The application of data analytics with a spatial dimension is extending across most industry segments and yielding best results. Powered with Cloud technology, a comprehensive database helps companies in predictive analysis, identifying the errors and finding the best solution in less time.
IBM Research has developed a technology that can get insights from geospatial-temporal data in minutes, thus saving a considerable amount of time.
The technology, known as IBM Pairs Geoscope, is a cloud-based service that utilizes hyperspectral imaging. Recently IBM successfully deployed it for reducing power outages by assisting energy companies to predict where trees and other vegetation may pose a risk to power lines.
Hendrik Hamann, Manager, Physical Analytics, IBM Research, says “PAIRS Geoscope is a unique cloud-based data store and data analytics service that can rapidly explore and analyze massive complex geospatial-temporal data sets and find new insights. What would typically take months with other existing technologies can now be done within minutes”.
Drastically reducing time, cost-effectiveness and high accuracy makes Pairs Geoscope a pioneering technology.
Processing multiple data streams
Geoscope collects and utilizes data from multiple sources, including satellite and drone images, IoT devices, cell phones, social media, population records and weather reports. This not only enhances precision but also greatly reduces the probability of error margin.
“Hyperspectral images, usually captured by remote-sensing satellite or aerial systems, are one source of data that PAIRS can ingest, curate, and seamlessly integrate with other forms of geospatial-temporal data for efficient retrieval, query and downstream analytics”, Hamann adds.
PAIRS Geoscope has opened a new chapter in the analysis of multiple datasets in the shortest time span. This is a seminal development in analytics and will spur further innovation.
PAIRS Geoscope acquires more than 10 terabytes of geospatial-temporal data, which is a new breakthrough considering that prior to this, inability to access, query and analyze this class of big data in a scalable way was the reason much of this data remained relatively “dark”
“PAIRS represents a dramatically new way to discover insights from this frontier of big data, with a ready-made catalog of carefully indexed, diverse, and continuously-updated geospatial-temporal information, enabling scalable access to complex queries and machine learning-based analytics and AI without the need for downloading data”, says Hamann.
Also Read: Amazon launches AWS Ground Station and how it would turn earth observation industry on its head
Versatile applications
Hamann also adds that while PAIRS is the underlying engine for this new vegetation management solution for the energy & utility sector, it is generally available and can be used across industries and use cases.
In the energy and utilities sector PAIRS offer a host of advantages.
“With this new solution, hundreds of miles of transmission and distribution lines will be regularly monitored to provide continuous insight about the state of growth and maintenance, replacing costly on-site inspections or unnecessary trimming”, adds Hamann.
Apart from identifying and predicting outage threats, geospatial-temporal insights can help with overall grid reliability and compliance, wildfire prevention, storm management and assessment, he says.
Citing an example of a use case, he says, PAIRS is already being used in the farming sector.
IBM PAIRS is suitable for agriculture to accurately predict the mix of various crop yields essential to agribusinesses.
“Using massive amounts of satellite data in combination with weather, soil, and land use information, PAIRS Geoscope helps forecast the supply of major commodity crops such as soy, corn, wheat, and sugar at a national level”, says Hamann.
Emphasizing the versatility of PAIRS, he further says, any industry that can benefit from analyzing massive amounts of data, such as retail, insurance, emergency management, government and more.
Geospatial information, in spite of the great developments and technological advancements, is yet to be harnessed. The emergence of technologies like PAIRS Geoscope could prove to be a game changer in this regard. This would also enable us to address grave challenges like climate change.
Enhanced capacity though convergence
In the coming years it has to be ensured that the humungous increase in geospatial-temporal data is also matched with proportionate enhancement in our ability to process the data using the latest technologies like cloud-based big data and emerging technologies like Machine-Learning and Artificial Intelligence.
Another crucial factor would not be the ability to analyze individual datasets, but to gain insights from thousands of datasets simultaneously, says Hamann.
IBM PAIRS can also be seamlessly merged with AI-based models for greater efficiency
A lot of AI solutions are built on PAIRS. The ability to rapidly develop and experiment with solutions using geospatial data is one of the primary use cases for PAIRS, adds Hamann.
He also believes that considering that a large amount of geospatial data comes in gridded form, applications of deep learning methods are obvious.
“Geospatial data is very heterogeneous, which historically made the rapid development of new models that exploit different data sources a challenge. PAIRS resolve these problems by curating data sets that are relevant to a wide audience”, says Hamann.
Also Read: Geospatial data analytics: The way forward in analytics
