Latest Enabling Technology Article
Latest update: May 2019
What is GIS?GIS is a computer system built to capture, store, manipulate, analyse, manage and present geospatial data - information with a geographical component such as GPS data, satellite imagery and geotagging. Ordnance Survey estimate that 80% of all information globally contains a geographical component, hence the location dimension of GIS can add another layer to conventional analytics. GIS allows users to perform calculations and analysis on geospatial data, edit the data in maps and presents results. Key applications of GIS are spatial planning, decision-making and asset management.
How will GIS impact the rail industry?GIS based asset management systems can be used to improve the safety, reliability and efficiency of operations for rail. For both static and dynamic infrastructure, GIS can determine the potential impacts of an asset’s location and scale, as well as its proximity to other assets. For example, the Jubilee, Northern and Piccadilly lines on the London Underground (LU) have linked their GIS and Enterprise Asset Management System (EAMS) to provide a clear view of works and assets. Geospatial information is used to plan maintenance work packages, optimising use of limited maintenance windows overnight and saving both time and money.
Data integration: GIS provides a platform to integrate layers of data using the common geographical component. Layers can be adjusted according the user’s requirements and can show changes over time. Ordnance Survey data, or other contextual data, can create good quality background mapping. Operators can establish a comprehensive GIS to maintain current and accurate location-based representations of their networks.
Geospatial data can improve the relationship between disparate information systems, such as Computer Aided Design (CAD) and asset management. Likewise, Building Information Modelling (BIM) can be geographically enabled using GIS. Projects are using GIS to provide greater visualisation at each stage including planning, operation and reporting. Crossrail created map-based reporting directly from the Tunnel Boring machines. Up to date information is accessible to all parties via mobile devices, creating more informed decisions throughout.
Flood risk prediction models: A GIS-based model was created by the LU Comprehensive Review of Flood Risks (LUCRFR). Flood hazard maps can be produced automatically, and flood risks and assets are easy to search for. Real-time weather data is beginning to be incorporated to further enhance the functionality.
What other industries are using it?Emergency services create crisis maps by collecting data from sources including satellite imagery, remote sensors and crowdsourced data. GIS is being used to generate a common operational picture of incidents in shared databases, helping to coordinate responses. For time-sensitive decisions, accurate, real-time geospatial data is important, and a greater volume of data can improve accuracy and precision.
Uber uses GIS to track the changes in demand at locations throughout the day and notifies drivers to maximise revenue. A large volume of data can be managed by the intelligent algorithms, classification and predicative capabilities of artificial intelligence (AI). Geospatial artificial intelligence (Geo.AI) combines AI with GIS. Traditional GIS are primarily outdoor focused and plotted in 2D. Indoor wayfinding can help plan for maintenance and emergencies. Building-wide maps enable the user to navigate through the indoor environment without physically entering it. Furthermore, a 3D GIS enables the user to depict additional detail in a third dimension, which could be geographical height, or location characteristic.
What is the current state of R&D?
- A pilot Geo-Referenced Safety Risk Model (GeoSRM) was developed by RSSB and the University of Southampton in 2016 as part of the project T972 Piloting a geo-reference safety risk model for the rail network in GB. It covered 10% of the GB rail network and created localised risk estimates of three Safety Risk Model hazardous events (train derailment, suicide, and slips, trips and falls), by considering local features.
- The Rail Infrastructure Network Model (RINM) project uses aerial surveying and LiDAR to construct a picture of the railway network for surveillance. After a successful pilot, it is aiming to add real time GIS data to the model.
What challenges remain for GIS?The accuracy of GIS is affected by data sources. To be effective, GIS needs up-to-date, reliable geospatial data, to accurately represent a situation. Reliable data is also essential for an AI system.
To enable real-time updates, data architectures will need to support the quantity and variety of incoming information. Geo.AI is one solution to GIS data management. Furthermore, a large volume of visual data requires significant computational power and storage. As commercial software and cloud-based solutions can be expensive, users are transitioning to open data sources such as OpenStreetMap (OSM). Users should be aware of open data policies and consider cybersecurity risks.
What should the rail industry do?
Rail could expand/improve GIS applications in the following areas:
- Develop appropriate methods to manage the quality, accuracy and availability of the data used. For example, Crossrail created a Spatial Data Infrastructure as an extension to its GIS.
- GeoSRM could be expanded to model a larger set of hazardous events and a greater geographical area.
- Incorporate weather susceptibility mapping into the GIS for a better understanding of extreme weather conditions and impacts on rail infrastructure. The LUCRFR project could be used to learn good practises for flood risk mapping.
- New industry datasets could be made suitable for use with GIS.
Rail could investigate applications where GIS can be integrated to increase the capabilities of other technologies. LiDAR could be used in conjunction with satellite observations to manage both weather and climate related risks and for general track surveillance. Augmented reality could be combined with interactive GIS maps to better visualise topography and weather conditions, such as expected snowfall across the network.
Banner image: Image author Daniele Masi [CC BY-SA 4.0], via Wikimedia Commons