Solutions

Predictive Maintenance

Many companies offer predictive maintenance solutions, but our approach is faster, more cost effective and requires fewer resources. We achieve this using templates that follow a six-step process.

• 1Data collection — We collect data from multiple sources for any given time period, yielding only the necessary information for analysis and real-time results.
• 2Signal processing — We remove outliers and noise to get clean data.
• 3Feature extraction — We convert raw, clean data into meaningful features to train our AI algorithms.
• 4Health assessment — Our AI model creates a fingerprint of healthy and unhealthy machines and compares performance over time.
• 5Trending/prediction — We identify trends and calculate remaining useful life (down to the day, cycles, or shifts).
• 6Fault diagnosis — When a failure is likely, we determine root cause so maintenance teams can respond before downtime.

Industrial Robots

Our industrial robot templates observe each axis to monitor servo-motor torque and provide fleet-based health assessments and failure prediction. Unlike other offerings, our solution:

• Does not require any equipment sensors to be installed on the robot
• Can monitor any robot regardless of manufacturer, age or controller
• Is quick to deploy and can be operating in a few weeks
• Reduces maintenance time and costs because technicians know the exact component that requires attention
• Reduces spare part inventories

Machine Tools

Our smart machine templates analyze all critical components of a CNC machine to assess health and predict failure. They work on any CNC machine and any controller and are quick to deploy. Templates include:

• Tool condition monitoring — Assess misalignment, chatter and vibration that could compromise part quality
• Spindle performance monitoring — Using vibration sensors, health assessment within weeks, not months
• Feed axis (ball-screw) monitoring — Analyze health without the need for sensors
• Coolant analysis — Oil analysis sensor to monitor pH and determine when coolant needs to be changed
• Alarm (fault) mining — Capture and analyze alarm faults to determine patterns and root cause

Stamping Presses

We monitor the health of stamping presses (hydraulic, servo, electric and giant) via two templates: Vibration-based fault protection—sensors inside the machine to analyze press and die vibration—and Acoustic-based analytics when sensors can't be installed inside, using an acoustic sensor near the machine to capture frequencies and measure press degradation.

Ancillary Equipment

We monitor HVAC compressors and chillers, pumps and motors. Compressors and chillers: we analyze current and voltage data from the controller (no additional sensors). Pumps: vibration, pressure and flow sensors to assess health and predict failures. Motors: current and vibration assessment for health and failure prediction.