Additionally, with only six underwriters, Expion could only respond to about 200 RFPs per year, limiting the company's ability to bring in new business.
machine decision
In August 2023, the Expion IT team, with support from management, launched an internal project to automate complex underwriting processes for prescription cost management.
Expion's two AI developers spent approximately six months creating ExpionIQ Advisor, a tool that uses linear regression, multiple algorithms, and custom-built AI models to automate prescription RFPs. The IT team plans to further enhance the application using the XGBoost machine learning software library to predict drug use in the target population.
ExpionIQ Advisor has reduced the time it takes to calculate the numbers needed for an RFP to hours instead of days, Kumar says. He added that the new application will allow employees to see their work.
“Now we know exactly what calculations we're using and what risks we're taking. And if the model is predicting something wrong, we know exactly what that problem is. We know exactly what to do to solve the problem,” says Kumar. “In the past, anyone could easily make a mistake. Now, when a model moves away, it automatically alerts you and fixes it.”
A web-based .NET application on the front end of ExpionIQ Advisor allows insurance companies and other users to log into the system and view numbers. “The .NET application pulls everything together, does the final calculations, presents that data in an easy-to-understand way, and delivers the printout to the end customer,” says Kumar.