Practical Guide: Tanker and Multi-Compartment Transport Optimization

Key takeaways

• A multi-compartment truck is planned compartment by compartment. Each compartment has its own limit, and one that is full (on collection) or empty (on delivery) can force a return trip even though the truck as a whole is not.
• Product compatibility and cleaning rules shape the schedule far more than the order of stops.
• Each flow has its own dominant constraint: keeping supply chains separate for milk, avoiding run-outs for fuels and gases, cleaning and empty returns for chemicals and food liquids, cross-contamination for animal feed.
• Knowing actual levels at the customer site (milk tank gauges, tank and silo telemetry) turns fixed routes into demand-driven routes.
• These constraints are too intertwined to manage by hand. A multi-compartment solver handles them all at once, and its performance is measured in cost per unit shipped (€/1,000 L or €/tonne).

Contents

1. What is tanker and multi-compartment transport?
2. What makes planning compartmented vehicles different?
3. Multi-compartment route optimization (MCVRP)
4. Key use cases
5. Which KPIs to track for tanker transport?
6. Tanker and multi-compartment transport glossary
7. Frequently Asked Questions

1. What is tanker and multi-compartment transport?

A tanker or compartmented truck cannot be planned like a pallet truck: its capacity is split into compartments, and each compartment comes with its own rules. Six types of flow share this logic.

The common thread: assign the right products to the right compartments, chain compatible loads, then build routes that make the most of that assignment without breaking it.

2. What makes planning compartmented vehicles different?

Five constraints set these flows apart from standard parcel and pallet distribution, and they compound one another.

• Multi-dimensional capacity: capacity is not a single overall figure; each compartment has its own limit. Once the compartment assigned to a product is full, the truck cannot take any more of it, even if there is room elsewhere, which can force a return to the depot or plant before the truck is full. Planning must account for volume per compartment, gross vehicle weight and axle loads, and density: a dense product hits the weight limit before filling its compartment.
• Product-compartment compatibility: some products cannot share a compartment, or not without cleaning (organic and conventional milk, diesel and petrol, two animal feed formulas, a food product and a chemical).
• Cleaning and certification: between two incompatible products, the tank goes through a wash station, which issues a certificate (an ECD in the chemical sector). Every cleaning costs time, money and extra kilometres.
• Fixed or flexible compartments: bulkheads may be fixed (typical of fuel tankers), combinable, or absent altogether (single-compartment tanks). The more flexibility, the more room for optimization.
• Operations at each stop: pumping, gravity or pneumatic discharge, hose connection, sampling. Service times vary widely from one stop to the next.

Depending on the flow, further constraints apply: ADR rules for dangerous goods, temperature control (insulated tanks, heated bitumen or chocolate) and site time windows (plant, depot, terminal, wash station).

3. Multi-compartment route optimization (MCVRP)

The reference framework is the MCVRP (Multi-Compartment Vehicle Routing Problem): a VRP in which each vehicle has several compartments and each customer orders one or more products.

In practice, the problem is solved on two levels. First, assignment: which products go in which compartments, for which customers. Then routing: the order of stops, subject to time, access and regulatory constraints.

Constraints to configure every time:

• capacity per compartment and total weight per vehicle;
• product × compartment and product × product compatibility matrix, with the associated cleaning time and cost;
• customer and site time windows (plant, depot, terminal, wash station);
• accessibility (vehicle size, weight limits, low-emission zones, narrow roads);
• volume-dependent service times;
• drivers’ hours rules and, where relevant, ADR rules;
• intermediate unloading or reloading during a route (multi-trip).

4. Key use cases

a. Fuel and heating oil tankers

Here, product flows from the depot to filling stations, industrial customers or households (domestic heating oil). The challenge is to prevent customer run-outs while filling every compartment to the maximum.

Specific constraints

• Fixed compartments (often 4 to 6 on a semi-trailer tanker), one product per compartment, no partial discharge of a compartment without a flow meter.
• Customer tanks of different sizes for each product: the deliverable volume is capped by the ullage available on arrival.
• ADR: routes and tunnels prohibited depending on the restriction code, regulated parking, driver training and documentation.
• Loading time at the depot, with queues at the loading bays, is often the bottleneck of the day.
• Domestic heating oil: strong seasonality and price sensitivity, with order peaks that are hard to smooth out.

Optimization levers

• Vendor Managed Inventory (VMI): the distributor decides when and how much to deliver based on tank telemetry. The routing problem becomes an IRP (Inventory Routing Problem), which balances transport cost against run-out risk.
• Automatic order splitting to fill vehicles to capacity.
• Load plan building: consolidate single-product orders, combine orders from 2 or 3 stations to fill every compartment, then sequence the stops.
• Consumption forecasting per tank: historical throughput, calendar effects (weekends, holidays), weather for heating oil.
• Depot loading smoothing: allocate loading slots to reduce waiting time.
• Night deliveries where the site allows, to free up daytime slots and avoid traffic.

b. Milk collection

The challenge is to collect milk from farms at the right frequency, without mixing supply chains, and deliver to the plant within its reception slots. Transport is a major share of the cost of collected milk, especially in mountain areas or regions with scattered farms.

Specific constraints

• Collection frequency: daily, or every two or three days, depending on the bulk milk tank capacity and health rules on how long milk may be stored on the farm.
• Keeping supply chains separate: organic, conventional, PDO, GMO-free, goat or sheep milk. A premium supply chain cannot be downgraded by mixing, so pickup order matters (organic first if one compartment must carry both).
• Sampling and traceability at every farm, with pumping time depending on volume.
• Multiple plants: milk can be sent to one plant or another depending on the day’s processing needs.
• Seasonal variability: spring peak, autumn trough.

