Skip Navigation

Reduce costs by optimizing actuarial risk compute on Azure

Posted on September 6, 2018

Principal Program Manager, Industry Experiences Cloud + Enterprise

Actuarial risk modeling is a compute-intensive operation. It employs thousands of server cores, with many uneven workloads such as monthly and quarterly valuation and production runs to meet regulatory requirements. With these compute-intensive workloads, actuaries today find themselves trapped by traditional on-premises systems (grid computing on hardware) which lack scale and elasticity. Many find they cannot even finish simple tasks like production runs with enough time to spare to correct an error and rerun before a deadline. With scalability and elasticity being the cornerstone of the cloud, risk modeling is incredibly well suited to take advantage of this near bottomless resource. With Azure, you can access more compute power when needed, without having to manage it. This translates to a great savings in time and money.

image

Your immediate savings comes from reallocating your costs from hardware investments to operational expenses. With on-demand compute, personnel are released from hardware and software burdens. They can devote their time and expertise to other production costs, such as writing scripts for the optimum deployment of cores.

More power, less time, faster reports

New regulations, such as International Financial Reporting Standard 17 (IFRS 17), Solvency II, and Actuarial Guideline XLIII, are increasing the pressure by requiring significant actuarial modeling and additional compute power. The efficiency of the compute grid helps reduce the time that senior actuaries and expensive outside consultants need to wait for results. This reduces the overall cost of generating needed reports. This efficiency also leads to improved predictive modeling, which helps enhance risk management. Finally, the increase in modeling speed yields more accurately priced products and better optimized financial positions.

Cope with the increase in reports

Partner-based solutions running on Azure have features that facilitate reporting under International Financial Reporting Standard (IFRS) 17, the new standard governing insurance contracts. These partner-based solutions provide insurers with a structure to meet the 2021 implementation deadline for the standard. Solutions must offer insurers an automated, secure, and flexible way to manage a complex, data-intensive process and perform key calculations. The solutions do this while helping to bridge actuarial and financial reporting functions with integration to the existing actuarial systems.

Reduce the friction between systems

IFRS 17 forces interactions between finance and actuarial systems. This necessitates the exchange of more data between the two domains, while placing additional time constraints on the actuarial risk process. And that’s where a managed cloud environment for your applications comes into its own. Free from the constraints of fixed data centers, the cloud’s pay-as-you-go cost model allows you to scale hardware rapidly up and down to your changing requirements. At the same time, you can free up in-house IT resources to add value in other areas – transferring any related technology risks to your managed cloud service provider, with its combined knowledge of both the software application and the cloud.

Modular capabilities

Most insurance companies work with a purchased third party modeling solution. They customize the solution with their add-ons to build their risk analysis solutions. Many of the solutions include add-on packages which allow actuaries to run models in an on-premises grid or Azure. In cases where the vendor does not offer a cloud solution, insurance companies augment the systems with process schedulers and cloud scaling tools. To interpret the results, the insurer adds reporting functionality as needed. On Azure, they can use tools like Power BI to summarize the terabytes of data. Individual users can even use bookmarks in Power BI to tell a story with the reports. 

Swing at will from low use to high

For insurance companies, actuarial workloads are increasing. With increasingly complicated risk models and new broader regulations, the demand for calculation capacity will grow as well. The cloud is the only feasible solution to meet this growing need. Azure can meet these compute needs with a wide range of hardware and service options available globally. Azure can meet the wild swing of compute demand throughout the year with calculation capacity on demand orchestrated by different services.

Variable sizes to fit

Insurance companies can choose to build solutions to meet these needs or work with our partners to deliver a solution. Mid and small size insurers may prefer the simplicity of using Software as a Service (SaaS) solutions from one of our partners. Large insurers will utilize a mix of SaaS and non-SaaS solutions from one or more of our partners—again backed by the reliability of Azure. With a non-SaaS solution, insurers can add an orchestration layer to managed automation tools and manual processes.

Recommended next steps

Read the Actuarial risk compute and modeling overview to understand your options for moving your compute workload to Azure.

I post regularly about new developments on social media. If you would like to follow me, you can find me on Linkedin and Twitter.