Among route optimization APIs for waste transport, Kardinal dominates this sector ranking by a wide margin. The sector's most critical constraint, the limit on how many vehicles can be at a tipping site at the same time, is covered by none of the other ten solutions tested. Without it, the routes produced remain unusable in production, whatever the headline coverage figure.
Waste transport is by far the sector worst served by the route optimization API market. Not because vendors have not tried, but because this business concentrates constraints most engines simply cannot model: simultaneity constraints at tipping sites, limited intake capacity at treatment facilities, and highly specialized vehicles per waste type.
This article compares eleven route optimization APIs (Kardinal, LogisticsOS, Solvice, HERE, Google, NextBillion AI, GraphHopper, Timefold, Verso, Routific, Mapbox) on their ability to handle the specific constraints of waste transport and collection.
Key takeaways
What makes waste transport genuinely complex
Intake capacity and simultaneity at sites
This is the sector's most distinctive constraint, and the one generalist optimizers handle worst. A civic amenity site, a transfer station or a treatment plant has a physically limited intake capacity: often a single vehicle can unload at a time, with a tipping duration of 45 to 60 minutes. The same logic applies to collection points where physical access for several trucks at once is impossible: a narrow yard, a single gate, a single dock.
An optimizer that ignores this sends eight trucks to the same site first thing in the morning. The result: seven trucks waiting, the day's plan collapses, and drivers rack up unplanned overtime. Theoretically optimal routes become unusable in the field, and the plans always need manual rework.
Highly specialized vehicles
Waste transport runs vehicles that are not interchangeable: rear-loader refuse trucks, crane trucks, hook-lifts, walking-floor trailers, tankers for liquid waste, hazardous waste vehicles. Each vehicle type can only collect a specific waste stream, in suitable containers. The optimizer has to handle vehicle/waste compatibility on top of weight and volume capacity.
Roll-off container swaps
In skip and roll-off operations, the vehicle drops an empty container and removes a full one at the same site or in a defined sequence. That means tracking empty and full containers as distinct loads carried on the vehicle, a mechanic few engines model natively.
Multiple return trips to the tipping site
A collection truck does not run a single end-to-end route: it collects until full, returns to the tipping site to empty, then leaves for another sequence. Without that modeling, the optimizer produces over-long routes that ignore reloading.
Heavy-goods speed profiles and hazardous materials
Collection trucks travel on roads subject to size and weight restrictions. Some waste streams (oils, solvents, chemicals) are classified as hazardous materials and require regulated routing.
The critical constraints for a waste collection API
Three levels of requirement structure the market: the standard baseline everyone covers, the intermediate constraints that are indispensable in practice, and the advanced constraints that genuinely make the difference in the field.
Distance and fleet size
The baseline objectives of any optimization.
Covered by: All vendors except Mapbox.
Vehicle capacity
Weight and volume per vehicle.
Covered by: All vendors.
Prioritizing full containers
Emptying the containers closest to overflowing first.
Covered by: All vendors except Mapbox.
Vehicle / waste-type compatibility
Preventing incompatible streams from sharing a vehicle.
Covered by: Kardinal, Timefold, NextBillion AI, Solvice, GraphHopper, Google, HERE and LogisticsOS. Missing from Routific, Mapbox and Verso.
Heavy-goods routing with hazardous materials
Compliant routing for hazardous waste streams.
Covered by: Kardinal, Timefold, NextBillion AI, Solvice, GraphHopper, HERE and LogisticsOS. Missing from Routific, Google, Mapbox and Verso.
Predictive traffic
Essential for routes starting early in the morning in urban areas.
Covered by: Kardinal, NextBillion AI, Solvice, GraphHopper, Google, HERE and Mapbox.
Truck speed profile
Without a truck profile, heavy-goods travel times are simply wrong.
Covered by: Kardinal, NextBillion AI, GraphHopper and HERE. Partial elsewhere.
Maximum distance from depot
Capping how far a route can range from its depot.
Covered by: Kardinal only. Partial at NextBillion AI, Solvice and LogisticsOS.
Alternative treatment facilities
Dynamically choosing among several eligible unloading sites.
Covered by: Kardinal, Solvice, HERE and LogisticsOS.
Simultaneity limit at facilities
The maximum number of vehicles present at the same time at a tipping or treatment site. The most critical constraint in this sector.
Covered by: Kardinal only: none of the other ten solutions covers it.
Roll-off container swaps
Tracking empty and full containers as distinct loads on the same vehicle.
Covered by: Kardinal only.
Lexicographic optimization
Ranking objectives instead of weighting them: compliance first, distance second. Without it, every criterion needs its own weight, and the configuration quickly turns into a sprawling setup to fine-tune.
Covered by: Kardinal and HERE only.
Per-site delivery or collection cap
Capping what a site can receive or ship in a day, even across several vehicles.
Covered by: Kardinal and Solvice only.
Cost charged per facility visit
Counting gate fees once per visit rather than per drop. A billing model specific to some sites.
Covered by: Solvice only: the one constraint in this sector Kardinal does not yet cover.
Flat-rate thresholds and blended billing
Modeling subcontracting thresholds and optimizing the blended own-fleet / subcontractor invoice.
Covered by: Kardinal and LogisticsOS. Partial at Solvice, Google and HERE.
Route optimization APIs for waste transport, compared
Kardinal
98% coverage · 1st/11Kardinal is the only vendor in the panel that models the simultaneity limit at facilities, meaning a transfer station can only take one truck at a time for 45 minutes. That single constraint is enough to make the other solutions hard to run in production as soon as tipping capacity is constrained.
