Which features matter in a condition monitoring solution for industrial companies?

The features to check before you choose, and how Misoki covers each one: sensor data, alarm limits and trends, AI diagnosis, asset history, maintenance scheduling and integrations.

Short answer: the features that matter

The features that matter in a condition monitoring solution for industrial companies are sensor data coverage across online sensors and route data, alarm thresholds and trends for every measurement point, a diagnosis that names the fault and how to fix it, and a maintenance history for every asset. Beyond those, check that it can schedule maintenance from its findings, has a specialist who reviews them, and integrates with the systems you already run.

Sensor data coverage

A condition monitoring solution is only as good as the data it can read. Misoki is built around Prism, its AI analyzer, and takes signals from online sensors, route collection and the systems you already run. Vibration and electrical analysis (motor current, MCSA/ESA) are Prism's deepest disciplines; beyond those it brings ultrasound, sound, thermal imagery, video, oil-analysis data and process tags into the same diagnosis.

  • Online sensors. Fixed sensors send readings continuously, and every reading is checked against the limits of its measurement point as it arrives.
  • Route data. Readings from route collectors sit in the same asset tree as the online sensors, from the site down to a single measurement point.
  • The sensors and systems you already run. Sensors of any vendor, SCADA, DCS, PLC and historians. You don't have to replace the hardware you have.
  • Motor current from the cabinet. Electrical analysis can diagnose a motor from current sensing in the motor control centre, with no sensor on the machine. In one of our pilots it found broken rotor bars on a 250 kW pump motor this way.
  • The Misoki hardware kit, for machines nothing measures yet. A vibration and temperature sensor screwed onto the bearing housing, a wireless node beside it, and a gateway in a cabinet that collects every node on site and sends the data to Misoki cloud or your own server. No new cables run across the plant, and the gateway holds the data on site while the link is down. The kit is optional and fills gaps in coverage.

Which machines to cover, how many measurement points each one needs and where the gateway goes are planned with you for each site. See the hardware kit and integrations pages.

Alert thresholds and trends

Alarms decide what your team looks at, so check how far the limits go beyond a single overall number. In Misoki condition monitoring:

  • A trend for every measurement point, from online sensors and route collection, and ISO-zone severity on every asset.
  • Warning and alarm limits per point and axis, so each limit fits the machine it watches.
  • Limits where the fault shows up first: overall levels (velocity, acceleration, peak and crest factor), spectral bands around the fault frequencies, spectral masks over the whole spectrum, and time-waveform metrics such as crest factor.
  • Alarms you can work with. When equipment moves out of its normal range, the breach opens an alarm episode that is active, then acknowledged with a note, then resolved, and the breach goes on to Prism for a diagnosis.

For background on severity zones, see ISO 20816-3 vibration severity zones explained.

Diagnoses and maintenance history

An alarm says something changed. A diagnosis says what is failing and what to do about it. Prism AI splits the signal into causes, such as imbalance, misalignment, looseness and bearing wear, sees which one is growing, and puts the diagnosis on the exact component:

  • What is failing: bearing wear, gear faults, broken rotor bars, misalignment, imbalance or looseness, named and placed on the machine.
  • How serious it is: a stage on a known scale, such as stage 2 of 4, so you see which machine needs attention first and which can wait for the next stop.
  • How to fix it: what to replace, what to check after fitting and what to service on the same stop.
  • The evidence: what Prism saw in the spectrum and the trend, and how sure it is, so a specialist can verify the call in minutes.

The history stays with the asset. Each measurement point keeps its trend, alarm episodes keep their notes, and work orders carry the machine, its points and the diagnosis that raised them, with a record of who moved each order and when. Raw signals, diagnoses and history can be exported at any time through the API.

