This paper highlights the authors’ attempts to develop a software environment to help analysts (1) improve their understanding of the dynamics of the energy system and (2) build confidence in a complex, highly defined system dynamics model of energy supply and demand. The envisioned organizational learning environment has at its center a human interface design from which the user controls simulations, exercises, guided tours, model modifications, experiments, etc. While efforts to date have focused upon developing a micro-computer based system, the learning environment will be extended to workshops in the future.Any energy simulation model can be intimidating to users who lack a sufficient understanding of supply/demand concepts and knowledge of the computer language employed. The situation becomes more complicated as the complexity of the simulation model increases and new operational (or system performance) concepts are introduced.We provide a brief description of an energy supply/demand simulation model, ENERGY 2020, that central Maine Power uses for demand forecasting. Developed by George Backus and Jeffrey Amlin, the long-term energy policy model is used by analysts for both utility-level planning as well as state energy policy analysis. The model is coded in PROMULA for IBM-compatible personal computers.The problem is to create a means by which new users can explore and understand the complexity of the energy system (the ENERGY 2020 Model) in a structured and non- intimidating manner. The goal is to bring the users to a higher level of understanding, mastery, and ownership of the model.