View all projects

Leveraging AI-driven LiDAR analysis to enhance track maintenance efficiency, improve safety and reduce manual inspections

Sydney Trains: Advancing track maintenance with digital inspection
Sydney Trains 2
Transportation
Rail & Transit

In 2021, Sydney Trains sought solutions for a series of track inspection use cases to consolidate maintenance and compliance efforts across its network. The goal: improve safety, accuracy, repeatability and efficiency of track inspections by generating greater value from regular LiDAR (Light Detection and Ranging) data collection. 
 

The focus was on using the LiDAR outputs to replicate and automate several tasks traditionally performed in the field, such as assessing tunnel and track clearances, track centers and vegetation encroachment, identifying ballast deficiencies and analyzing signal sighting.
 

The solution

Jacobs’ digital solutions team co-developed a suite of digital tools using rules-based algorithms, artificial intelligence and machine learning techniques to analyze LiDAR data captured during routine rail corridor inspections. These tools streamline inspection processes across six core use cases: track center determination, kinematic envelope clash detection, vegetation management, ballast deficiency, signal sighting and rail head position.


Jacobs worked closely with Sydney Trains to define use case requirements and parameters, including existing rolling stock specifications (used to define the kinematic envelope — the area trains occupy in motion), sleeper types and track geometry. This foundation allowed the team to build automated workflows that deliver reliable, repeatable and accurate measurement.  


By combining these parameters with regularly updated LiDAR data - captured via a mobile laser scanning system mounted on Sydney Train’s Mechanized Track Patrol Vehicle (MTPV), the team automated several previously manual tasks. These include detecting vegetation encroaching on the kinematic envelope, identifying potential clearance issues, spotting ballast deficiencies, and pinpointing signal sighting. These processes are performed at one-meter intervals across Sydney Trains’ 932 mile (1,500 kilometer) operational network. 


Data from the latest MTPV scans is incorporated into Jacobs’ digital solutions for each use case, generating detailed analytics and reports. The outputs provide actionable, easy-to-understand insights that enable proactive maintenance across the track corridor. 

  • Sydney Trains 1
  • Sydney Trains 2
  • Sydney Trains 3
  • Sydney Trains 4
Reality capture within the rail corridor. Point cloud and imagery provide the basis of the digital twin geometric framework.
Identifying infringements to a kinematic envelope has been automated through the entire rail corridor.
Passing clearances based on rail track centers, highlighting areas of potential rolling stock clash. 

AUTOMATED IDENTIFICATION OF VEGETATION MANAGEMENT NEEDS.

2.5

Years spent co-developing digital solutions to meet Sydney Train’s operational goals.

6

Number of use cases for which Jacobs developed automated inspection solutions: track Center determination, kinematic envelope clash detection, vegetation management, ballast deficiency, signal sighting, and rail head position)

2,000

hours spent designing and implementing data driven, automated inspection outputs.

Project status

The project has transitioned from trial phase to business as usual. 

The suite of digital tools is already helping Sydney Trains make more informed and strategic decisions about maintenance priorities. Ultimately, the solution aims to eliminate manual track inspection processes entirely, reducing health, safety and environment risks to Sydney Trains’ staff and lowering costs associated with manual inspections. 

Regular acquisition of LiDAR data provides a geometric framework to support more advanced digital twin concepts. Future enhancements, using data from additional sources, will enable further customization.

  • Jacobs has consistently delivered high-quality outcomes within tight timeframes, providing accurate and timely data that reduces the need for manual track measurements across a range of inspections. The service also strengthens Sydney Trains’ predictive, data-driven asset maintenance capabilities.

     
    Kamran Kermanshahi
    Senior Manager, Mechanised inspection Services, Sydney Trains

Key team members

  • Richard Lemon, Project Manager and Survey

    A highly respected industry leader in the survey and geospatial professions, Richard brings expertise in major development programs to the project.

  • James Leech, Data Scientist and Rail SME

    James leads the development and implementation of data analytics tools to optimize specific use cases tied to compliance needs for Sydney Trains’ track inspection activity. 

  • Brett Sheehan, Surveyor and Data Analytics

    Brett is a highly credentialed surveyor with specific expertise in mobile laser scanning. He uses his deep domain knowledge in measurement science to workshop and develop solutions for each use case.