Optimization levers

• Periodic routes: plan over a 2- to 4-day cycle rather than day by day, choosing a feasible collection schedule for each farm.
• Tank level monitoring: level sensors avoid visiting a farm whose tank is half empty, or arriving too late.
• Relay trailers and transfers: a rigid truck collects in hard-to-reach areas and offloads into a parked semi-trailer, which runs the long shuttle to the plant.
• Cross-dairy cooperation: neighbouring operators swap collection areas to reduce overlapping routes.
• Seasonal re-zoning: redraw collection areas in spring and autumn rather than keeping the same routes all year round.

c. LPG and liquefied industrial gases

Bulk LPG (propane for households, farms and industry) and cryogenic industrial gases (nitrogen, oxygen, argon, CO2) are delivered in single-compartment tankers to many customers per route. This is the textbook case of the Inventory Routing Problem: the distributor manages its customers’ stock itself.

Specific constraints

• ADR Class 2, pressurized or cryogenic tanks, specially trained drivers.
• A safety stock to guarantee at every customer, with no run-outs.
• For cryogenic gases, gradual product evaporation (boil-off), which penalizes long routes and waiting time.
• Many small, scattered customers, sometimes with difficult access for residential LPG.

Optimization levers

• Telemetry gauges and consumption forecasting per customer (history, weather for heating).
• Plan as an IRP: deliver when a visit is cheap (the customer is on the way) rather than at the last minute.
• Maximize the volume delivered per stop to lower the unit cost.
• Choose the loading source dynamically (plant, depot) for each route.

d. Chemical tankers

Chemical tank transport depends less on the order of stops than on how loads are chained together: every product change may require a cleaning, and many return legs are run empty.

Specific constraints

• ADR classification for each product, with specific equipment, routes and tunnel restrictions.
• Compatibility with the previous product carried and with the tank material (stainless steel, lining, seals).
• Cleaning at an approved wash station, with a cleaning certificate (ECD, European Cleaning Document) required by the shipper.
• Shippers’ quality and safety requirements: SQAS assessments, loading and unloading protocols.

Optimization levers

• Chain compatible loads to avoid a cleaning between two jobs.
• Treat wash stations as stops on the route, with their opening hours, durations and costs.
• Look for backhaul loads rather than running empty.
• Weigh dedicated tanks against multi-product tanks: a dedicated tank eliminates cleaning but reduces flexibility.
• Shift long distances to ISO tank containers by rail or sea, with road handling first and last legs.

e. Food-grade liquid tankers

Wine, oils, juices, liquid sugars, chocolate and egg products travel in stainless steel tanks, often insulated, under strict hygiene rules. Optimization focuses on product sequencing and temperature control.

Specific constraints

• In the European Union, bulk tanks for liquid foodstuffs must in principle be reserved for food (Regulation (EC) No 852/2004).
• Cleaning between products according to a sequencing matrix: allergens, colours, odours, organic, kosher or halal certification.
• Temperature control: heated chocolate and fats, chilled juices and egg products.
• Strict reception slots at manufacturers’ sites, with quality checks on arrival.

Optimization levers

• Sequence loads from least to most restrictive product to limit full cleanings.
• Dedicate tanks to a product family on regular flows (wine, oil, liquid sugar).
• Link transport with customers’ production planning to smooth loading and delivery slots.
• Use compartments to deliver two compatible products to the same customer or to two nearby customers.

f. Dry bulk and animal feed tankers

Cement, lime, flour, sugar, plastic pellets and animal feed are delivered in pneumatic discharge tankers. Animal feed is the archetypal multi-compartment truck use case: several formulas delivered to several farms on the same route.

Specific constraints

• Cross-contamination: loading and delivery order imposed between formulas (medicated feed, organic or GMO-free supply chains).
• Very different densities across products: payload is often the limit before volume.
• Long pneumatic discharge times that vary with the product, the distance to the silo and the compressor.
• Access to farms and construction sites: narrow lanes, weight-restricted bridges, limited turning space.

Optimization levers

• Assign formulas to compartments following the anti-contamination order, then sequence the farms.
• Synchronize routing with production at the feed mill or cement plant to cut waiting time at loading.
• Group orders by area and fixed delivery day to fill every compartment.

5. Which KPIs to track for tanker transport?

The key metric is cost per unit shipped (€/1,000 L, €/m³, €/tonne), supplemented by the load factor per compartment.

Tracking these indicators by area and by day of the week reveals structurally loss-making routes, which are often hidden in an overall average.

Conclusion

Milk, fuel, LPG, chemicals, food liquids or dry bulk: tanker transport always combines constraints that cascade into one another. The choice of compartment determines which customers can be served. Those customers set the order of stops, which in turn decides the next return trip or the next cleaning.

No planner can evaluate all these combinations by hand, let alone recalculate everything after each disruption. A multi-compartment routing solver does it in minutes: it assigns products to compartments, builds the routes at the same time, and recalculates the plan as soon as any input changes. Combined with telemetry, it even makes it possible to anticipate deliveries instead of reacting to them.

For a real-world example, read our customer case on route optimization for compartmented trucks.

Tanker and multi-compartment transport glossary

Frequently Asked Questions