Transport Plan Optimization for 3PLs and Carriers: Why Choose the Atoptima Solver

Key takeaways

• Carrier margins no longer leave room for guesswork. Road haulage margins in France are estimated at 1 to 2%, while diesel hit €2.34/L on 15 September 2026, up more than 50 cents since July (Créditnews, Sept. 2026).
• One kilometre in five is driven empty. In 2025, 21.8% of European road freight vehicle-kilometres were run empty, rising to 26% in national transport (Eurostat, 2026). This is the single biggest source of savings in a transport plan.
• Carbon pricing arrives in 2028. The ETS2 carbon market will cover road fuels from 2028, adding an estimated 15 cents per litre of diesel on average (AUTF, 2026).
• A transport optimization solver cuts transport operating costs by 20 to 30% by acting on fleet size, mileage, load factor and the own-fleet vs subcontracting balance (benchmarks from Atoptima customer projects).
• For a service provider, multi-customer consolidation is the number one lever. A plan built customer by customer simply mirrors commercial silos; a plan built at network level turns the flows of several shippers into optimal routes.

Contents

1. What is a transport plan, and how does it differ from a route?
2. Why is planning so complex for a 3PL or carrier?
3. Which optimization levers matter, from strategic to operational?
4. Use cases by business line: LTL, FTL, distribution, cold chain, bulk
5. What savings to expect and which KPIs to track?
6. How to choose transport plan optimization software?
7. How to deploy a solver without disrupting operations?
8. What trends will shape 2026–2030?
9. Frequently Asked Questions

1. What is a transport plan, and how does it differ from a route?

Short answer: the transport plan defines the structure of the network (which flows go through which sites, with which resources, how often); a route is the daily execution of part of that plan by one vehicle and one driver. Optimizing routes without questioning the plan means optimizing only part of the problem.

The VRP: the mathematical foundation of route optimization

The Vehicle Routing Problem (VRP) is about serving a set of locations with a fleet at minimum cost. Its variants map directly onto transport business lines: time windows (VRPTW), pickup and delivery (the core of FTL and LTL), multi-depot, heterogeneous fleets, multi-trip, and more.

The number of possible solutions explodes with problem size: a few dozen stops are enough to rule out any form of enumeration. That is why only a mathematical optimization solver can find optimal plans within minutes.

What sets 3PLs and carriers apart from shippers

A shipper optimizes its own flows. A service provider optimizes the flows of many customers, each with different contracts, service levels and rates, on shared resources. This has three consequences:

• The goal is not just cost, but margin per customer and per lane: a cheaper plan that degrades service for a key account can destroy value.
• Capacity is hybrid: own fleet, dedicated subcontracted hauliers, spot freight, sometimes a freight exchange.
• The plan is also a sales tool: pricing a tender response means simulating a new transport plan that includes the prospect’s flows.

2. Why is planning so complex for a 3PL or carrier?

Short answer: a service provider stacks every customer’s constraints on shared resources, with a margin that leaves no room for error. Five factors compound one another: multi-customer diversity, cost pressure, hybrid owned/subcontracted capacity, driver scarcity and tighter regulation.

Multi-customer diversity

A single depot may serve a manufacturer in full pallets, an e-retailer in parcels, a retail chain in roll cages and a B2B distributor through a groupage network. Each customer sets its own time windows, handling units, handling rules, penalties and contractual service level.

The plan must honour all these commitments while pooling resources. This is exactly where manual planning hits its limits: planners think customer by customer, whereas margin is made by consolidating flows.

Margins measured in one or two points

Fuel accounts for nearly 30% of operating costs according to FNTR, the French road haulage federation, and road freight margins in France are estimated at 1 to 2% (Créditnews, Sept. 2026). Fuel surcharges exist, but they are passed on late and are open to negotiation: in the meantime, every unnecessary kilometre is funded out of cash flow.

With margins this thin, a 5% cut in operating costs is not an improvement: it can double or triple net profit.

Empty running: the biggest source of savings

In 2025, 26% of vehicle-kilometres in European national road freight were driven empty. In international transport, France is among the most exposed countries, at 29.1% (Eurostat, 2026). Cutting these empty returns requires knowing, at planning time, every compatible pickup across all customers and the freight exchange.

