Optimized Robust Combinatorial Human Incentive Design Structures (ORCHID)

While the DoD has tried to take advantage of collaborative APIs across many programs, parties often make a series of seemingly unimportant decisions that unintentionally lead to the development of customized systems and APIs. As a result, there has been a stream of stove-piped system development that puts the DoD at a disadvantage in rapid evolution, restricting innovation, and making the integration of new capabilities more complex. Many of these current processes result from policies that cannot be easily changed; instead, the DoD needs a system of incentive mechanisms that drives stakeholders away from stove-piping and encourages the use of collaborative APIs. To address these needs, Charles River Analytics proposes to design and demonstrate the feasibility of a platform to explore Optimized Robust Combinatorial Human Incentive Design Structures (ORCHID). ORCHID will draw from the empirical foundations of behavioral economics, mechanism design, and judgment and decision-making science to provide decision-shaping nudges through probabilistic, incentive-structure choice architecture models. To test and optimize these incentive structures to contexts and decision-making bottlenecks, ORCHID will combine a first order simulated model of DoD ecosystems and actors with discrete combinatorial optimization via stochastic search, assessing the most effective incentive configurations through Monte Carlo simulations.