Generative AI for Intelligent Scheduling

Create optimized schedules in seconds after PhySS learns from your past schedules.

Shift scheduling is enormously complicated in medicine. New office staff can take months to learn how to create a schedule that meets all the rules and preferences. With PhySS, you can generate schedules in seconds after it learns from your past schedules.

We developed and published a paper on the first generative machine learning model for anesthesia shift scheduling. Instead of writing rules for the machine to follow, you can let the machine learn your scheduling rules, contractual obligations, preferences, and schedule equity from your historical data.

The machine will then generate optimized schedules that meet all of your requirements based on actual schedules, ensuring consistency with your established practices while adapting to new constraints.

Multi-Stage Schedule Generation

Call schedules are the most important to get right for providers. PhySS enables you to create shift schedules in stages. For example, you can create a balanced call schedule and then use that as a base to create the rest of the daily schedule.

Flexible Rule Management

Contractual and regulatory obligations can be codified in the system to ensure that all generated schedules meet those requirements. If you have specific scheduling rules that are not easily learned from historical data, you can also input those rules directly into the system.

The machine will then generate schedules that meet those rules in addition to rules learned from your past schedules. By incorporating both learned and explicitly inputted rules, rare situations can be handled while still generating optimized schedules based on historical data.