U.S. federal government agencies oversee a wide array of citizen benefits, which affect millions of Americans. Federal benefits administration is a complex interaction of systems that can be approximated with a modular framework and digital twin. Rather than focusing on individual elements of the benefits administration operation, we aim to minimize interface issues between the elements by modeling the entire operation using a holistic modular framework. We also present a digital twin discrete event simulation of the benefits administration system to measure how much new Artificial Intelligence (AI) technologies improve government services.
Federal agencies responsible for adjudicating health, food, financial, and other benefits are increasingly asked to deliver services faster and cheaper with fewer resources. Inefficiencies within benefits administration can negatively impact vulnerable citizens in need of time-sensitive critical services.
AI solutions are increasingly being used to meet this demand and have consistently been touted as bringing massive innovation to government. Modeling the benefits administration system as a collection of independent agents allows agencies to customize and explore system interactions with minimal business risk.
Discrete event simulation has been used to reduce wait times and improve customer service in healthcare and other citizen service settings. A modular macroscopic-level framework with a benefits application system, application intake, decision-maker network, and decision-maker processes was created to approximate benefits administration operations. In addition to developing the framework, the impact of introducing an AI solution was quantified using a digital twin.
A modular macroscopic framework for modeling the federal benefits administration system is presented. The framework and digital twin can model physical and virtual systems and approximate the interaction of people, processes, and AI technology within the benefits administration context.
The framework comprises a benefits administration system, application intake, a decision-maker network consisting of individuals who determine benefits eligibility, and agency-defined decision-making processes. Customized variables can be applied to any component depending on the agency’s context.
Benefits administration operations can be simulated using discrete event simulation tools such as Simio. The digital twin includes a benefits application system with simple and complex applications routed to either NoAI or AI processes, a pool of decision-makers, and a set of processes used to reach eligibility decisions.
The digital twin is adaptable to system changes and provides a general approach for improving efficiency, operations, and customer service.
A baseline current-state digital twin representing the benefits application system was simulated, and permutations were analyzed. Simple and complex applications arrive independently and are routed through NoAI or AI processes to decision-makers before exiting the system.
Throughput and processing time efficiencies were measured, demonstrating improvements achieved through the introduction of AI technologies.
Simulation methods enable agencies to explore “what-if” scenarios using a digital twin. By perturbing the model, agencies can understand how changes in one part of the system affect overall performance, allowing improvements with minimal operational risk.
Proceedings of the 2020 Winter Simulation Conference K.-H. Bae, B. Feng, S. Kim, S. Lazarova-Molnar, Z. Zheng, T. Roeder, and R. Thiesing, eds.
Accenture Federal Services
800 North Glebe Road
Arlington, VA 22203, USA