MultiViz AI Vibration Analytics

Viking Analytics is a company founded in 2017 in Gothenburg, Sweden, by a group of engineers and researchers. It builds software that helps vibration analysts and maintenance teams find developing machine faults earlier, pairing AI with human expertise instead of a black-box model.

MultiViz 2.0 uses unsupervised machine learning to build an explainable model for each sensor point, then applies detectors for symptoms such as modulation, rising kurtosis and harmonic shifts. The platform ranks every sensor by priority, keeps full-spectrum waveforms for root cause analysis, and connects to more than 50 vibration sensor brands and gateways via MQTT. Users include manufacturing plants, service providers, OEMs and sensor makers.

• AI prioritization: ranks which machines need attention first and explains why.
• Deep signal insights: full-spectrum waveforms and frequency trends for root cause analysis.
• Adaptive anomaly detection: learns each machine and sensor without thresholds or a fault library.
• Analyst-first workflow: users tag patterns, confirm anomalies and capture expert insights.
• Sensor-agnostic integration: connects to 50+ wired or wireless sensors and gateways via MQTT.
• Enterprise connectivity: works with SCADA, CMMS and data historians through an API.

A paper or pulp mill can rank hundreds of pumps and fans by urgency across several sites. A maintenance service provider in Vietnam can overlay MultiViz on existing sensors to serve more clients with fewer analysts. A foundry or forging plant can detect bearing wear on rotating machines weeks before failure.

The vendor reports 95% fewer false alarms compared with threshold-based monitoring. Analysts are said to save 2-3 hours per day, and early warning time of 1-2 months is stated. No threshold setup or fault library is needed because models learn each machine's own behavior, and the platform is sensor-agnostic.

Plants are adding wireless vibration sensors faster than analysts can review the alarms, which creates demand for software that filters noise and ranks assets. Vietnamese and wider Southeast Asian plants in cement, paper, steel and food are expanding condition monitoring and face the same shortage of vibration experts.

Based in Gothenburg, Sweden, founded 2017. Raised a Series A of more than EUR 3 million in January 2023. Partnerships listed on its news page include Weir (2025-09), CTC (2025-09) and MachineSaver (2026-01). Södra signed an enterprise-wide framework agreement in 2026-06.