Willis Towers Watson has launched Radar 3.0, the latest edition of its pricing software, which includes machine learning approaches for greater pricing sophistication.
Building on the last major release of Radar in 2016, Radar 3.0 implements a wider range of machine learning models, enabling a greater number of analytical techniques to deliver more effective pricing approaches, the company explained in a statement.
“Processing speed is also up to five times faster, meaning companies can refine and test pricing approaches more rapidly,” Willis Towers Watson said.
“With competition intensifying, changes in regulation and distribution, and the ongoing redefinition of consumer expectations, we are seeing a clear and widespread focus on pricing sophistication and effective customer management,” said Duncan Anderson, Willis Towers Watson’s Global P&C Pricing and Product Management leader.
“As part of that, insurers in many markets are now actively incorporating machine learning models in their pricing approaches, not only in backroom analytics but also in the live deployed rates. Radar 3.0 supports all of this,” he added.
Willis Tower Watson provided the features of Radar 3.0 that aim to facilitate improvements to modeling and pricing:
- New machine learning capabilities – Machine learning models in Radar include Gradient Boosting Machines (GBMs), Random Forests and Elastic Nets.
- Performance – A new high-performance mode for 64-bit environments enables faster processing for large datasets by making greater use of available RAM. Processing speed improvements of up to five times can be achieved using suitable hardware.
- Connected reporting- Radar 3.0 allows users to create hyperlinks that allow dashboards to be created with integrated workflow or guidance.
- Enhanced platform support – The latest version of Radar also provides enhanced support for cloud-based platforms via Microsoft Azure and is qualified for use on the Windows 10 anniversary edition.
Source: Willis Towers Watson
Was this article valuable?
Here are more articles you may enjoy.