Hybrid capacity: own fleet, subcontracted hauliers, spot freight

Most carriers and 3PLs combine their own resources with subcontracting. The make-or-buy decision is often made day by day, on gut feeling: whatever overflows gets subcontracted. A solver can instead decide upfront which lanes to keep on the own fleet (those that keep it fully utilized) and which to subcontract, based on the real cost of each option.

Driver scarcity

Around 502,000 HGV driver positions remain unfilled in Europe, a 13% shortage rate (IRU Annual Report 2025). For a carrier, turning down a contract for lack of drivers is now more common than running out of trucks. Every poorly used driving hour is lost sales capacity.

Tighter regulation

• EU Mobility Package: drivers must return home every 3 or 4 weeks, regular weekly rest in the cab is banned, and cabotage is capped at 3 operations within 7 days, followed by a 4-day cooling-off period (CJEU, press release No 173/24).
• ETS2: a carbon cost built into the price of diesel from 2028 (AUTF, 2026).
• Access restrictions: low-emission zones and local delivery charters constrain vehicle choice and delivery hours in city centres.

A plan that ignores these rules may be mathematically good, but it is operationally unworkable.

3. Which optimization levers matter, from strategic to operational?

Short answer: the biggest savings lie where few people look for them: in the transport plan itself (assigning customers to branches, lanes and shuttles, consolidation), before route sequencing even starts. A solver covers all three planning horizons with the same constraint model, which ensures the strategic plan remains workable day to day.

Six levers that make the difference for a service provider

• Multi-customer consolidation: combine the flows of several compatible shippers in a single transport plan (handling units, temperatures, time windows, safety rules).
• Cutting empty kilometres: build loops and triangular routes that chain deliveries and pickups, rather than out-and-back trips.
• Multi-dimensional load optimization: optimize 3D vehicle loading by handling weight, volume, pallet count and loading metres at the same time, while respecting stackability rules and loading/unloading sequence.
• Cost-based make-or-buy decisions: keep on the own fleet the lanes that fill it best, and subcontract irregular or unbalanced lanes at real market cost.
• Making the most of driving hours: plan around statutory driving and rest times, duty spans and driver relays (double-manning) to maximize productive hours.
• Simulation-based pricing: price a new customer at its marginal cost by inserting it into the existing plan, rather than using an average rate card.

4. Use cases by business line: LTL, FTL, distribution, cold chain, bulk

Short answer: each transport business line has its own optimization problem with its own constraints. A transport plan optimization solver must cover all these variants natively to serve every activity of the provider.

A special case: responding to tenders

For a 3PL, optimization is not just about running operations: it is about winning contracts at the right price. Inserting a prospect’s flows into the existing plan reveals their true marginal cost, which may be well below an average rate card if they fill empty returns, or above it if they overload an already stretched area.

The provider can then bid lower where it creates synergies and decline lots that would erode its margin, with pricing it can defend in front of procurement teams.

5. What savings to expect and which KPIs to track?

Short answer: benchmarks from Atoptima customer projects put the reduction in transport operating costs at 20 to 30%. For a provider with a net margin of around 1 to 2%, even a fraction of that gain changes the economics of the business.

Where the savings come from

• Fewer vehicles for the same volume, thanks to consolidation and multi-trip routing.
• Fewer kilometres, and above all fewer empty kilometres, thanks to backhauling.
• Better load factor by weight, volume and floor space.
• Less unplanned subcontracting: outsourcing becomes a costed decision, not an overflow valve.
• More contracts served with the same headcount, because driving hours are better used.
• Lower CO2 emissions, as a direct result of lower mileage and fuel consumption.

KPIs for a service provider’s transport plan

6. How to choose transport planning optimization software?

Short answer: for a 3PL or carrier, the deciding criteria are coverage of business lines and multi-customer constraints, algorithmic performance at network scale, re-planning speed, planner autonomy and integration with the IT landscape. Functional coverage comes first: a constraint that cannot be modelled wipes out the gain.

Criterion 1: Functional coverage across business lines and customers

The solver must natively model each customer’s rules (time windows, handling units, compatibilities, service levels) and the provider’s business-line variants: groupage, FTL, distribution, cold chain. Every rule worked around by manual edits erodes part of the gain and rebuilds dependence on a few expert planners.

Criterion 2: Performance at network scale

A service provider plans thousands of orders a day across several branches. The real test is not a demo on a small dataset, but how the solver performs on your actual volumes, on a peak day, with every constraint switched on.

