Mapping Flexibility - Data Pipelines, Innovation Ecosystems, and Strategic Decision-making in Distribution Networks

Abstract

Electrification is increasing the pressure on electricity distribution grids, causing congestion and capacity problems, while grid infrastructure is characterized by long planning and construction lead times. Consequently, Distribution System Operators (DSOs) must make better use of the existing grid and its existing resources through flexibility, defined as ability to shift loads away from peaks and reduce congestion risks through different technologies. Thus, a DSO must increasingly rely on flexibility market mechanisms to manage capacity. A central challenge is to identify where, when, and how flexibility in the grid, and how it can be captured and deployed to solve grid related issues. One way to address this challenge is to analyze the large amounts of smart meter data generated by electricity consumption in the city, while also better understanding how the organization can utilize these resources. This thesis examines how data-driven analysis and strategic decision-making jointly shape the deployment of flexibility through an interdisciplinary approach. The study combines advanced quantitative data analysis and clustering techniques with a qualitative organizational and innovation ecosystem study. It examines how interdependent actors jointly enable flexibility in the grid, while creating a data pipeline and identifying relevant use cases and trade-offs based on interviews with key stakeholders within the DSO organization and its surrounding external actor network. The quantitative analysis develops consumption baselines and applies clustering to identify recurring demand patterns among households and small to medium-sized enterprises. The qualitative analysis examines how these analytical outputs can be interpreted, acted upon, and embedded in organizational decision-making and flexibility market development. The findings show that smart meter data can be used to show where, when, and how much flexibility exists. The study also finds that, within the regional flexibility market context, geographical clustering provides limited additional value, allowing analytical efforts to focus instead on clustering customers based on similarities in consumption behavior. Furthermore, the results highlight that clustering outputs must remain operationally actionable, where fewer, clearer, and more explainable customer groups are more useful for decision-making than highly granular segmentations. These findings imply that flexibility deployment is both a technical analytics challenge, and an organizational and ecosystem challenge. By linking advanced data analytics and clustering methods to the practical needs of DSOs, aggregators, and other ecosystem actors, the study shows how smart meter data can support more resource-efficient, lower-risk, and strategically informed flexibility deployment in modern electricity distribution systems.