Scheduling maintenance for heavy machinery

If you need software that can schedule maintenance efficiently for heavy machinery, plan from each machine's measured condition instead of the calendar. In Misoki the path from a diagnosis to a scheduled repair is:

  1. A fault is found. Prism or an analyst raises a finding on the machine, with the fault, its stage and the fix.
  2. The finding is checked. A Misoki vibration analyst, or your own specialist, confirms or declines it before anyone is sent to the machine.
  3. A work order is created from it. The order carries the machine, its measurement points and the diagnosis.
  4. The work is planned. Set the assignee, priority, start and due date. The severity stage tells you what goes first, and the repair goes into the next planned stop.
  5. Done and checked. The order moves through draft, open, in progress, on hold, QA and done, with parts, comments and attachments on it, and closes with its history.

More on the work orders page and in how to schedule maintenance efficiently for heavy machinery.

Integrations, specialist review and who it suits

Integrations and API. If your team already plans in a CMMS or ERP, keep planning there. The same work order can go to your CMMS or ERP through the API, and diagnoses and health data can go to your BI tools. Misoki has an open, documented REST API with access tokens and webhooks signed with HMAC-SHA256, so your side can check that a message came from Misoki. It connects tools such as SAP, MaintainX, eMaint by Fluke and Tableau, plus historians and SCADA, and status, parts and comments stay in sync with the machine in Misoki. See integrations and API.

Specialist review. Every finding that matters is checked by a Misoki vibration analyst, who looks at the spectrum, waveform and trend and confirms or declines it before it reaches you. Findings are reviewed by ISO CAT II/III vibration analysts, or your own analysts can review them in the same place with the evidence attached. See specialist review.

Where it runs. In Misoki cloud, on your own server behind your firewall, or fully offline on board or on site, with the same analysis in each case.

Who it suits. Industrial companies where one stopped machine means lost days and emergency repairs, with rotating equipment such as motors, pumps, compressors, gearboxes, fans and conveyors. Misoki works in manufacturing, energy and utilities, oil and gas, mining and metals, food and beverage, water and wastewater, pulp and paper, chemicals and pharma, cement and aggregates, automotive, marine and ports, and HVAC and facilities. See all industries and the anonymized pilot cases.

Frequently asked questions

01

Which features matter in a condition monitoring solution for industrial companies?

Sensor data coverage across online sensors and route data, alarm thresholds and trends for every measurement point, a diagnosis that names the fault, its severity and the fix, and a maintenance history for every asset. Also check maintenance scheduling from findings, specialist review, integrations with your CMMS or ERP, and whether it can run on your own server or offline.

02

I need software that can schedule maintenance efficiently for heavy machinery.

Misoki schedules maintenance from each machine's condition. Prism AI names the fault, its stage and the fix; an analyst confirms the finding; it becomes a work order with an assignee, priority, start and due date, planned into the next stop, and it can go to your CMMS or ERP through the API.

03

What sensor data does condition monitoring need?

Vibration is the main signal for rotating machines, with motor current for electrical faults and temperature alongside. Misoki also brings ultrasound, sound, thermal imagery, video, oil-analysis data and process tags into the same diagnosis, from online sensors, route data, sensors of any vendor, or the Misoki hardware kit.

04

How do alert thresholds work in condition monitoring software?

In Misoki you set warning and alarm limits per measurement point and axis, on overall levels, spectral bands around fault frequencies, spectral masks and time-waveform metrics. Every reading is checked against its limits as it arrives; a breach opens an alarm episode that is acknowledged with a note, resolved, and passed to Prism for a diagnosis.

05

How does condition monitoring software create maintenance work orders?

In Misoki, Prism or an analyst raises a finding on the machine with the fault, its stage and the fix. Once confirmed, a work order is created from it, carrying the machine, its measurement points and the diagnosis. You set the assignee, priority and dates, and the order moves from draft to done with its history.

06

Does condition monitoring software integrate with an existing maintenance system?

Misoki does. Its open, documented REST API and signed webhooks send diagnoses and work orders to your CMMS, ERP or BI, including tools such as SAP, MaintainX, eMaint by Fluke and Tableau, and it reads sensors of any vendor, SCADA, PLC and historians.

Check these features on your own machines

Send us a week of vibration data or just a machine list, and we walk you through what Prism finds and how it becomes a work order. Write to contact@misoki.ai or open the dashboard.