Energy management systems currently focus primarily on optimizing a best guess scenario based on a feed-in forecast. Renewable energy feed-in forecasts are subject to uncertainty, resulting in deviations from the best guess, flawed operating plans and higher opportunity costs. Some uncertainties can be modeled using probabilistic forecasts. [1] Main Goals: • Develop a reproducible workflow for generating probabilistic PV Data • Automatically create realistic test datasets that include probabilistic PV feed-in forecasts for use in downstream optimization and analysis