OEE is the most gamed metric in manufacturing.
We build the dashboard that surfaces the gaming.
Real-time OEE per asset, refreshed every 30–60 seconds via OPC-UA into Microsoft Fabric Real-Time Analytics. Six Big Losses Pareto, audit trail, mobile shift-handover view. Built on Power BI Direct Lake. First production OEE in 6 weeks.
The Problem
Patterns we see in every engagement
Most boards see 85% OEE. The plant is producing 62%. The line operator pads availability by reclassifying changeover as planned. The quality reject goes into rework instead of scrap. These are the four patterns we surface.
Availability gamed by reclassifying stops as planned.
Setup, cleaning, minor stoppages and changeover get moved into the 'planned production time' denominator. Availability looks fine. Real availability is 12–18 points lower. The dashboard logs every reclassification with timestamp, operator and reason — gaming becomes visible.
Performance gamed by lowering the ideal rate.
If the ideal cycle time gets edited down to match what the line actually does, performance always looks like 95%. The audit trail catches every edit to the standard. Anyone changing the rate has to log why.
Quality gamed by reclassifying scrap as rework.
If quality holds and rework do not count against quality rate, the number stays high. The dashboard tracks rework-as-percentage-of-scrap as a separate trend so quality gaming surfaces over weeks, not hides indefinitely.
Manual operator entry burden kills adoption.
If the operator has to log every stop on a clipboard, they will not, and the data will not be there. We design with the manual entry burden at 30 seconds per shift maximum — automated capture from PLC tags wherever the signal exists.
What we build
What we build
Six dashboards. Each tile maps to a specific decision a plant manager, shift supervisor or operations director makes that day.
Real-time OEE tile per asset
Replaces
The shift-end Excel OEE that arrives Monday morning, three days after the decisions that mattered.
- Refresh every 30–60 seconds via OPC-UA or MQTT from PLC tag into Fabric Real-Time Analytics
- Availability × Performance × Quality computed continuously — no shift-end Excel reconciliation
- Drill-through to current run rate, downtime cause and operator on shift
- Mobile-friendly tile so plant manager sees current OEE from anywhere in the plant
OEE visibility shifts from end-of-shift Excel to live dashboard. Decisions move during the shift, not at next morning's standup.
Six Big Losses breakdown per shift, per line
Replaces
The 'we know there's downtime but we don't know why' conversation in the operations review.
- Breakdown, setup, minor stops, reduced speed, defects, startup losses — categorised per shift per line
- Operator stop-reason entry via Power Apps tablet form, sub-30-second log time
- Pareto chart of which causes drive 80% of the loss — drill to which crew, which time of day, which material
- Trend over 30 / 90 / 365 days so seasonal patterns surface
Maintenance and operations stop guessing. The Pareto names the cause. The next intervention has a target.
Audit trail of every reclassification
Replaces
The shift supervisor who quietly reclassifies a 90-minute changeover stop as 'planned downtime' so OEE looks acceptable.
- Every stop reclassification logged with timestamp, user, original code, new code, reason
- Daily exception report to plant manager surfaces unusual reclassification patterns
- Quality hold-to-rework conversions tracked separately so quality gaming is visible over weeks
- Read-only audit log — even admins cannot back-date or rewrite the original capture
Gaming does not stop because operators are watched. It stops because gaming is visible to the person who reviews the number.
Target vs actual per line, per crew
Replaces
The 'we hit target this week' narrative that compares actual to a target that was already lowered last month.
- Per-line OEE targets locked at quarter-start with formal change control
- Crew-level trending — same line, same shift pattern, three crews, real comparison
- Bench-mark vs the best-shift-ever value as a stretch reference
- 30 / 90 / 365 day trend so quarter-end target-game-the-system patterns surface
Operations leadership stops congratulating the line for hitting a soft target. The conversation shifts to closing the gap to the best crew.
Mobile shift-handover view
Replaces
The clipboard handover where the outgoing supervisor briefs the incoming one in 90 seconds and 60% of context gets lost.
- Plant manager sees yesterday's OEE on their phone before the morning huddle
- Shift handover form on Power Apps tablet — open issues, quality holds, maintenance backlog
- Auto-summary of the previous 12 hours with named exceptions and current line state
- Read directly from the same Fabric Lakehouse — no duplicated data, no sync lag
Shift handover quality improves. Open issues do not get dropped between crews.
Stack — Microsoft Fabric end to end
Replaces
The mix of MES OEE module, Excel pivot, in-house SQL warehouse and Power BI that nobody can reconcile.
- Fabric Real-Time Analytics (KQL database) for stream
- OneLake for historian / time-series — 5+ years of granular history
- Power BI Direct Lake for the semantic model and dashboards
- Azure IoT Hub if you need cloud-side device management at scale
One stack. One refresh strategy. One semantic model. Maintained by a small team — not a five-vendor integration project.
How we work
From plant walk to live OEE in 6 weeks
We do not start with the dashboard. We start with the PLC tags, the operators, and the stop-reason taxonomy nobody has agreed on yet.
01
Discover — walk the lines, audit the tags
Two weeks. We walk every line in scope, document which PLC tags exist for stop reason, run rate and counts. We facilitate the per-line target-setting workshop with operations. We score which assets have enough signal for automated OEE — and we are honest about which do not.
02
Prototype — one critical line live
Two weeks. Build the OEE dashboard for one critical line. Refresh every 30 seconds. Run it for 2 weeks in parallel with whatever exists today. Operators see both numbers. Trust gets built before scale.
03
Deploy — across all lines, then predictive
Four to six weeks. Roll out to remaining lines. Train operators on the audit trail. Train shift supervisors on the morning huddle workflow. Add predictive maintenance, energy intensity, scrap analytics — same Lakehouse, same Power BI workspace.
Technology stack
Edge & OT
Cloud Streaming
Lakehouse
Visualisation
Operator UX
Source Systems
Common questions
What buyers ask us
We already have an OEE module in our MES. Why a separate dashboard?
Most MES OEE modules do not expose the data outside MES. The plant manager cannot see it from the office. The Group Operations Director cannot compare across plants. The cross-line comparison does not exist. We pull the same data MES is calculating and put it in Power BI alongside cost, quality and supply chain data. Same OEE number — more useful.
Our PLCs are 20 years old. Can we still do this?
Often, yes. We have connected to PLCs that predate Ethernet via a Kepware or Ignition edge gateway. The honest test: if the PLC has tags exposed via OPC-UA, Modbus or even MQTT, we can read them. If it is a true island with no comms layer, we add an edge module first.
Can the dashboard read from Siemens MindSphere / Rockwell FactoryTalk / AVEVA?
Yes. We connect to MindSphere, FactoryTalk and AVEVA via their APIs, land the data in OneLake, and the dashboard sits on top. Vendor lock-in works the other way after we ship — your OEE history is now portable.
How much does it cost?
Discover phase USD 8,000–12,000 fixed-fee. Prototype (one line) USD 18,000–28,000. Deploy across 5–15 lines USD 40,000–120,000 depending on line count, PLC count, edge integration scope. Quoted precisely after Discover.
What if our operators do not want to log stop reasons?
We design the experience so logging takes under 30 seconds per shift. The bigger blocker is usually trust — if the data has been used to punish operators historically, getting buy-in takes longer. We facilitate the operator workshop and we design the audit trail to surface gaming by supervisors, not operators. The dashboard becomes a tool for the floor, not just leadership.
Ready to move
Book a 30-minute OEE diagnostic
30 minutes with Amit. No slides. No pitch deck. No obligation to proceed. We walk through your current OEE definition, your PLC signal coverage, and where the dashboard would land first.