The integration of small-scale, distributed flexibility—such as electric vehicles (EVs)—into power system operation is increasingly pursued through market-based mechanisms, including local flexibility markets and emerging concepts of market-based redispatch. This is particularly relevant in the ongoing transition from historically cost-based redispatch, which mainly activated large centralized assets, to market-based redispatch, where decentralized small-scale flexibilities become operationally and economically relevant. In this setting, robust ex-ante flexibility estimates and continuous plausibility checks are both required. Field data alone remain sparse and biased toward early adopters, making a hybrid approach necessary that combines real observations with synthetic scenario generation, while retrospective monitoring is too slow for dynamic market behavior and therefore requires a live system to assess offers and schedules against technically plausible flexibility in near real time.

This research project addresses these gaps by using data-driven flexibility simulation for verification of plausible redispatch behavior. We develop a hybrid synthetic charging framework that integrates field data with stochastic scenario generation to estimate expected local EV flexibility across user groups, charging contexts, and temporal conditions. The model captures key drivers such as plug-in behavior, connection duration, energy demand, and mobility patterns, and translates them into context-specific flexibility envelopes.

These envelopes are used within a machine-learning-supported live system to compare offered or scheduled flexibility against physically plausible ranges under current conditions. This enables the identification of implausible bids, information asymmetries, and strategic behavior, including baseline manipulation and INC-DEC gaming. In contrast to purely statistical or market-based approaches, the framework links ex-ante flexibility estimation with operational monitoring, allowing for a more consistent assessment of flexibility provision.

The results demonstrate that incorporating physically grounded flexibility estimates significantly improves the detection of unrealistic or strategically biased market behavior. This approach provides a foundation for enhancing the robustness and effectiveness of market-based redispatch and local flexibility markets by explicitly accounting for the technical feasibility of decentralized flexibility.