Skip to content
kint
Decision Intelligence for Logistics

kint manifest

Move more with less. Provably. The decision layer for operators of asset-heavy logistics nodes that makes expensive capacity, equipment and throughput decisions optimal instead of merely visible.

Provably optimal EU AI Act ready Made in Germany
kint engine
run 4e1a · solving
Describe the problemlanguage + data
Build the modelwhite-box / learned
Solve to optimalitysolver
Minimum fleet to clear 3 vessels < 3hcertificate
optimality gap 0.0%provably best · defensible

The decisions nobody computes

Your most expensive decisions are made on gut feel.

How much equipment to run. Which capacity to commit. Berth and yard strategy. The answer is buried in distributed data and in a simulation nobody runs by hand. So planners reach for Excel and heuristics.

Your terminal operating system shows you what is happening. It does not tell you what to do. kint manifest sits above it and answers the capex and operating questions your TOS leaves open.

The cost of deciding on gut feel
4h 10m
Average vessel turnaroundon manual planning. Every hour over plan risks demurrage and a missed berth window.
+27%
Equipment over-provisionedto feel safe. Capex that sits idle for most of the shift.
€1.8M
Yearly late-handling penaltiesat a mid-size terminal. Bottlenecks the TOS never flagged in time.
Industry-typical figures. Illustrative, not a kint measurement.

Terminal & Port Ops

One screen. Every expensive decision, already optimal.

Berth allocation, yard strategy and equipment sizing computed together, not in separate spreadsheets. Each number carries a solver certificate.

Terminal & Port Ops
OperationsYardEquipmentAudit
PORT-HMB-04 · shift 06:00–18:00
Vessel turnaround
2h 41m
▼ from 4h 10m
Rehandles
−14%
▼ 3,120 moves saved
Crane productivity
+9%
▲ 32.4 moves/hr
Truck turn time
7 min
▼ from 28 min
TEU throughput
9,840
/ shift · TEU
Yard utilization · rehandle hotspots8 × 5 blocks
A1
A2
A3
A4
A5
A6
A7
A8
B1
B2
B3
B4!
B5
B6
B7
B8
C1
C2
C3
C4
C5
C6
C7
C8
D1
D2
D3
D4
D5
D6
D7!
D8
E1
E2
E3
E4
E5
E6
E7
E8
lowhigh! rehandle hotspot
Optimizer: minimum equipment to unload 3 vessels in under 3h, given current yard state and gate load.
Recommendation
8 straddle carriers + 4 STS cranes
● optimality gap 0.0% · solver certificate
What-if: +1 crane saves 22 min but costs more than +2 straddle carriers. The carriers are the cheaper path to the same deadline.
Want this on your terminal's data? Bring one shift. We will show you the optimal plan.Book a terminal walkthrough
Works alongside your stackkint docks on top via API or OEM. No rip and replace.
Any TOS / TMSERP & WMSYour data lakeBring one shift of data

One engine, four modules

From the network down to the quay crane.

Each module answers the decisions its layer leaves open. Run one, or run them together on the same optimization brain.

network & flow design

Where capacity should live

Network design across locations and flows. Decide which nodes carry which volume and how cost-to-serve moves when demand shifts.

fleet & routing

How many vehicles, which tours

Fleet sizing and vehicle routing with time windows. CVRP, VRPTW, multi-depot. The right number of assets, not the comfortable one.

warehouse & yard

Storage and yard strategy

Slotting and yard strategy that cuts dwell time and rehandles. Put boxes where the next move is cheapest.

★ flagship module

Terminal & Port Ops

Berth allocation, quay crane split, yard strategy, equipment sizing and gate flow. Tuned with the reference customer and exclusive to their segment.

Forecasts tell you a storm is coming. kint tells you which cranes to move, and proves it was the cheapest way to clear three ships by noon.
The category gap, in one sentence. Decisions, not dashboards.

How kint decides

What is the minimum fleet size to unload three vessels in under three hours?

Same data, different question. The TOS cannot answer this. kint does, in milliseconds, with the reasoning attached.

Minimum fleet to clear 3 vessels < 3h

run id: hmb-2026-0615-04
solved in 380 ms
solver: kint engine
Recommendation
8 straddle carriers + 4 STS cranes
Binding constraint is crane availability, not carrier count. Sensitivity: the plan stays optimal up to 4,300 TEU per vessel.
−1h 29m
★ Decision humans miss
Shift MAERSK 40 min earlier
Frees crane STS-2 during the MSC peak. One fewer crane needed, same deadline met.
★ Decision humans miss
Pre-stage 2 blocks for HAPAG
Removes 410 rehandles at the gate. Truck turn time drops to 7 min during the afternoon surge.
gut-feel default: 11 straddle carriers + 5 STS cranes · turnaround 4h 10m · 2 assets over-provisioned

Why operators choose kint

Built for the people who own the P&L.

01

We do not replace your systems

kint sits above your TOS and your data. It makes the decisions they leave open. Nothing to rip out.

02

It works even with no formula

When there is no ready model, kint learns the relationship from your data and finds the best operating point in milliseconds, with confidence gating.

03

Every answer has a reason and a record

Each recommendation ships with the binding constraint, the sensitivity, and an audit trail you can present to liable management later.

The category gap

The optimization brain above the TOS.

Others execute, model by hand, or predict. None of them decide. Forecasts are not decisions.

Their approachkint manifest
Terminal operating systems
TOS / TMS
Run the terminal and show the plan. Leave the sizing open and never improve it.The optimization brain above the TOS. Decisions, not dashboards.
Hand-built OR models
consultant-built
One model per terminal, built by consultants. Months to adapt, rigid when operations change.Generates the model from description plus data. Adapted in days, not months.
Forecasting tools
arrivals & congestion
Forecast vessel arrivals and congestion, then hand the decision back to you.Decides, provably optimal, and answers the capex and equipment-sizing questions.

Powered by the kint engine

Proof, not promises.

Provably best solution
Optimality gap plus solver certificate on every run.
Fully traceable
Audit-ready defense package for liable management.
Learns when no formula exists
Models the relationship from your data, with confidence gating.
EU AI Act ready
GDPR, EU hosting, Made in Germany.

Stop sizing on gut feel.

Bring one shift of your terminal data. We will show you the optimal plan, the reason behind it, and the record you can defend.

Book a Demo