Case Study

Top 20 US gas utility resists encroachments with AiDash


AI analysis of satellite images identifies precise location of encroachment issues

Found 15 previously unnoticed encroaching structures
Right-of-way violation identified in first 50 miles analyzed
570 pipeline encroachments detected and prioritized

Company: Top 20 U.S. Gas Utility
Location: United States of America
Solution: AiDash Intelligent Encroachment Management System (IEMS)


Historical methods have benefits — and known drawbacks
Employing nearly 20,000 people across more than 10 U.S. states, this leading energy company is charged with supplying many forms of energy, including natural gas, to over 6 million customers.

Along with many other natural gas utilities, they have directed their pipeline encroachment management with conventional aerial, vehicular, and manual surveillance. Although these methods have been effective in many areas, they have also brought challenges. For example:

  • The gas utility’s traditional pipeline patrol methods had several issues, including blind spots, location errors, subjective evaluations, and cost.
  • Field investigations of pipeline encroachments are inherently limited in scope and efficiency.
  • Use of aircraft with human observers have limited effectiveness with higher expense. Inconsistencies and inaccuracies included missed encroachments, mismatched images, and difficulty detecting patterns over time.
  • Aerial patrols also presented a significant safety risk.

With these challenges in mind, the gas utility chose to pursue fresh ideas and innovations in gas pipeline management, including satellite-based pipeline surveying. 


Satellites and AI bring practical innovation

With a combination of satellite images and AI, AiDash Integrity and Encroachment Management System™ (IEMS™) uses high-resolution multispectral satellite images and on-ground reports to increase visibility of encroachments on pipeline rights of way.

AiDash conducted a satellite-based pipeline survey of 60-foot and 1,400-foot corridors to detect 12 kinds of encroachment, including vegetation, construction activity, erosion, excavations, debris, changes in river courses, and even dead or mired livestock. Rigorous AI models analyzed the complex inputs from geospatial and time-based datasets. They identified encroachments and assessed risk for compliance and other integrity threats — all without sending up a single aircraft.


Reliable, efficient, cost-effective management of  encroachments  

AiDash IEMS quickly and efficiently found 570  encroachments in the initial pipeline survey, including construction of new roads, foundations, and timberland  clearing. It even found a right-of-way violation and 15 structures that had never been noticed before. These were automatically assigned to the nearest crews for follow-up, marked as complete in the application, and verified in future scans.

This utility has found 3 sources of value from AiDash Integrity and Encroachment Management System:


  • All encroachments were located more precisely, helping crews resolve issues quickly.
  • Assignments to crews were automated, accelerating  response.
  • Surveys were accurate in both 60- and 1,400-foot pipeline corridors.


  • Analysis of earlier evaluations allowed trends to be identified and managed.
  • Easy-to-use digital workflows saved time and reduced errors.
  • AI’s self-learning ability helps it improve its own accuracy over time.

Lower cost  

  • Satellite imagery and AI offer better information at a lower cost than aircraft patrols.
  • Wider coverage finds more encroachments per day and consumes fewer resources.


Self-learning AI improves results over time

Satellites and AI can replace traditional aerial inspections of  pipelines and improve encroachment management. These  modern, intelligent tools also help to find improvements in field-operations productivity, encroachment-management efficiency, class-location analysis, and more.

The value increases over time as growing historical records are used to spot trends, predict danger zones, and plan accordingly. The AI models continue to refine their predictive abilities by comparing their estimates to confirmations from the field teams. As a result, the system grows more efficient and more accurate the more it’s used.