Fleet Size Quantization: The Vehicle Count Cliff Route Optimization Reports Rarely Show in 2026

Fleet size quantization is the effect where a route optimization plan’s total cost changes in discrete steps rather than smoothly as demand, service area or delivery density changes, because the number of vehicles a solver assigns to a plan is a whole number, not a continuous value. A route optimization report will show cost per stop trending down as density improves, but the underlying vehicle count moves in integer jumps, and at the specific points where the solver’s optimal fleet size crosses from one integer to the next, cost per stop can rise even as the input conditions improve. Locus, the world’s first Decision-Intelligent, Agentic TMS, surfaces the vehicle count behind every route plan rather than only the aggregate cost, specifically so a fleet planning team can see this threshold before it produces a confusing result.

Why Fleet Size Quantization Matters: The Business Case

Vehicle routing software optimizes two things that behave very differently. Route sequencing, the order stops are visited in and the path between them, is a close to continuous optimization problem: small changes in demand produce small changes in distance and time. Fleet size, the number of vehicles committed to serve a given volume, is not continuous at all. A plan uses three vehicles or four, never three and a fraction. This distinction is well established in operations research as the fleet size and mix vehicle routing problem, and specifically its step cost function variant, which studies exactly this behavior: total network cost that jumps at specific volume thresholds rather than rising smoothly.

The size of each jump is not trivial. ATRI’s operational cost research put the average all-in cost of running a Class 8 truck at $2.336 per mile in 2025, an all-time high, with driver compensation alone accounting for roughly 44 percent of that figure. At a typical 100,000 miles a year, that is well over $200,000 in annual cost committed the moment a plan crosses into needing one more vehicle. A route optimization report that shows cost per stop as a single smooth trend line is hiding the fact that a meaningful share of that trend is actually a sequence of discrete five and six-figure commitments, not a continuous efficiency gain.

The practical consequence shows up when a planning team reads a cost-per-stop chart at face value. If cost per stop ticks up at a certain volume or density level, the intuitive read is that the plan got less efficient there. Frequently what actually happened is that volume crossed the exact threshold where the solver’s optimal fleet size stepped from N vehicles to N plus one, and the new vehicle’s fixed cost has not yet been offset by enough additional volume to bring cost per stop back down below its previous level. The plan did not get worse. The fleet got one vehicle bigger, and the chart is showing the transition cost of that step before the benefit catches up.

This is not a hypothetical edge case confined to academic papers. Any operation that periodically reviews cost per stop across service areas, shifts or seasonal periods, and asks why a specific segment looks worse than a neighboring one, is implicitly asking a question that fleet size quantization can answer if the underlying vehicle count is available, and cannot answer if it is not. Two service areas with genuinely identical routing quality can show meaningfully different cost-per-stop figures purely because one sits comfortably inside a fleet-size band and the other just crossed into a new one. Without the vehicle-count series next to the cost series, that difference reads as a performance gap between the two areas rather than what it actually is, a difference in where each area happens to sit relative to its own nearest threshold.

How Fleet Size Quantization Shows Up in Route Optimization

Step 1: The solver evaluates fleet size as part of the same optimization as sequencing

A route optimization engine does not decide fleet size first and then sequence stops. It searches over both together, because the cheapest sequence for three vehicles and the cheapest sequence for four vehicles are different plans entirely, each with its own total cost. The reported cost is the lower of whatever integer fleet sizes the solver actually evaluated.

Step 2: A demand increase can leave fleet size unchanged for a range, then jump

Within a band of demand, the existing fleet size remains the cheapest option even as utilization rises, because adding a vehicle’s fixed cost is not yet justified by the marginal volume. Cost per stop improves smoothly through this band as fixed costs spread across more volume.

Step 3: At a threshold, the optimal fleet size steps up by one vehicle

Once demand exceeds what the current fleet can serve within its time and capacity constraints, without another vehicle, the solver’s cheapest available plan requires one more vehicle. That vehicle’s full fixed cost enters the plan immediately, while its marginal volume has not yet filled its capacity.

