Time Series Aggregation for Optimization: One-Size-Fits-All?

One of the fundamental problems using optimization models that use different time series as data input, is trade-off between model accuracy and computational tractability. To overcome intractability these full models, dimension input size commonly reduced through aggregation (TSA) methods. However, traditional TSA methods often apply a one-size-fits-all approach based on common belief clusters best approximate also lead to aggregated approximates model, while metric really matters – resulting output error in results not well addressed. In this paper, we plan challenge show output-error with theoretical underpinnings have unprecedented potential efficiency accuracy.