Kardinal is also alone in handling roll-off container swaps and in fully covering maximum distance from depot. Its only gap here concerns a specific billing model: cost charged per facility visit, available at Solvice alone.
LogisticsOS
74% coverage · 2nd/11LogisticsOS comes straight in second, carried by its specialty: cost modeling. It is the only solution besides Kardinal to fully cover flat-rate and overtime thresholds as well as blended invoice optimization between own fleet and subcontractors, and it joins Kardinal and Google on guaranteed minimum revenue per subcontracted resource.
Solvice and HERE
73% coverage · 3rd/11Solvice and HERE finish level with different profiles. Solvice is the only solution in the panel, Kardinal included, to cover cost charged per facility visit, and it joins Kardinal on the per-site collection cap. Its limit: no lexicographic optimization, so every criterion needs its own weight and the configuration quickly becomes a heavy machine to fine-tune. HERE is the only solution besides Kardinal to offer lexicographic optimization, and covers hazardous-materials routing and truck profiles well.
Neither covers the simultaneity limit at facilities, which remains the decisive constraint in this sector.
Google covers the fundamentals and predictive traffic well, but offers neither a complete truck speed profile nor hazardous-materials routing, and remains, like the entire market outside Kardinal, blind to the simultaneity limit at facilities. It joins Kardinal and LogisticsOS on guaranteed minimum revenue per subcontracted resource.
NextBillion AI
65% coverage · 6th/11NextBillion AI covers hazardous-materials routing, truck profiles and predictive traffic well. Like most of the market, it models neither simultaneity at facilities nor roll-off container swaps.
GraphHopper, Timefold, Verso, Routific and Mapbox
49% → 26%- GraphHopper (49%) and Timefold (48%) cover hazardous-materials routing and, for GraphHopper, truck profiles, but miss the advanced cost and constrained-site constraints. GraphHopper remains suited to small volumes.
- Verso (40%), Routific (35%) and Mapbox (26%) lag across the whole scope.
None of these five models the simultaneity limit at facilities. Using them for collection with constrained tipping sites means producing routes that will have to be reworked manually from end to end.
Summary table
| Solution | Coverage | Site simultaneity | Roll-off swaps | Hazardous materials | Truck profile |
|---|---|---|---|---|---|
| Kardinal | 98% | Full | Full | Full | Full |
| LogisticsOS | 74% | No | No | Full | Partial |
| Solvice | 73% | No | No | Full | Partial |
| HERE | 73% | No | No | Full | Full |
| 66% | No | No | No | Partial | |
| NextBillion AI | 65% | No | No | Full | Full |
| GraphHopper | 49% | No | No | Full | Full |
| Timefold | 48% | No | No | Full | Partial |
| Verso | 40% | No | No | No | Partial |
| Routific | 35% | No | No | No | Partial |
| Mapbox | 26% | No | No | No | Partial |
Our verdict
Fleet size is not what should drive the choice. What matters is how much of your operation goes beyond the classic constraints. Start from the five leading solutions on waste, then narrow the list according to your needs.
That is exactly the use case Kardinal was built for. Test the optimization on your own data, with no commitment.
FAQ
Which route optimization API should you choose for waste collection?
Kardinal is the only solution covering the critical constraints of waste collection, with 98% sector coverage, and the most competitive on price among the top 5. LogisticsOS (74%), Solvice and HERE (73%, tied) follow, ahead of Google (66%). For hazardous waste, the shortlist narrows to Kardinal, LogisticsOS, Solvice and HERE; with several depots on top, to Kardinal, LogisticsOS and Solvice. As soon as tipping sites have limited intake capacity, only Kardinal models simultaneity.
What is a simultaneity constraint in waste transport?
A simultaneity constraint caps how many vehicles can be in the same place at the same time. In waste transport it applies mainly to tipping sites: if a transfer station can only take one truck at a time for a 45-minute unload, the optimizer has to stagger arrivals accordingly. Of the 11 solutions tested, only Kardinal models it.
How do you optimize routes with roll-off container swaps?
Skip and roll-off operations, where the vehicle drops an empty container and picks up a full one, require tracking containers as distinct loads carried on the vehicle. Kardinal is the only solution in the panel that models this natively.
Do route optimization APIs handle hazardous waste routing?
Hazardous waste transport (oils, solvents, chemicals) is subject to specific routing restrictions. Kardinal, Timefold, NextBillion AI, Solvice, GraphHopper, HERE and LogisticsOS support heavy-goods routing with hazardous materials. Google, Routific, Mapbox and Verso do not.
Can an API optimize multi-stream separate collection routes?
Multi-stream separate collection involves different vehicles per waste type and strict vehicle/collection compatibility rules. That compatibility is covered by 8 of the 11 solutions tested, including Kardinal, Solvice, Google and HERE. Kardinal remains the only one to add simultaneity management at drop-off points.
Why don't generalist optimizers work for waste transport?
Generalist optimizers are built to maximize stops while minimizing distance. They ignore the physical constraints specific to waste: site intake capacity, simultaneity, vehicle/waste compatibility. The result: routes that are mathematically optimal but operationally unusable, which explains the failure of many optimization projects in this sector.
Further reading
Parcel delivery
Stop density, tight time windows and daily disruptions: the constraints that separate the APIs on the last mile.
Read the full analysis →
Heavy and bulky transport
Vehicle dimensions, restricted access and multi-dimensional capacity: the most discriminating sector in the benchmark.
Read the full analysis →
Long-haul / FTL
Regulation, driving hours and rate modeling on full truckloads.
Read the full analysis →