Criterion 3: Network optimization and consolidation

The solver must be able to challenge the scope of the decision itself: assigning customers to branches, choosing lanes, consolidating across customers. A simple route calculator that sequences pre-built routes leaves most of the savings on the table.

Criterion 4: Computation speed

Late orders, an absent driver, a breakdown, a shipper running late: the morning plan does not always survive the day. A solver’s value lies in its ability to rebuild a workable plan within minutes, while locking in what has already been committed.

Criterion 5: Cost-driven planning and make-or-buy

The tool must reason in real costs: full own-fleet cost, subcontractor rates, service penalties. This is what makes it possible to arbitrate automatically between own resources and subcontracting, and to price a quote.

Criterion 6: Planner autonomy

Teams must be able to adjust a constraint, reweight an objective or test a scenario without development work or vendor involvement. An interface designed for operations (clear route views, scenario comparison, manual edits with measured impact) is what drives adoption.

Criterion 7: Integration with TMS, WMS and AI agents

The solver plugs into the existing IT landscape through standard APIs, with no overhaul. It can also be driven by an AI agent, including the provider’s own in-house agent, which makes data preparation, configuration and scenario analysis easier while keeping control of sensitive data (rates, customer volumes).

7. How to deploy a solver without disrupting operations?

Short answer: a pilot scope (one branch, one flow type, one business line) benchmarked against historical data proves the gain before rollout, with no IT overhaul and no disruption to operations.

• Set the objective: cut empty running, absorb a new customer, reduce subcontracting, make quoting more reliable. A precise, quantified objective guides the configuration.
• Choose the pilot scope: a representative branch or customer, with available data and a willing team.
• Replay history: re-optimize several real weeks after the fact and compare with the plan actually run. This is the most honest measure of achievable savings.
• Model constraints with the planners: every unwritten rule (the customer who refuses deliveries before 7 a.m., the dock that cannot take an articulated truck, etc.) must be made explicit and built in.
• Run in parallel: the solver proposes, the planner approves or corrects; the gaps reveal missing constraints.
• Integrate with the TMS and/or WMS: automate order import and route feedback via API.
• Roll out and steer: extend branch by branch, track plan KPIs and use simulation for network decisions and tenders.

8. What trends will shape 2026–2030?

Short answer: carbon pricing, fleet electrification and AI will turn the transport plan into a constant trade-off between cost, service and emissions. Providers that can simulate these trade-offs ahead of their competitors will set market prices.

• Carbon becomes a planning parameter. With ETS2 from 2028, carbon will be built into the price of diesel and passed on through fuel indexation; beyond 2030, AUTF expects it to become a defining factor in the cost of fossil-fuelled road transport (AUTF, 2026).
• Electrification rewrites the routing rules. Range, charging times and locations, time-of-day energy prices: planning a mixed diesel/electric fleet brings new constraints that only a solver can handle properly.
• Shippers demand measured emissions. Tenders increasingly include a CO2-per-unit criterion; providers that calculate it from their actual plan stand out.
• Consolidation reaches beyond company boundaries. Carrier partnerships, capacity sharing and freight exchanges open up savings that optimization can capture, provided planning happens at network level.
• AI makes the solver even more accessible. An AI agent prepares the data, turns a business instruction into parameters and explains scenarios; the solver guarantees the plan is feasible and optimal. The two work hand in hand: AI operates upstream and downstream of the solver, but never replaces its expertise.

Conclusion

European 3PLs and carriers operate on one- to two-point margins, with volatile diesel prices, a carbon cost scheduled for 2028, a structural driver shortage and one kilometre in five driven empty. In this context, the transport plan is no longer just an operational matter: it is the primary lever for profitability and commercial competitiveness.

Not all tools are equal. What makes the difference is the ability to model a service provider’s reality without workarounds (multi-customer constraints, multiple business lines, hybrid capacity, EU regulations), to optimize at network level rather than route by route, and to integrate with the existing IT landscape.

These are the dimensions — algorithmic performance, functional coverage across business lines and network optimization — on which Atoptima supports logistics service providers and carriers in building and managing their transport plans, with RouteSolver for transport planning, FlowSolver for flow optimization, PlanSolver for scheduling and resource management, and PackSolver for load plan optimization.

Frequently Asked Questions