Court scheduling has always been a complex problem with many variables. The task is currently performed by a department of individuals. Three Trees was tasked with finding a way to capture the team's institutional knowledge allowing senior schedulers more flexibility when taking vacations or retiring.

We opted for a supervised learning system that relied on complex visualizations. The AI determined the probability of every actor's utilization and assembled the most workable schedule. Scheduling staff would then review the effectiveness and either reinforce the assertions that led to the correct decision or reject those that the human reviewer felt was incorrect.

The system state was depicted through visualizations with mutli-level symbology. Interactive filters called attention to various aspects of the data. The visualizations can be browsed to focus on various regions and facets. By the end of the project a yearly schedule could be determined by AI and viewed in one animated interface.

Zynergy’s primary focus is moving projects through various stages of review. The interfaces adjusted to highlight the information and tasks most relevant when users of different departments used the system. User could toggle their optimized view to another job focus entirely and gain greater understanding of the project. This system uses an adaptive interface pattern throughout to optimize departmental views and provide flexibility to managers without maintaining multiple interfaces.