Which municipal assets to electrify first

Most municipal electrification plans fail the same way. Someone commits to a fleet-wide target, the first orders go to whatever is due for replacement that year, and two years later there are three chargers on a 200-amp panel, a plow truck that cannot finish a storm shift, and a shop with nobody qualified to open a battery pack.

The alternative is not to refuse. It is to screen — sorting your inventory into assets that work today, assets that work once you have done something specific, and assets that should stay on liquid fuel until the market catches up. That last category is real, and saying so out loud is what makes the first two credible in front of a council.

Filter 1: duty cycle

Three characteristics make a good candidate, all observable from data you already have.

Predictable daily distance — the distribution, not the average. A sedan averaging 40 miles a day but hitting 210 once a quarter is a different problem from one that never exceeds 70. Pull 12 months of odometer or telematics data and look at the 95th-percentile day.

Return to base, at a facility you control, every night. That is what lets you charge on Level 2 overnight instead of building public-grade DC fast charging. DOE makes the point about heavier work specifically: the applications with models on the market today — school and transit buses, delivery, refuse — are the ones that already have return-to-base duty cycles and short predictable routes.

Dwell time, because charging is a function of hours plugged in, not charger size. A unit parked 4pm to 6am needs very little power per port. A unit turning over between shifts with a 45-minute gap needs DC fast charging, which changes the service question from “can we add a circuit” to “do we need a new transformer”.

That last row is the one people skip. A unit idling for hours running a lift, a compressor or a scene light is spending battery on work the range rating never contemplated.

Filter 2: class economics

Total cost of ownership is class-specific, and lumping the fleet together hides that.

That ordering is not a forecast. It falls out of two things you can check: whether a model in the class exists on a contract you can actually buy from, and whether the unit runs enough annual mileage to amortise a premium. Both move year to year, so do not accept anybody’s parity date — this table’s included — as applying to your fleet.

Two local inputs decide most comparisons: your delivered electricity price including demand charges, and the annual mileage of the specific unit. There is no national per-mile energy cost worth quoting, and a demand charge triggered by one afternoon fast-charging session can undo the whole saving on a low-mileage unit. Get the tariff before you model anything.

Maintenance is steadier. DOE estimates scheduled maintenance for a light-duty BEV at 6.1 cents per mile against 10.1 for a comparable ICE vehicle. Two cautions before that goes in a memo: it is a 2021 modelling estimate rather than observed fleet data, and it covers scheduled maintenance only — no repairs, no collision, no battery events. It is also the saving you forfeit entirely if every drivetrain concern goes to a dealer, which is the point of Filter 4.

Filter 3: electrical service capacity

This is where plans actually die. The charger is cheap; the service is not. Adding chargers can require panel upgrades, a new service entrance, trenching, sometimes a new transformer, and in the worst case a utility distribution upgrade. AFDC is blunt that labour is typically the largest line item and that existing infrastructure often needs significant work.

Before you order a single vehicle: get available capacity at each facility from the utility’s service record and a load study, not the panel schedule taped inside the door. Engage your utility account representative early — they know what capacity exists, what make-ready programs they run, and the current upgrade queue. Transformer lead times have been the long pole for years, and DOE’s Office of Electricity describes distribution transformers as facing supply-chain constraints. Confirm the permitting path with your building department, and size conduit beyond what you are installing, because trenching twice is the expensive mistake.

Filter 4: shop readiness

High-voltage systems change what the shop is allowed to do, not just what it knows how to do. You need technicians with formal high-voltage safety training, insulated tools and PPE, a defined de-energisation procedure, and a decision about where damaged or suspect packs get quarantined. ASE certifies two levels — Level 1 awareness for anyone working around the vehicles, Level 2 for technicians working directly on high-voltage systems. The tests are under $50 each, which is exactly why the certificate is not the capability: each is a short exam sitting on top of hands-on training the fleet still has to pay for. Budget the training, not the test fee, in the same year as the first vehicles. The avoidable outcome is a shop that can do tires and brakes but sends every drivetrain concern to a dealer, quietly erasing the maintenance savings that justified the purchase.

Cold-weather derating

Northern fleets size range around winter, not the window sticker. FuelEconomy.gov reports EV range dropping roughly 41% in mixed driving at 20°F versus 75°F with cabin heat in use, about two thirds of it going to heating the cabin; without cabin heat the loss is closer to 12%. Note the attribution — those figures come from a 2019 AAA test of then-current vehicles, not DOE dynamometer testing, and heat-pump vehicles generally do better.

Use roughly 40% as a conservative planning derate, then derive your own: once a handful of units have been through a winter, compare observed kWh per mile in January against June on the same routes. Derate rated range, then apply the 95th-percentile day test against the derated figure — a 250-mile sedan at 40% is a 150-mile sedan for screening. Preconditioning on shore power recovers part of this, because cabin and pack heat comes off the building meter instead of the battery. Write it into the SOP; it only helps if drivers do it.

What should not be electrified yet

Being specific here buys credibility for everything else in the plan.

Storm and emergency duty is the clearest case: a plow truck runs long variable shifts in the coldest conditions of the year, with high auxiliary loads, at exactly the moment the yard may be on generator. Seasonal equipment used a few dozen hours a year cannot recover a premium through fuel savings. Assets that overnight away from base have no charging story. Specialty vocational units may have no viable product on any contract you can buy from.

Write these down as “revisit in three years, on these conditions” rather than “no”. The condition list is the useful artifact: a model exists in class, service capacity is in place at the yard, and a comparable fleet has run one through a full winter.

Start with the rightsizing question DOE asks first: does this unit need to exist at all? An underused vehicle eliminated saves more than the same vehicle electrified. The usage history you pull for replacement scoring is the same data Filter 1 runs on, so do that pass first and electrify what survives it.