Improving the way farmers monitor crops
CNH uses Databricks to analyze geospatial data, boosting crop outputs
Faster analysis of geospatial data to monitor crop conditions
CNH is a worldwide producer of agricultural and construction equipment. However, agricultural data operates on widely different scales, which makes it challenging for the company to effectively gather insights. That’s why CNH turned to Databricks Data Intelligence Platform. Now, the company can drastically reduce the time it takes to analyze massive volumes of geospatial data to identify how weather patterns can impact crop ouputs. Not only does this save costs and boost collaboration, but it also allows CNH to more quickly act on this data to deliver value to its customers.
Scaling issues hampered data insight collection
In modern agriculture and construction, data plays an absolutely crucial role in driving success. For CNH, a worldwide manufacturer of agricultural and construction equipment in over 100 countries, that data presents a unique challenge. Agricultural data, for example, operates on very different scales, from single field-level metrics to identifying global trends.
While connected machines help CNH better compare different planting or harvest seasons and monitor individual machine performance, they can’t provide this at the scale the company needs to further refine their products. One particular challenge is managing geospatial analysis data at scale.
“With tens of thousands of polygons being ingested daily, it was difficult, if not impossible, to add even basic geographic insights into our reports,” William Kafes, Senior Data Engineer at CNH, said. “This meant we couldn’t dig deeper into our data to understand what we could really improve upon.” It was clear that CNH needed a solution that could help them ingest large amounts of data regularly and turn that around into actionable insights, which is why they turned to Databricks and Mosaic AI.
Databricks and Mosaic AI unlock data efficiencies
To better manage large amounts of data, CNH used multiple aspects of the Databricks Data Intelligence Platform. The company used Delta Lake to ingest and store their data, ensuring ACID (atomicity, consistency, isolation, durability) transactions and scalable performance. With Unity Catalog, CNH could implement fine-grained access controls and data governance. Then, the collaborative notebooks allowed the team to fully utilize their skill sets to remove roadblocks quickly and maintain high-quality data.
CNH also tapped into the built-in H3 indexing system and Mosaic AI’s spatial library to bolster their ability to analyze geospatial aggregations, which can help farmers better analyze crop conditions. All of this has led to a more collaborative, agile organization that can operate at scale.
“Before Databricks, many of these aggregations simply weren’t possible to analyze,” William said. “Now, we can elastically scale our resources in a way that benefits our end users.”
Delivering AI-driven agricultural innovations at scale
Thanks to Mosaic AI and Databricks, CNH significantly reduced the time it took to run geospatial aggregations. What previously took a full day can now be done in just two hours.
This greatly reduces storage and compute costs and allows for daily refreshes. As a result, anyone on the team can now efficiently add geographic information to their reports without the need for a specialized geospatial background. This democratization of data has opened up the doors for greater collaboration, time -to -value, and scalability. CNH will only continue using Databricks and Mosaic AI to create more AI-driven efficiencies and find new ways to help their customers do their work more effectively.
“We’re incredibly excited with what’s possible with Databricks and Mosaic AI,” William added. “The power of the platform and the AI has opened new doors for us to gather previously unattainable insights to innovate the technology we provide to agricultural and construction professionals everywhere.”