Step 4: Cost per stop rises immediately after the step, even though volume grew

Right after the step, the new vehicle is underutilized relative to the existing fleet, so average cost per stop is temporarily higher than it was just before the threshold, despite demand being higher. This is the specific point that looks, on a chart, like the plan getting worse.

Step 5: Cost per stop resumes falling as volume fills the new vehicle’s capacity

As demand continues to grow within the new fleet size’s range, the added vehicle’s fixed cost spreads across more stops and cost per stop resumes its downward trend, until the next threshold is reached and the pattern repeats.

Step 6: Report fleet size alongside cost per stop, not cost per stop alone

The only way to distinguish a genuine efficiency regression from a fleet-size step is to see the vehicle count at each point on the chart. A cost-per-stop trend line without a fleet-size series next to it cannot be correctly interpreted at the points where it changes direction.

Fleet Size Quantization vs Continuous Cost Assumptions

A network that does not track fleet size alongside cost per stop will periodically see a service area, a shift, or a seasonal peak reported as newly inefficient, prompt a review, and find nothing wrong with the routing itself, because the actual cause was a fleet-size step that the report never surfaced as its own variable. The review consumes analyst time looking for a routing defect that does not exist, while the real explanation, an integer threshold that was crossed, sits one column away in data the dashboard never displayed.

What to Look for in Route Optimization Reporting to Catch Fleet Size Quantization

Vehicle count reported alongside every cost-per-stop figure. If a dashboard shows cost trends without the underlying fleet size at each point, a genuine fleet-size step and a genuine efficiency regression are indistinguishable from the outside.

Marginal cost-per-vehicle visibility, not just fleet-wide averages. The software should be able to show what the Nth vehicle’s utilization and cost contribution actually is, not only a blended average across the whole fleet.

Threshold alerts when a plan is near a fleet-size step. A planning team benefits from knowing when demand is close to the volume that will force the next vehicle addition, so the decision can be made deliberately rather than discovered after the fact in a cost report.

Historical fleet-size series alongside historical cost series. When reviewing trend data, both series need to be available together so a step in one explains a step in the other, rather than analysts re-deriving fleet size from cost alone.

Scenario modeling before committing to a fleet-size step. Before adding a vehicle, the software should be able to show the cost-per-stop path both with and without that addition across a plausible demand range, so the decision reflects the full step, not just the immediate volume that triggered it. A team that can see both paths side by side is in a position to decide whether to add capacity now or to wait for a few more weeks of volume to arrive first, rather than reacting to a single week’s utilization number.

Fleet Size Quantization in Action: What This Looks Like in Practice

A Fortune 50 parcel enterprise running Locus across more than 4,500 drivers and 51 sites uncovered more than $14 million in annualized operational opportunity by making fleet and capacity decisions visible inside the same planning system used for routing, rather than as a separate fleet-sizing exercise disconnected from the day’s actual route plan. Visibility into what each unit of committed capacity was actually contributing was part of what surfaced that opportunity.

A leading North American retailer consolidating six legacy systems onto one platform reached 95 percent-plus route compliance and break-even in year one, an outcome that depends on fleet decisions and route sequencing being evaluated together rather than fleet size being fixed upstream and routing optimized only within that fixed constraint.

Across Locus’s enterprise deployments broadly, the platform has delivered $320 million-plus in aggregate logistics cost savings across 1.5 billion-plus deliveries, evidence consistent with the idea that a meaningful share of routing savings comes from getting fleet-size decisions right at the threshold, not only from sequencing stops more efficiently within a fleet size that was decided separately.

Common Fleet Size Quantization Mistakes to Avoid

Reading a cost-per-stop increase as a routing failure without checking fleet size. The first diagnostic step when cost per stop rises should be checking whether the optimal fleet size changed at that point, before investigating the route sequencing itself.

Adding a vehicle reactively the moment utilization looks tight. Because the new vehicle’s fixed cost lands immediately while its marginal volume fills in gradually, adding capacity too early extends the period where cost per stop sits above where it was before the addition.

