineMetrics does plug-and-play CNC and discrete machine monitoring and prices by quote (we model $320 per machine per month as a budgetary assumption). Tulip lists its no-code frontline apps at $100 to $250 per interface per month, with Enterprise quote-based. QAD Redzone bundles connected-workforce coaching with OEE and is also quote-based. Vorne XL sells one-time LED hardware scoreboards without SaaS fees. An owned OPC UA, MTConnect, Sparkplug B, and TimescaleDB stack has no per-machine fee across multi-plant rollouts.

  • In our model of a 3-plant, 90-machine manufacturer over 36 months, Tulip totals roughly $884,000, QAD Redzone roughly $927,000, and MachineMetrics roughly $1,221,800 once edge kits, implementation, and recurring per-machine or plant subscriptions add up. The vendor rates are budgetary assumptions, since two of the three price by quote. Engineering an owned IEC 62541 OPC UA, MTConnect, MQTT Sparkplug B, and TimescaleDB OEE platform in CEE at $75/hr totals roughly $310,200 all-in including $150,000 of software CapEx and $18,000 of industrial edge gateways, saving $573,800 to $911,600.

Still unanswered
Ask us directly

A senior engineer replies under 4 hours.