AI

How IBM PAIRS Helps Agribusiness Generate Better Forecasts, Yields

Share this post:

On February 6, IBM announced it would make IBM PAIRS Geoscope – the company’s AI-enabled cloud technology built for geospatial-temporal data – widely available, offering it to data scientists and software developers in industries as diverse as agriculture, finance, retail, energy, and government. IBM had previously made PAIRS (Physical Analytics Integrated Data Repository & Services) available as an experimental research offering. PAIRS Geoscope lets users quickly search massive, complex weather, satellite, and other geospatial-temporal datasets, to help uncover insights that can improve their business or solve other problems.

IBM PAIRS makes generating insights from geospatial-temporal data sets – which can exceed hundreds of petabytes across various formats — much less labor-intensive than it is now. One key factor setting PAIRS apart from other analytics platforms is its ability to ingest, index and manage geo-spatial data, such as high-resolution images into common formats that can be more easily searched. That means data scientists and developers can spend far less time preparing data, freeing them up to focus on running queries and analyzing the results.

Among other things, PAIRS is currently being used to help government agencies predict the impact of natural disasters, including the spread of wildfires. It’s also the underlying technology for Watson Decision Platform for Agriculture and The Weather Company Vegetation Management Predict solution, enabling these solutions to tap into terabytes of multi-layered geospatial-temporal data to help agriculture, as well as energy and utility companies better predict everything from crop yields to vegetation-related outage risks.

Curtis Jones, Vice President, Global Economic Analysis, Bunge.

One such company taking advantage of PAIRS is Bunge, a leading global agribusiness and food company. The THINK Blog recently sat down with Curtis Jones, Vice President of Global Economic Analysis for Bunge, to talk about the practical application of PAIRS and its potential to reshape industries that rely heavily on large amounts of geospatial data. The following is an excerpt:

THINK Blog: How does Bunge use PAIRS?
Curtis Jones:
PAIRS provides a very granular resolution of historical weather data, and the satellite data it offers is accurate to within 250 meters. For crop forecasting that level of accuracy helps us determine what ’s likely planted in a given location, what the yields were and all the weather variables for that particular location. That enables us to build much more intricate models than we could if we were working only with large, discrete data sets whose resolution is not as fine as what PAIRS offers.

THINK Blog: What impact has PAIRS had on the way Bunge manages and analyzes market data?
Jones:
We’re a global company with a very complex global supply chain, so it’s natural for us to rely on geospatial temporal data for our business. We have lots of proprietary data we analyze in a traditional sense — that hasn’t changed — but the market also has a lot of public information, such as weather, U.S. crop production and satellite data that we can take advantage of. PAIRS allows us to apply advanced statistical methods and lots of computational power to those public data streams as well as our proprietary data. That lets us do far more sophisticated analysis of crop production and land utilization data than we were capable of previously. We simply didn’t have the computing power or a way to gather that data in a timely fashion. PAIRS has also opened new doors that allow us to work with much larger databases and apply machine learning techniques to that data.

THINK Blog: How does access to larger amounts of data and additional processing power enhance Bunge’s ability to apply machine learning?
Jones:
The numeric methods that underlie machine learning have been known forever, but they require huge computational capabilities to make them useful. They also need a mountain of data to push through it, otherwise you get very spurious results. Think of it this way: statistically there are nice, neat mathematical packages such as regression analysis that can find important relationships among different variables. To optimize that process, you need a lot of granular data and a lot of computing power. PAIRS arranges all of that data geospatially and is a good architecture for running queries. The more you pound on it the better it performs, actually, which is pretty remarkable.

THINK Blog: What’s the benefit of all of this being available via the cloud?
Jones:
Computational power has increased massively with the use of parallel processing and graphics processing unit (GPUs) as opposed to CPUs, and data storage prices have plummeted. Cloud computing makes all of those benefits accessible to anybody with a laptop or a terminal.

THINK Blog: What potential does Bunge see for the future of PAIRS across industries?
Jones:
Think of it as a platform on which anybody who has a geospatial problem set—such as mining, petroleum exploration, or agribusiness—and a lot of proprietary data can do incredibly sophisticated analysis. Having that data on demand is probably our biggest proprietary edge. For an agriculture company, the technology is here to stay, you can’t walk it back. The way people do things, such as in crop forecasting, where there’s a lot of public data that you can make use of, that won’t change. In fact, the need will just grow.

More AI stories

How AI, IoT and Weather Tech Can Help Better Detect Deadly Wildfires

As summer temperatures in the northern hemisphere rise, so too does the risk of wildfires and the threat to life and property from the western U.S., to Europe and across Siberia. In fact, firefighting teams are working to contain wildfires across central Portugal as I write. Unfortunately, this trend shows no signs of slowing. According […]

Continue reading

FOX Sports, IBM Team Up to Transform Production

The eighth edition of the FIFA Women’s World Cup™ is well underway, with teams from 12 countries battling it out for the championship title. While millions of soccer fans stay tuned to the excitement in France, IBM is teaming up with FOX Sports to help transform production of the event by infusing AI analysis and […]

Continue reading

Making Monitoring AI Bias a Little Easier

When we launched Watson OpenScale late last year we turned a lot of heads. With this one solution, we introduced the idea of giving business users and non-data scientists the ability to monitor their AI and machine learning models to better understand performance, help detect and mitigate algorithmic bias, and to get explanations for AI […]

Continue reading