Previsão de demanda de serviços emergenciais em concessionárias de energia elétrica

Abstract

The electric utilities need to guarantee the quality of their services. One of the most significant problem that affects this quality is the interruption in the supply, which is many times related to emergencies occurrences that need human intervention to restore the service . Precisely because of the nature of these and their impact due to the generated number, having some knowledge about the forecasted demand for these services regarding the necessary number of hours is relevant and advisable. This study presents a exponential smoothing application to forecast emergencies services in a electric utility. Beyond the definition of the method's parameters, this study also defines the way how the forecast should be considered using the observation of the demand's deviation in days of the week and in the hours of the day. A study case developed with real data shows the methodology application and indicates important questions about the method application in the context.