Processed Input Data – PipeFusion

Input Data, Data Quality, Analysis and Machine Learning

InfoTiles processes data through three distinct pathways. Understanding which pathway a dataset uses helps you interpret its contents and update frequency correctly. See 1.2 How InfoTiles Handles Data for an overview of all three types.

Processed Input Data – PipeFusion

All datasets starting with pf_ are loaded via a machine learning tool called PipeFusion. PipeFusion uses data from the Asset Database as its source and will by default apply a filter for status: Active. Pipes from the Asset Database with other status values — such as removed, replaced, or projected — will not be visible in the PipeFusion results.

As part of quality assurance, missing nodes, missing connections, and incorrect pointers from the asset database will be corrected. The original value from LSID/PSID is preserved in the field pf_id_source, and changes are documented under pf_history. The total number of nodes and length of pipes may be updated as a result of the correction.

For analysis of raw data, dedicated data views exist for the respective tables used, with the prefix Asset Database.

PipeFusion Analyses

PipeFusion performs several analyses on datasets from the pipe database and work order systems, as well as any other relevant data sources such as the subscriber database, sensor data, and measurements (e.g. SCADA data, weather data, watercourse data). Examples of analyses and data generation include:

• Network connections and zone divisions
• Probability of failure/break
• Consequence of failure/break
• Risk of failure/break based on probability and consequence
• Inflow and infiltration calculations
• Water consumption for calculation and alerting of water leakage

Related Articles

• How InfoTiles Handles Data
• Datasets Uploaded Directly by User – Client
• Data Directly from Source
• Network Connections and Zone Divisions
• Risk Calculations
• Inflow and Infiltration Analysis