Skip to content

The maintenance system your technicians actually use — off WhatsApp in days.

AI maintenance execution for operational excellence: work orders, preventive maintenance, spare parts, and condition monitoring that turns alerts into work orders. Technicians always know what to do, with checklists and a set fix for every known problem.

Contact Us
app.opexmx.com/predictive/coolant-pump-2
OpexMX
AR
PredictiveCoolant Pump #2 · Line B
Example data · OpexFlow edgeISO 20816 · Zone C

Health score

61 ↓ from 92

Detected fault

Outer race defect

DE bearing · 94% confidence

Remaining life (illustrative)

Calculating…

Weibull model

Vibration velocity, RMS (mm/s)Now 4.2 mm/s · 30 d
ABCD
Predicted failure · 17 d
FFT spectrum, 0 to 400 HzBPFO 107 Hz · 2× · 3×
Next best actionby total expected cost
Analysing fault signature…
See the Full Story

Example condition alert

  1. Predict
  2. Plan
  3. Execute
  4. Verify
Example: bearing fault detected in the vibration spectrum

Built for the floor you actually run

Most CMMS assume a perfect process. Yours isn't. OpexMX is built around how work actually moves, skills, capacity, and the people behind both.

01

Work goes to the right person, not the same person

OpexMX sees every technician's skills, shift, and current load, so the job lands on the right hands, every time.

02

Your best technicians stop being the bottleneck

Spread the load before they burn out. Capture how they work so their knowledge stays when they move on.

03

One system fast enough to actually use

When the tool fits the job, your team stops working around it. Every job gets logged, every decision becomes data.

From WhatsApp chaosto running in days

No rip-and-replace. No multi-month rollout. Here's how your team goes from scattered chats to one system.

01
Step 1

Map your floor

We map how work actually moves: your technicians' skills, your machines, and the real flow of jobs. Not the process on paper.

Team capacity3 on shift
AR
Ahmad Rizkion-shift
70%
SN
Siti Nurhalizaavailable
30%
BS
Budi Santosooverloaded
92%
02
Step 2

Go live in days

Guided setup configures OpexMX for your operation. Your team starts on day one with training built in, not bolted on.

Go-live checklist3/4
Map assets
Import team
Assign shifts
First ticket livein progress
03
Step 3

Run the floor

Work lands on the right person automatically. See who's doing what, who's overloaded, and where it's piling up, in real time.

Auto-assignmentjust now
medium#1253

Replace Bearing, Coolant Pump #2

Coolant Pump #2

AR
Ahmad Rizkiauto-assigned
04
Step 4Optional

Connect your stack

Optional, never required. Pull readings straight from machines and PLCs, sync work orders with your ERP, auto-replenish spares. Less manual entry, fewer errors, faster closes.

Integrations2 live
Machine dataPLC + SCADA
live
ERPWork orders
syncing
Spare partsAuto-replenish
ready
auto-synced42 events/min

Talk to OpexMX in plain language

Connect Claude, Cursor, or any AI agent you already use. Ask it to create a ticket, check inventory, find the right technician, or assign a work order. OpexMX does the rest.

Chat onWhatsAppWhatsAppTelegramTelegram
Or your own AI agent
Claude CodeClaude CodeCursorCursorHermesHermesOpenClawOpenClaw

Connect any AI agent

Model Context Protocol

Add OpexMX to Claude, Cursor, or any agent you already use. Your instance serves it over HTTP — nothing to install — with ready-made tools that stay up to date and scoped to each person's permissions.

  • Ready-made tools: assets, tickets, work orders, inventory, parts
  • New features reach your AI automatically, no rebuild
  • Per-person keys: the AI can only do what that person may do
  • One endpoint, no binaries: connect with a URL and an API key
Setup Guide
OpexMX
online · via WhatsApp

What one demo attendee told us

Maintenance managers who walk through OpexMX see the difference in minutes. The quote below is anonymous feedback after a product demo, not a customer reference.

“
OpexMX has a good master data structure that forces good execution. The system grows with the team as our processes mature, the platform adapts instead of getting in the way.
Maintenance Manager
Manufacturing, Surabaya, Indonesia
Anonymous feedback after a product demo — not a customer quote

See it for yourself. Book a walkthrough and get your own first impression.

What's Coming Next

v4.0 is here with the Digital Twin foundation, energy management, ERP integration, and AI Reliability Engineering. AI Predictive Maintenance is planned next — here's what's coming.

Recently Shipped

Latest major releases now available

v3.6

ERP & External System Integration

v4.0

Energy Management & Sustainability Tracking

v4.0

Digital Twin Foundation & Floor Plans

v4.1

AI Reliability Engineering

Shipped earlier

Offline-First Native Mobile AppMulti-Site & Enterprise ScaleAI Copilot for Maintenance OperationsPredictive Maintenance & IoT IntegrationCustom Dashboard Builder & Scheduled ReportsInventory Optimization & Smart ProcurementReliability Engineering Suite (FMEA, RCM)

Coming Soon

What we're building next

AI Predictive Maintenance

v5.0
Planned

Planned: multivariate AI models trained on your sensor data and failure history to forecast remaining useful life, to be validated against your own historical failures before you rely on them.

SOC 2 & ISO Compliance Framework

v6.0
Planned

Built-in audit trails, compliance checklists, and automated reporting for SOC 2 Type II, ISO 55001, and ISO 14001 certification readiness.

Have a feature request?

We build what our users need. Tell us what would make your team more effective.

Request a Feature

Condition Monitoring to AI Predictive Maintenance

From connecting your machines to alert-to-work-order today, and AI forecasting and prescriptive actions as planned next steps. The illustrations below are examples, not live data.

01
Machine connectivity3 live
Modbus TCP
OPC-UA
MQTT
edge → cloud42 events/min
Layer 1Industrial Edge

Connect

OpexFlow reads any machine over Modbus, OPC-UA, MQTT, and more — no rip-and-replace of existing PLCs, sensors, or historians.

02
Fleet health2 alerts
PUMP-P10192%
MOTOR-M20464%
CONV-C30188%
PUMP-P101—Vibration rising
MOTOR-M204—Temp 86°C
Layer 2Condition Monitoring

Monitor

FFT bearing-fault detection, ISO 20816 vibration alerts, parameter anomalies, and health scores — with regime and PCA models flagging drift before it becomes an alarm. Work orders auto-create when thresholds cross.

03
Failure forecast · illustrationExample: 17 d · CI 11–26 d
Illustration only · planned capability, example data
Layer 3Planned · AI Forecasting

Predict

Planned: Weibull models forecast remaining useful life with a confidence interval, trained and backtested against your own sensor history and failures before you rely on them.

04
Recommendation
Bearing 6205-2RSExample · 17 days
Action:Replace at Sep 5 shutdown
Part:Vendor B — cheapest, arrives Sep 3 · before windowfallback: Expedite by Sep 1 if the schedule slips
Downtime:2h
Risk:78% → 5%
Ranked by total expected cost — action + residual failure risk · stock-first · expedite fallback
Layer 4Planned · Prescriptive Action

Prescribe

Actions ranked by total expected cost — the intervention plus the residual failure risk it leaves behind — down to the spare part: stock first, else the cheapest vendor that delivers before the window, expedite as fallback.

Frequently Asked Questions

Everything you need to know about OpexMX

Get your techniciansoff WhatsApp in days.

The maintenance system your technicians actually use. Request trial access and our team will approve it within 4 business hours, or talk to us and we will show how OpexMX fits your specific setup.

Contact Us
Trial by request
No credit card required
Approved within 4 business hours