Mixed EV-ICE Fleet Cost-Per-Mile: Why Utilization Decides EV Payback in 2026

The Mixed-Fleet Cost-Per-Mile Question CFOs Are Framing Wrong

For a CFO evaluating a mixed EV-ICE fleet, the deceptively simple question is: what does each mile actually cost? Most total-cost-of-ownership models answer it as a vehicle problem, comparing purchase price, energy, and maintenance per vehicle class, and producing a cost-per-mile figure that looks fixed once the fleet is bought. That framing hides the variable that actually decides the answer.

Cost-per-mile in a mixed fleet is a utilization problem. An electric vehicle and the internal-combustion vehicle it replaces have very different cost structures, and the electric one only delivers a lower cost-per-mile if it is used enough to amortize its higher fixed cost. The same vehicle can be the cheaper or the more expensive option depending entirely on how many productive miles it runs each day, which routes it is assigned to, and how much of its day is lost to charging. None of that is set by the purchase decision. It is set by dispatch and utilization.

This piece narrows the mixed-fleet TCO question to the lever a finance leader can actually move after the vehicles are on the road: utilization. It lays out why cost-per-mile is utilization-driven, defines the EV utilization threshold that determines payback, and shows how predictive fleet analytics changes the payback math without touching a single vehicle-cost input. The dollar values belong to your own fleet; the framework is what turns them into a defensible decision.

The ICCT’s commercial-vehicle TCO analysis finds that “an uptick in annual mileage can move the breakeven point substantially,” and shows a battery-electric van’s cost per mile only reaches diesel parity at sufficient utilization (equal at ~$0.73/mile in its California case).

Why Cost-Per-Mile Is a Utilization Problem, Not a Vehicle Problem

Every vehicle’s cost-per-mile is the sum of two things: its fixed costs spread across the miles it runs, plus its variable cost for each mile. Fixed costs, the vehicle’s capital cost, financing, and any dedicated charging or fueling infrastructure, do not change with distance. Variable costs, energy and maintenance, scale with it.

This is where EVs and ICE vehicles diverge. An electric vehicle typically carries higher fixed costs, from acquisition and charging infrastructure, and lower variable costs, from cheaper energy per mile and reduced maintenance. An internal-combustion vehicle is the reverse: lower fixed cost, higher variable cost. The exact figures belong to each fleet’s own quotes and energy contracts, but the structure holds across them.

Also Read: 10 Tips for AI Fleet Management and Utilization in 2026

The consequence is decisive for a CFO. Because the EV carries more of its cost as fixed, its cost-per-mile falls faster as utilization rises, since a larger fixed base is being spread over more miles. At low utilization, the EV’s fixed cost dominates and its cost-per-mile can exceed the ICE vehicle it replaced. At high utilization, the EV’s lower variable cost wins and its cost-per-mile drops below ICE. The vehicle did not change. Its utilization did.

This is why a mixed-fleet TCO model that treats utilization as a fixed assumption produces a misleading answer. Utilization is not an assumption. It is the output of thousands of daily dispatch and assignment decisions, and it is the single largest swing factor in whether the EV investment pays back.

Consumer Reports’ analysis of real-world owner data finds EV maintenance and repair costs run about 50% lower than ICE (roughly $0.031 vs $0.061 per mile over a vehicle’s lifetime).

The Five Utilization Levers That Move Mixed-Fleet Cost-Per-Mile

1. Utilization Rate: The Master Variable

Utilization, the productive miles a vehicle runs per day relative to its capacity, is the master variable in mixed-fleet economics. Every other lever works by protecting or raising it. Because the EV’s cost-per-mile is more utilization-sensitive than the ICE vehicle’s, the finance model should treat daily utilization as a decision variable, not a planning constant. A fleet that books EVs at low or uneven utilization is paying premium fixed costs to run them like economy vehicles, and its TCO model will never reconcile with reality. The first question in any mixed-fleet business case is not which vehicles to buy but how high and how consistent their utilization will be once deployed.

2. The EV Utilization Threshold

Every EV in the fleet has a utilization threshold: the daily productive mileage at which its cost-per-mile crosses below that of the ICE vehicle it would replace. Below the threshold, the EV is the more expensive choice; above it, the cheaper one. The threshold rises with the EV’s fixed-cost premium and with anything that steals productive time, and it falls as energy and maintenance savings accumulate over more miles. Each fleet must compute its own threshold from its own inputs, but the strategic point is universal: payback is not a date on a depreciation schedule, it is a utilization line each vehicle must stay above. Managing a mixed fleet to payback means managing each EV to clear and hold its threshold.

