Posters

Abstract

The IEA Wind task 51 “Forecasting for the weather-driven Energy System” initiative “Probabilistic Forecasting Games and Experiments” in collaboration with the Max-Planck Institute for Human Development's WEXICOM project has three main objectives: (1) to empirically investigate the psychological barriers in the industry when dealing with probabilistic forecasts and (2) to gain a better understanding of communication pitfalls and challenges when introducing uncertainty forecasts to end-users and (3) enabling them to exploit the benefits of the enhanced information in their decision process. For this, the group is conducting experiments that appear to be simplified realistic scenarios for many decision-makers in the (wind) energy industry. However, the simplification allows to test the behaviour of participants in a well-defined environment. Feedback from participants in the first two experiments confirm that the simplification and gamification of a complex problem helps break down barriers and enables participants to go through the process of decision-making in a structured way. Thus, the learning-by-doing strategy in a safe environment has been received very positively, even though it may not reproduce the entire context of the decision-making problem. Moreover, the platform allows users to play repeatedly and test different strategies for their final decision— and thus enables professionals to familiarise themselves with a complex topic, to train and to understand challenges and benefits of the new technologies. In this presentation we report first results of our new experiment, where participants had to make decisions given the risk of a high-speed shutdown for a wind farm in complex terrain in 22 randomised cases; decisions were made in randomised blocks, either first based on deterministic or based on probabilistic forecasts. The design allows to compare decisions based on both types of forecasts - as well as the confidence with which participants made their decisions.