From Cooperative Attribution to Dispute-Aware Settlement in Integrated Charging–Swapping V2G Stations

Fair allocation in an integrated charging–swapping station depends on both how an event defines value and how evidentiary uncertainty propagates into settlement. Existing vehicle-to-grid (V2G) studies commonly treat contribution attribution and trusted transaction recording as separate problems. We formulate an event-conditioned cooperative game in which service requirements, available capabilities, and a service prior shape the characteristic function before Shapley attribution. An evidence–confidence vector further links measured performance to attributable performance. A calculation-provenance certificate and an evidence–dependency matrix then connect attributed amounts to component-level settlement states under dispute. Under the stated budget and dependency conditions, the formulation establishes conservation and dispute locality. In reproducible synthetic cases, event conditioning produced distinct allocations across three service types and a smaller normalized core deficit than the equal-weight game in 86.4% of sampled events. A position-balanced estimator reduced Shapley RMSE by factors of 2.6–3.4 relative to random sampling and 1.6–1.9 relative to antithetic sampling, although position-stratified sampling remained more accurate. Missing dependency edges released unsupported components, whereas spurious edges froze otherwise releasable amounts. The framework therefore connects event-specific cooperative attribution with evidence-dependent settlement and identifies dependency specification as a first-class condition of dispute-aware V2G revenue allocation.