Guest blog: 'Geospatial: Emerging Technologies in Utilities'
Utilities are required to be one of the most stable and reliable service items. If lights don’t illuminate, or the tap runs dry, it’s an instant and costly problem. How then, can new technologies be introduced to transform ways of working without causing major disruption to services?
Geospatial techniques provide one of the largest areas of potential growth for utility providers. Locational data is already tightly interwoven within other ‘big names’ in research and development big data, Internet of Things sensors, Machine Learning and predictive maintenance, to name just a few. But to realise their benefits, utilities will need to have the basics really nailed down to reap the benefits of these technologies, without causing undue disruption to services.
The core issue is locational data. Traditional Common Data Environments such as Projectwise struggle when confronted with spatial formats, so a Spatial Common Data Environment (sCDE) is a must-have. Current uptake of network-wide central data management is low among utilities, with many dependent on data silos within individual departments. This is a limiting factor when pinpointing spots for proactive maintenance, cleansing data for largescale analytics, and even scheduling engineers site visits.
Utility providers need to take a close look at their processes to analyse where an sCDE could improve their data.
Does data have unknown provenance?
Dates on most recent reports or age of system elements can be key to predictive maintenance processes, so having these comprehensively recorded across the dataset, in a single date format, is critical.
What is the precision and accuracy of the data?
Knowing the margin of error when examining the spatial accuracy of data will enable better application of techniques such as Machine Learning thanks to an increased ability to understand limitations of results and build this into decision-making.
Where are the network data’s geometry errors?
Having perfect virtual networks that reflect the real-life connections between assets is a key step towards building a digital twin. However common errors in spatial data can include mismatched connection points between wire or pipe links, or ‘island’ assets without linkages at all. Automated checks around geometry, built into a spatial common data environment, can keep data clean.
Is the data incomplete?
Running checks across silos of data would simply add costs and mean analysing for incompleteness while the silo itself is incomplete. Using automated checks within an sCDE can uncover where gaps exist in data.
Spatial data management runs the full gauntlet, from origin to revisioning and final archiving, and with very large datasets commonplace in utilities, expert versioning, potentially at feature-level, is a core enabler for making data a contributor to profit rather than loss. BIM is of course heavily entrenched to provide guidelines, with alignment to either ISO 19650 or ISO 19115 dependent on which side of the pond networks are on.
Better spatial data can make utility operations healthier well into the future, acting as a firm foundation for digital twins, complex analytics, and predictive maintenance. For utility asset managers, getting data right is just the first step.