Compound effects of traffic and climate on electric vehicle HVAC energy consumption: a spatiotemporal framework with city-level attribution

Electric vehicle (EV) energy consumption in real-world driving departs significantly from rated values due to the joint influence of ambient temperature, traffic congestion, and route characteristics. Existing studies typically evaluate these factors in isolation or under static conditions, leaving their compound effect on HVAC energy unquantified. This study develops a spatiotemporal simulation framework that couples traffic-aware driving speed, location- and time-specific ambient temperature, and physics-based energy submodels (cabin HVAC, traction, and battery thermal management) at the segment level, and decomposes HVAC energy variability into its temperature and trip-duration components per route. The framework is applied through a structured comparative design across seven UK cities and eight radial routes, varying departure time, regional climate, and road-network type. While total energy consumption varies by 14% across cities, the HVAC component varies by up to 89%, identifying cabin thermal management as the primary differentiator between routes in winter. The decomposition attributes 79–85% of London’s above-average HVAC energy to its extended trip duration rather than to temperature, the range reflecting how the temperature–time interaction is allocated; a same-route counterfactual shows that departure-time congestion accounts for about half of this excess duration and the network’s typical operating speed for the remainder. These attributions are robust to temperature, parameter, and correlated stochastic uncertainties, and the inter-city spread collapses in summer as heating demand vanishes. The decomposition yields a closed-form model estimating HVAC energy from three inputs (ambient temperature, average speed, and trip distance) with physical interpretability and negligible computational cost. Demonstrated for a heat-pump vehicle under UK winter conditions, the framework transfers to other vehicles and regions by re-fitting three coefficients, although the dominance of trip duration is specific to the narrow temperature range and high congestion studied. These findings indicate that the energy intensity of real-world EV operation, and hence achievable range, is substantially shaped by traffic and road-network characteristics.