An AI model developed to create “digital twins” of injured people can assist insurers in evaluating the efficacy of policy and parameter changes, its developer says. University of Melbourne’s Associate Professor Jason Thompson, from the psychiatry department, reveals that this tool can generate a “synthetic population” of over 20,000 injured individuals and study the impact of alterations in operational models.
Professor Thompson told that the digital twins function as artificial societies and injured client populations. Insurance schemes can test policies and operational models on them, rather than in the high – risk real world. These digital twins have detailed attributes such as names, ages, injuries, and more. The tool can track how thousands of people progress through the system under different scenarios and how their interactions with various system aspects affect overall scheme performance metrics. It examines how changes like increased early intervention and improved service access impact financial sustainability, health outcomes, and client satisfaction.
Professor Thompson adds that the model can help strategy managers assess numerous parameters to determine the best results. Policies can be implemented quickly in the model, and it can show not only how different operational settings affect key performance metrics but also why. The team can identify where problems arise in the model and check if the real system has similar issues. He mentions that his team has designed similar platforms for nearly a decade with funding from various councils. As the tool evolves, it could be applied in other insurance areas.
Related topics:
