SDG&E and SCE use bottoms‑up approaches for medium‑ and heavy‑duty EV forecasts; SDG&E finds higher energy due to vehicle mileages

SDG&E and SCE presented bottoms‑up methods for forecasting medium‑ and heavy‑duty (MDHD) electrification at the May 22 workshop, telling stakeholders these forecasts aim to capture facility‑level charging requirements that are not well represented by system‑level IPR scenarios.

Why it matters: MDHD charging can be highly concentrated and may require significant distribution upgrades at depots, ports and freight corridors. Bottoms‑up, facility‑level forecasts can identify localized capacity needs that a system‑level forecast might understate.

SDG&E (Anderson Boles, Clean Transportation Analytics) described a three‑pillar methodology: (1) locate facilities using Department of Transportation (RigDig) records, DMV registration aggregates and port data; (2) categorize facilities against California medium/heavy‑duty electrification mandates and estimate likely adoption timing by regulatory class; (3) compute energy by combining vehicle counts, annual vehicle miles traveled and vehicle efficiency to generate facility‑level energy forecasts that are then aggregated to circuits. Boles said SDG&E’s first iteration is a “first‑of‑its‑kind” bottoms‑up forecast that focuses on vehicles domiciled locally. He told the workshop SDG&E found its resulting energy forecast was larger than the IPR MDHD energy forecast primarily because DOT facility data contained higher annual mileages than the datasets the CEC used.

"This bottoms up approach really showed us very clearly that medium and heavy duty load is not evenly distributed throughout our territory," Boles said, and he added that peak estimates are sensitive to charging behavior assumptions: compressed charging windows (i.e., many vehicles charging in a short time) materially increase peak megawatt demand.

SCE (Chris Mervarzi) described using Guidehouse’s vehicle analytics and simulation tool (VAST) to produce a circuit‑level MDHD forecast. Guidehouse identified local fleet operators and their vehicles, combined that with macroeconomic trends to forecast adoption, inferred charging infrastructure needs and used that to estimate circuit‑level loading impacts. SCE said it used the Guidehouse circuit‑level output to derive allocation factors that distribute the CEC MDHD forecast across SCE circuits. SCE noted the Guidehouse study was conducted for SCE and is not currently public, but SCE offered to discuss the methodology with stakeholders.

Backtesting and limitations: Regulators and stakeholders asked whether utilities have backtested feeder simulations against realized installations. Presenters said there has been limited backtesting to date; SDG&E and SCE both said this is an area for improvement. SDG&E also noted its current forecast focuses on fleets domiciled within its territory and does not yet include non‑domiciled corridor traffic or cross‑border charging in this iteration, though the utility expects to iterate the model in future cycles.

Practical takeaway: The MDHD bottoms‑up forecasts surface highly localized needs and tend to increase short‑term peak estimates relative to system‑level IPR energy allocations when local mileage and operational data indicate heavier use than assumed in the IPR. Utilities plan to refine methods and incorporate more observed charging and application‑level data in future cycles.