Skip to content

Patterns

Falkonry Patterns is a core capability within Falkonry's Time Series Intelligence (TSI) platform, designed to provide deep insights into complex system behaviors using time series data. At its heart, Patterns leverages Falkonry’s proprietary and patented PatternIQ™ technology, purpose-built for industrial time series visualization, anomaly detection, classification, and collaboration.

Purpose and Capabilities

The primary goal of Falkonry Patterns is to discover and classify meaningful multivariate, temporal correlations in streaming data. This includes identifying:

  • Early warning patterns
  • Stages of deterioration in complex systems
  • Recurring behaviors — whether previously known or entirely unknown

Patterns helps organizations unlock the full potential of their time series data, transforming raw sensor streams into timely, actionable intelligence. Aimed at reliability engineers and data analysts, Patterns supports:

  • Continuous monitoring of specific events and known conditions
  • Root cause analysis
  • Unsupervised condition modeling

Key Advantages

Falkonry Patterns offers several benefits over traditional analytics approaches:

  • No Code / ML Expertise Required: Designed for engineers and domain experts with no programming experience.
  • Robust Data Handling: Automatically tolerates missing values, outliers, irregular sampling rates, and signal gaps — no need for heavy preprocessing.
  • Rapid Learning: Efficiently trains on large operational datasets.
  • Actionable Insights: Explanation scores help pinpoint signal-level causes of detected patterns, speeding up diagnosis.
  • Iterative Refinement: Models can be continuously refined by adjusting training periods, parameters, or inputs based on evaluation.

Integration and Ecosystem

Falkonry Patterns outputs — such as predictions, confidence scores, and explanation scores — are new signals that can integrate seamlessly with other Falkonry TSI features: Rules

  • Use Patterns outputs as inputs to Falkonry Rules for condition-based alerting and low-latency notifications when patterns are detected.
  • While Patterns focuses on known or recurring behavior classification, Falkonry Insights specializes in detecting unexpected anomalies — making them complementary tools.
  • Patterns outputs can be grouped into Signal Trees, helping organize signals hierarchically by asset or function, improving discoverability and context within the platform.