CRN Impact Awards: Kablamo helps fight bushfires with AWS machine learning technology

Sydney-based Kablamo has been named one of the finalists in the upcoming 2021 CRN Impact Awards, with its work on helping fight bushfires with technology securing a spot under the Business Transformation category.

The company was tapped by the Department of Environment, Land, Water and Planning Victoria (DELWP) to develop a cloud-based machine learning platform to improve its computing backbone to power the Department’s bushfire prediction modelling.

DELWP was using existing modelling framework to manage bushfire risk, calculating the likelihood and severity of a bushfire caused by fuel levels on public land.

The platform, called Firetrail, was built on Amazon Web Services, using its machine learning capabilities to handle large increases in workloads and processing with existing and future prediction models.

Kablamo and Firetrail was brought in to address some challenges like overcoming compounding issues in the previous metric due to outdated data layers and models; bottlenecks that prevented scalability; adding new opportunities like increased spatial resolution and extending the application of new data resources (such as weather scenarios, temporal scales and ignition patterns).

The Department’s existing PHOENIXRapidFire bushfire modelling saw improvements from Firetrail, making it more accurate by feeding in clean data from geospatial, temporal and historical sources.

Firetrail also provided data storage and processing on a scale over 25x more than before; decreased processing time e from 16 hours to 4 hours; increased specificity of grid parameter from 5km to 2km; make 40 years of Victorian data instantly accessible for better predictions, adding new geospatial advances; and eliminating bottlenecks and manual workflows through automated data pipelines and serverless processes.

Some of the AWS tech involved include Amazon Glue to replace a PostgreSQL-based solution for post-processing, allowing to run several data transforms concurrently instead of serially; and Amazon SageMaker to better metabolise data from university-led research and models.

Aside from the CRN Impact Awards nomination, Kablamo also got the nod from AWS at one of its global events, securing the 2021 AWS Global Public Sector Partners Award for Most Innovative AI and ML solution.

AWS vice president of worldwide public sector partners and programs Sandy Carter said the winners, including Kablamo, are “essential in driving innovation, accelerating digital transformation, and delivering results for our AWS customers … and in helping public sector customers achieve their mission.”