Parvex turns motor current signatures and vibration telemetry into prescriptive maintenance intelligence — catching failures weeks before they happen, automating compliance, and routing your technicians like a chess master.
The problem
Over a million elevators in the US require mandatory annual safety inspections — yet maintenance decisions are still made on fixed schedules and gut feel. Unplanned outages cost building owners $500–$1,200 per incident. Technicians spend 40% of their time on unnecessary visits. Compliance certificates are managed in spreadsheets.
Parvex installs a compact IoT sensor node on the motor control cabinet — no wiring, no downtime — and begins streaming current signatures and three-axis vibration data to our cloud models within the hour.
The platform
Built for elevator service companies, facility management firms, and independent maintenance operators who want to move from reactive firefighting to proactive precision.
Our models analyse motor current signatures using Fast Fourier Transform decomposition and vibration spectral fingerprinting to detect bearing wear, winding faults, and brake degradation weeks before physical failure. Trained on motor telemetry patterns across hundreds of drive types.
State-by-state compliance logic built in. Parvex tracks inspection due dates, auto-generates certificate packages pre-populated with sensor data, and files directly with jurisdictional authorities where APIs are available — turning a 3-hour admin task into a one-click confirmation.
Constraint-based route optimisation across your technician fleet, factoring failure urgency scores, technician skill profiles, real-time traffic, and parts availability. Average travel time reduction of 30–40% in pilot deployments, with dynamic re-routing on new alerts.
Predictive demand signals for your parts storeroom. Parvex correlates failure forecasts across your entire monitored fleet to predict which SKUs — motors, contactor kits, brake shoes, door operators — need replenishment 4–6 weeks out, cutting emergency procurement costs.
How it works
A Parvex IoT node clamps onto the motor control cabinet power rail — no rewiring, no downtime, no certified electrician required. The device streams current waveforms at 1kHz and three-axis vibration at 800Hz over LTE or building WiFi. Full deployment per unit: under 45 minutes.
On-device edge processing computes frequency domain features locally, minimising data transmission cost. Cloud models run ensemble anomaly detection and component-specific degradation models — individualised per elevator's drive type, age, load profile, and usage pattern. Models update continuously.
Alerts surface in the Parvex dashboard ranked by failure probability and business impact. Each alert includes a plain-language diagnosis, recommended action, estimated parts required, and a one-click dispatch to your nearest available technician — with route already optimised.
Compliance
Every US state has its own elevator inspection requirements — frequency, documentation format, certifying authority. Managing a portfolio of 50+ units across multiple jurisdictions means a full-time compliance coordinator.
Parvex maps all 50 state code requirements and maintains a live compliance calendar for your portfolio. When an inspection is due, we auto-generate the certificate package — pre-filled with sensor-verified operational data — and route it to the right authority.
Founder
The most expensive failure is always the one nobody saw coming. Every elevator announces its failure in advance — the industry just hasn't been listening.
Nihar Gandhi founded Parvex to solve a problem he observed firsthand: the massive gap between how elevators are maintained today and how they could be maintained with modern sensor hardware and machine learning. The vertical transport industry has been locked in a time-warp of manual inspections, paper-based compliance, and reactive service dispatches — not from lack of will, but from lack of an accessible intelligence layer.
Parvex's architecture reflects Nihar's conviction that the right answer is non-invasive and additive — a sensor node that snaps onto existing equipment, a software platform that integrates with existing workflows, and a compliance engine that does the paperwork no one wants to do. No expensive elevator replacements required. No downtime for installation. No rip-and-replace.
The founding insight: motor current signatures carry an enormous amount of information about mechanical health — bearing condition, winding insulation quality, brake engagement friction — that trained signal processing models can extract with high reliability. Parvex's models were developed and validated against real-world failure datasets from commercial building portfolios, not just laboratory test benches.
Why now
Regulatory pressure is intensifying. Building owners are getting squeezed by liability exposure. And for the first time, the sensor and edge-compute cost curve makes per-unit monitoring economically viable at scale.
Get started
We're working with a select group of elevator service companies and facility managers for our initial deployments. If you manage 20+ units, we'd like to talk.