Risk Modeler 2.0: Commitment to Building the Ultimate Risk Modeling Cloud Platform

The transformation of large, complex, and monolithic applications to a cloud microservices architecture is no easy feat. The launch of Risk Modeler 2.0 – an innovative modeling application, offering one-of-a-kind proposition to catastrophe risk modelers and to all stakeholders in the (re)insurance industry – on the open, modular, and unified platform, Risk Intelligence™, reflects the culmination of eight years of evolution and commitment from RMS. A commitment to clients. And a commitment from clients.

A Vastly Improved Unified Modeling Experience

Risk Modeler 2.0 will be available to our customers in June 2020, offering our core version 18.0 and 18.1 RiskLink and Specialty models. Between June and September, we will also be making available 7 HD models. In addition to the U.S. Inland Flood and U.S. Wildfire HD models, we’ll be adding the following models in Risk Modeler 2.0.

• Europe Inland Flood that covers 18 river basins and over 8000 catchments across 15 countries in Europe
• Japan Earthquake & Tsunami and New Zealand Earthquake with both models covering including sub-peril coverage for tsunami, fire-following, liquefaction and landslides
• Japan Typhoon and Flood which incorporates the most recent events such as Jebi and Hagibis, and 
• Europe Severe Convective Storm, a brand new model that cover 17 countries with large pan-European event sets and fully integrated with our Europe Windstorm model. 

Risk Modeler 2.0 delivers major improvements for the modeling application:

• RiskLink and HD Models under one roof: It is not just combining modeling engines we are also bringing in the ability to combine model results together in Risk Modeler 2.0. You will also have side-by-side versioning of models. For example, going from version 18 to a newer version of our models is as simple as selecting the new model version from a dropdown
• It offers an all-in-one experience for RiskLink and RiskBrowser users. It is simple for our existing users because it is designed to replicate the familiar portfolio and account modeling workflows of RiskLink and RiskBrowser but with superior usability
• Risk Modeler is built to bring far superior total cost of ownership to surface. We see over 5X performance improvement in model execution times. That means, the platform comes with unmatched performance and bring the modern elasticity of the cloud, to give you the most savings
• Finally, Risk Modeler is part of the suite of integrated applications that’s part of Risk Intelligence. That means you don’t have to transform and translate information back and forth between Risk Intelligence applications. The exposures and losses you process through Risk Modeler are available to other applications within the Risk Intelligence “IQ applications”.

Risk Modeler’s enhancements extend beyond the features customers will notice. We have rearchitected Risk Modeler that allows engineering teams to make frequent, small releases that accelerates access to the latest models, as shown in Figure 1.

What makes the release of Risk Modeler 2.0 different? Two words: Occam’s Razor. William of Ockham, a fourteenth century philosopher and theologian proposed a problem-solving principle that states, “… the simplest solution is most likely the right one.” Simplest, in this context, means choosing the solution with the fewest assumptions. Software solutions are ever-increasing in size and complexity as the requirements of application solutions also expand. Complexity introduces uncertainty, and higher uncertainty means greater risk; the greater the risk, the higher the probability of failure. 

The tech industry has many examples where the principle of Occam’s Razor might have been abandoned. Microsoft Windows 8 sticks in the collective memory of the technology community. Launched in 2012, its Metro interface introduced a new, modern structure for users who had grown up on Windows, to work on and achieve a wide range of tasks. It worked fine on touchscreens using the company’s Surface tablets, but on laptops and desktops it left users out in the open ocean with no way to get back to shore.

In the earlier versions of Risk Modeler, we made a number of workflow changes to optimize specific tasks. However, some of these early assumptions generated a level of change complexity that was too much for some of our clients, many of whom had developed long-standing practices and tools based on 30 years of RMS catastrophe management solutions. 

We heard and listened, and for Risk Modeler, we took the lessons of Occam’s Razor to heart. In fact, over 400 users provided feedback to validate, enhance and simplify the model execution and risk analytics workflows in Risk Modeler 2.0 through live demo sessions, hands on workshops and a preview program where customers test the system for weeks.

The feedback helped us find ways to eliminate retraining for our existing users and unify our model engines. Enter a multi-phased project, ambitious in scope, that would first require building a cloud-based platform architected around an open standard. Next, re-organize and improve the design of the user interface. The result: a powerful modeling application, delivering innovative new functionality, with an intuitive and familiar look and feel, as shown in Figure 2.

In making these interface improvements, the application experience aligns with RMS existing users’ modeling workflows. RMS knows how users engage with their modeling products since these users have been doing it for three decades with the on-premise modeling applications, RiskLink and RiskBrowser. This is a shift that removes assumptions and uncertainty and favors simplicity over complexity. The transformation of Risk Modeler 2.0 captures the familiar experience embedded in user practices, so its clients can leap forward and benefit from all the advantages that the cloud has to offer. 

Learn More!

To really explore Risk Modeler 2.0 and Risk Intelligence, please join us at the Exceedance 2020 Virtual Event (May 4-7) with many sessions – that allow you to understand why these solutions really represent the true migration of models to the Cloud.

Find out more by clicking here.

¹ Alan Baker (2010) [2004]. "Simplicity". Stanford Encyclopedia of Philosophy. California: Stanford University. ISSN 1095-5054.