Condition monitoring software that shows only raw vibration amplitude without computing fault frequencies, trending against a baseline, or connecting alarms to work orders creates data that requires expert analysis to act on. The platform's job is to make the decision obvious, not to provide more raw data.
What Condition Monitoring Software Does
Condition monitoring software sits between sensor hardware and maintenance action. Its job is to transform raw sensor data into a decision: which machines need attention, what kind of attention, and how urgently.
The functional chain runs through four layers:
- Sensor data ingestion: Receives raw time-domain or frequency-domain data from connected sensors. Ingestion may be continuous (online monitoring) or periodic (route-based or timed acquisition).
- Spectrum analysis: Converts time-domain vibration signals to the frequency domain using Fast Fourier Transform (FFT), producing a spectrum that shows vibration amplitude at each frequency. The platform applies this analysis automatically, not on demand.
- Fault detection algorithms: Identifies signatures associated with specific failure modes. Bearing fault frequencies (BPFO, BPFI, BSF, FTF) are calculated from bearing geometry. Imbalance appears at 1x running speed. Misalignment appears at 1x and 2x. Resonance appears as broadband amplitude increase near a natural frequency. The platform's ability to automate this detection determines how much analyst time it requires.
- Alarm management and reporting: Determines when a condition warrants action and routes that information to the people who can act on it. This layer connects monitoring to maintenance workflow.
Evaluation Criteria
Bearing Fault Frequency Computation
This is the most consequential capability to evaluate. Bearing fault frequencies are mathematically deterministic: given the bearing designation and shaft speed, BPFO, BPFI, BSF, and FTF are calculable. A platform that maintains a bearing database and computes these frequencies automatically when the bearing designation is entered removes the single largest bottleneck in scaling vibration analysis.
Platforms that require a vibration analyst to manually identify and enter fault frequency bands for each bearing on each machine do not scale past a small number of critical assets. Ask vendors specifically: does the platform look up bearing designations and configure fault frequency tracking automatically?
Alarming on Change from Baseline vs. Fixed Threshold
Fixed-threshold alarming uses a single RMS amplitude value as the alarm point for all machines of a given type. This produces false positives on machines that run with inherently high vibration and misses early faults on machines that normally run clean.
Baseline-relative alarming establishes a machine-specific normal during an initial operating period, then alarms when readings deviate significantly from that normal. A machine running at 0.3 g RMS normally is worth attention when it reaches 0.6 g. A machine normally running at 1.2 g is not alarming at 1.3 g. Ask which approach the platform uses and whether baselines are updated as machines age.
Multi-Sensor Type Support
Rotating equipment failures often involve multiple physical signals. A bearing defect produces vibration. It also produces heat through increased friction. A motor with developing insulation issues shows changes in current draw before mechanical vibration is affected. Platforms that handle vibration only require separate systems for temperature and current, which fragments the data and makes correlation harder.
Ask whether the platform ingests temperature, motor current (MCSA), ultrasound, and oil analysis data alongside vibration, and whether these signals appear together in the asset view rather than in separate dashboards.
CMMS Integration Depth
The gap between a monitoring alarm and a completed repair is a workflow problem, not a technical problem. Platforms that generate email alerts leave the maintenance team to translate the alert into a work order manually, which introduces delay and loses context.
Evaluate integration depth by asking: does the platform create work orders in your CMMS directly from alarm events? Does the work order include the asset ID, fault description, and severity level? Is this integration native (direct API) or middleware-dependent?
Deployment Model
Cloud-hosted platforms minimize IT infrastructure requirements and support continuous updates. On-premise deployment keeps data within the plant network and satisfies certain security or regulatory requirements. Hybrid models keep raw data on-site while sending processed results to cloud dashboards. Confirm which models are available and what the IT requirements are for each.
Key Questions to Ask Before Selecting a Platform
- Does the platform maintain a bearing database and compute fault frequencies automatically from the bearing designation?
- Does it alarm on deviation from a machine-specific baseline, or against fixed thresholds?
- Which sensor types does it ingest natively: vibration, temperature, current, ultrasound?
- Which CMMS platforms does it integrate with, and what does a work order created from an alarm look like?
- Is data cloud-hosted, on-premise, or hybrid? What are the IT requirements?
- How does the platform handle new assets? What is the configuration time per machine?
- What does the alert-to-repair workflow look like from alarm to closed work order?
What Tractian Offers
Sensor Data Ingestion
Tractian's platform ingests data from its own wireless triaxial sensors, which measure vibration in all three axes simultaneously. Sensors transmit continuously to a gateway device that forwards data to the cloud platform. Measurement intervals are configurable per asset.
Automatic Fault Detection
The platform maintains a bearing database and computes bearing fault frequencies automatically when the bearing designation is entered for an asset. BPFO, BPFI, BSF, and FTF are tracked as independent channels in the frequency spectrum, with amplitude trended over time per fault type. This means a developing outer race defect shows a rising trend in BPFO amplitude weeks before the overall RMS level rises to a generic alarm threshold.
The platform also identifies imbalance (1x), misalignment (1x and 2x), and looseness (sub-synchronous and integer harmonics) based on spectral pattern recognition applied automatically.
Multi-Signal Support
Tractian's sensors capture vibration and temperature at the same point simultaneously. The platform displays both signals in a unified asset view, so rising bearing temperature that coincides with rising fault frequency amplitude is visible in a single screen rather than requiring correlation across separate tools.
Alarm Workflow and CMMS Integration
When the platform identifies a fault condition above a configured severity threshold, it generates an alert with the asset name, fault type (e.g., outer race bearing defect), severity level, and trending data showing how long the condition has been developing. The platform integrates with major CMMS platforms to create work orders directly from these alerts, including the fault context so the maintenance team has the diagnostic information before they open the machine.
Other Platforms
[PLACEHOLDER: Research platforms before publishing. Do not invent names, features, pricing, or reviews. Source from vendor sites and verified third-party reviews (G2, Capterra, peer publications). Include at least three competing platforms with: sensor compatibility, analytics capabilities (bearing database yes/no, baseline alarming yes/no, multi-signal yes/no), CMMS integration specifics, pricing model if public, and deployment model options. Follow BOFU competitor content rules: Tractian section only in full; all others as clearly labeled placeholders.]
See Tractian's Condition Monitoring Platform
Automatic bearing fault detection, baseline-relative alarming, and direct CMMS integration. Designed to scale across hundreds of assets without requiring a full-time vibration analyst.
See Mechanical Monitoring