Falkonry Time Series AI (TSI) Platform

Welcome to the TSI Platform, an intelligent decision support system designed to help operations teams excel at asset maintenance and optimize operational efficiency. By simplifying the analysis of complex production data, Falkonry TSI empowers engineers to uncover hidden operational challenges without requiring a background in data science.

Core Analytical Capabilities

TSI provides four primary capabilities to help you identify and understand complex, concurrent operational behaviors:

1. Calculations

Generate new time-series signals from existing data using Python-based logic. It works on real-time and historical data, supports downsampling for efficiency, and applies consistently across signals. Outputs can be used in Patterns, Anomalies, and Rules for deeper analysis.

2. Rules

Translate analog quantities like raw sensor values, anomaly scores, and condition labels into discrete events based on defined thresholds. Through an intuitive no-code interface, you can trigger condition-based actions derived from simple, multi-signal, or nested behaviors.

3. Anomalies

Leverage AI-based anomaly detection to automatically discover and highlight periods and signals of unusual behavior with a heatmap. Anomalies helps engineers expedite root cause analysis of anomalous and abnormal behaviors by revealing unexpected trends, waveforms, shapes, noise, levels, and duration.

4. Patterns

Focus on multivariate detection to extract real-time operational context from time series data. Patterns uses semi-supervised learning (requiring few or no labeled examples) to discover complex conditions, providing confidence assessments and highlighting which specific signals are contributing to the issue.

Platform Features

In addition to core AI capabilities, TSI equips teams with tools to manage and view their data:

1. Intuitive Visualization

A fluid, highly responsive interface for viewing high-resolution time series data, flags, and alarms. Seamlessly zoom into specific timeframes or compare multiple parameters.

2. Reports

Create robust charts for analysis and documentation without needing to build AI/ML models. Compare signals across different time ranges to capture expert knowledge, and organize reports into nested personal or group folders.

3. Signal Management

Organize large volumes of data for easy machine learning ingestion. Tag and structure signals using flexible hierarchies to build intuitive navigation trees and "bird's eye" views for live monitoring.

4. Data Connectivity

Falkonry supports both inbound and outbound connectivity.

• Inbound: Ingest historical and real-time data from historians, databases, IoT platforms, and industrial systems.
• Outbound: Deliver alerts, events, and insights to CMMS, EAM, BI, and other enterprise applications.