McKinsey’s mobility research documents the higher upfront capital of EVs driven by battery pack cost, the fixed-cost premium that makes EV cost-per-mile so utilization-sensitive (and which it expects to keep falling through 2030).

3. Charging Downtime: The Utilization Tax

Charging is where EV utilization quietly leaks. Every hour a vehicle spends charging is an hour it is not running productive miles, which lowers utilization and pushes the payback threshold higher. Treated as an afterthought, charging becomes a tax on the exact variable the EV’s economics depend on. Treated as a scheduling problem, it becomes manageable: charging planned into low-demand windows, overnight depot time, or natural gaps between duty cycles removes it from productive hours. The difference between charging that competes with revenue miles and charging that fills idle time can be the difference between an EV that clears its threshold and one that never does.

Also Read: How Fleet Utilization Impacts Last-Mile Delivery Costs: Five Mechanisms 2026

4. Vehicle-to-Route Matching

A mixed fleet’s advantage is optionality: the ability to send each mile to the vehicle that can serve it most cheaply. That advantage is only realized if EVs are matched to the routes where they win, in-range, high-density, predictable duty cycles that keep them utilized, while ICE vehicles absorb the long, irregular, or range-exceeding tail. Mismatching destroys value on both sides: an EV assigned beyond its range fails the route or wastes time charging mid-shift, and an ICE vehicle running a short, dense urban loop burns high variable cost on miles an EV would have served cheaply. Constraint-aware assignment, accounting for range, payload, charging access, and duty cycle, is what converts a mixed fleet from two separate fleets into one optimized system.

5. Dynamic Re-Matching as Demand Shifts

Utilization is not static. Demand moves by day, by season, and by service area, and a vehicle-to-route match that was optimal in one month may strand EVs at low utilization in the next. A fleet assigned once and left alone drifts away from its payback thresholds as conditions change. A fleet re-optimized continuously, reassigning vehicles to routes as demand and constraints move, keeps EVs loaded on the routes where they clear their thresholds across changing conditions. In mixed-fleet economics, holding utilization steady through demand volatility is worth as much as raising it, because payback depends on staying above the threshold consistently, not occasionally.

BNEF projects electric vans will reach about one-third (34%) of global segment sales by 2030, with medium- and heavy-duty electric trucks around 17%, driven largely by corporate fleet electrification.

How Predictive Fleet Analytics Changes the Payback Math

Each of the five levers is a dispatch and analytics capability, not a procurement one, which is why the payback math changes most after the purchase decision, in daily operation.

This is the layer Locus operates on. As the world’s first agentic Transportation Management System, Locus runs fleet assignment through a Capacity agent and a Dispatch agent inside a continuous Sense-Decide-Execute-Learn loop, optimizing against more than 250 real-world constraints including vehicle range, payload, charging access, and duty cycle. In a mixed fleet, that means EVs are matched to the in-range, high-density routes where they clear their thresholds, charging is scheduled into low-demand windows rather than revenue hours, and assignments are re-optimized continuously as demand shifts, so utilization stays high and consistent on the vehicles whose economics depend on it.

Also Read: Fleet Management and Utilization: AI Architecture Framework 2026

The effect on the payback math is direct: utilization rises and stabilizes without changing a single vehicle-cost input. Across 1.5B+ deliveries for 360+ enterprise customers in 30+ countries, at 99.99% uptime, this is the same optimization discipline that, in one anonymized deployment, lifted a Fortune 50 enterprise’s delivery execution rate from 75% to 92% across a 4,500+ driver operation. Higher, steadier utilization is exactly what pulls each EV above its threshold sooner and holds it there.

Analyst Validation

Locus’s optimization capabilities carry independent recognition: the G2 #1 position for Route Planning software, inclusion in the 2026 Gartner Hype Cycle across AI-powered logistics categories, a Leader designation in the QKS SPARK Matrix for Transportation Management Systems, and seven consecutive years of Gartner recognition across multiple research categories.

What This Means for a CFO or VP Finance

The mixed-fleet TCO question is real, but the version that produces a defensible answer is narrower than the vehicle comparison it is usually framed as. The vehicle inputs, acquisition, energy, maintenance, and residuals, belong in the model, but they are not the swing factor. Utilization is.

For a finance leader, three practical conclusions follow. First, model utilization as a variable, not a constant, and stress-test the payback under low and uneven utilization, because that is where EV economics break. Second, treat the EV utilization threshold as the operative payback metric, more actionable than a blended fleet cost-per-mile, because it tells you exactly which vehicles are earning their premium. Third, recognize that the largest lever on payback sits in operations, in dispatch and utilization decisions, not in the purchase order. The mixed fleet you approved pays back on how well it is run, and that is a lever finance can hold operations accountable for.

Learn more, visit locus.sh