Benchmarking cost per stop across networks or regions with different fleet sizes without normalizing for it. Two networks can have identical routing quality and show different cost-per-stop figures purely because one sits mid-way through a fleet-size band and the other just crossed a threshold.

Treating fleet size as a fixed input decided outside the routing system. When fleet size is set by a separate annual budgeting process rather than evaluated jointly with the route plan, the routing engine is forced to optimize within a constraint that may already be wrong for current demand. A budget cycle that runs once a year cannot track a threshold that a network can cross mid-quarter, which is why fleet size decided upstream of routing tends to lag the point where the plan actually needs it revisited.

How Locus Approaches Fleet Size Quantization

Locus, the world’s first Decision-Intelligent, Agentic TMS, evaluates fleet size as part of the same route planning system that sequences stops, rather than treating vehicle count as a fixed upstream input. Because Locus’s vehicle allocation engine recommends fleet size and type based on shipment count, weight, volume and service-area constraints inside the same run that generates the route sequence, a planning team can see the specific point where the optimal fleet size steps up, and what that step costs, instead of discovering it only as an unexplained rise in a cost-per-stop chart. Locus has been recognized by Gartner for seven consecutive years, featured in the 2026 Hype Cycle for Supply Chain Execution and Logistics Technologies, named a Leader in TMS by QKS Group’s SPARK Matrix, and ranked #1 in Route Planning on G2’s 2026 Best Software Awards.

A Fortune 50 parcel enterprise running Locus across more than 4,500 drivers uncovered more than $14 million in annualized operational opportunity in part by making previously invisible capacity decisions visible inside the planning engine. A leading North American retailer reached 95 percent-plus route compliance after consolidating fleet and routing decisions onto one system rather than managing them separately.

In October 2025, Ingka Investments, the investment arm of Ingka Group, the world’s largest IKEA retailer, acquired Locus. Locus continues to operate independently.

Fleet size quantization is not a routing defect. It is a mathematical property of the fact that vehicles are whole units with real fixed costs, and any route optimization report that hides the vehicle count behind a single cost-per-stop line will eventually produce a data point that looks wrong but is not. Locus surfaces fleet size as its own visible variable inside the same plan that sequences routes, so a planning team can tell a genuine efficiency problem from a fleet-size threshold on sight. If your cost-per-stop trends have unexplained jumps, schedule a demo to see how Locus makes the fleet-size decision behind every route plan visible.

FAQs

What is fleet size quantization in route optimization? Fleet size quantization is the effect where a route plan’s total cost moves in discrete steps rather than a smooth curve, because the number of vehicles assigned to a plan is a whole number. Cost per stop can rise at the exact point where the optimal fleet size increases by one vehicle, even though demand is growing.

Why does cost per stop sometimes get worse as delivery density improves? This usually happens at a fleet-size threshold, where the solver’s cheapest plan now requires one additional vehicle. That vehicle’s full fixed cost enters the plan before its capacity is fully utilized, temporarily raising cost per stop until enough additional volume fills it in.

Is fleet size quantization a sign that the route optimization software is wrong? No. It is a mathematical consequence of fleet size being an integer variable, studied formally in operations research as the fleet size and mix vehicle routing problem with step cost functions. The software is not wrong, the chart is simply incomplete without the vehicle-count series next to it.

How much does adding one vehicle typically cost a fleet? Per ATRI’s 2025 operational cost data, the all-in cost of running a Class 8 truck averaged $2.336 per mile, which at a typical 100,000 miles a year is well over $200,000 in annual fixed and variable cost committed the moment a plan adds a vehicle.

How can a planning team tell a genuine efficiency regression from a fleet-size step? By reviewing vehicle count alongside cost per stop at every point on the trend line. If cost per stop rises at the same point fleet size increased, the cause is very likely the step, not a decline in routing quality.

Should a fleet add a vehicle as soon as utilization looks tight? Not necessarily. Because the new vehicle’s fixed cost lands immediately while its volume fills in gradually, adding capacity earlier than the demand actually requires extends the period where cost per stop sits above its pre-addition level without a corresponding service benefit.