talk to us

test the plan.
before the day does.

A plan is only as good as the day it meets. Simulation runs the same decision against what could happen: more demand, a failed crane, a smaller budget. Your team sees which plan holds, which limit breaks first and what each option costs.

show us one decisionhow it works
Model 04 of 4 · scenarios and stress tests
one plan, many futuresone crosses the limit
Five questions, built into every run

the questions managers ask next.

The plan answers today's question. The next ones follow straight away. kint answers them on the same model and the same data, without exports or a second tool.

What-if analysis“what if we change this one thing?”

example: What if delivery windows widen from four hours to six? Routes, vehicles and cost, before and after.

Change one input, keep everything else, solve again. The new plan sits next to the original with every difference marked.

Scenario comparison“which strategy is best for us?”

example: Three berth strategies for next week, compared on waiting time, overtime and crane hours.

Your team defines the strategies: fastest, cheapest, lowest emissions. kint optimizes each one on its own terms. The results are compared, with each trade-off in numbers.

Sensitivity analysis“where should we push first?”

example: Is it berth length or the number of cranes that holds throughput back?

kint ranks the inputs by how strongly they move the result and states what one more unit of a resource is worth. Planners know which limit to negotiate and which to leave alone.

Trade-offs between goals“cost or speed? show us both.”

example: Lowest cost first, then the fastest plan that stays within a set margin of that cost.

When goals pull apart, kint finds the set of plans where improving one goal costs another. Planners set the priorities, by weight or in strict order, and pick from the set.

Learned outcomes“what does our history say?”

example: Which yard settings moved the most boxes per hour in the past, and how sure is the model? blackbox optimization

Where no formula exists, kint learns how outcomes respond to decisions and searches the learned model. Candidates come with an uncertainty range.

Stress test

one plan. many futures.

kint fixes the plan and replays it against scenarios the team defines or takes from past disruptions. Where the plan breaks a limit, you see which limit, in which scenario, and by how much.

Then your team chooses: keep the plan, add slack where it matters, or optimize again for the hard cases.

keep

The plan holds in every case your team cares about. Release it as it is.

add slack

Reserve capacity exactly where the stress test showed the break, and nowhere else.

optimize for the hard cases

Solve again with the critical scenarios as rules, and compare the cost of safety.

In practice

how a planning team uses it.

The tools work together. A typical week combines them in this order, and every step stays in the record.

  1. Solve the base case. The plan, its bound and gap, and its binding limit.
  2. Read the sensitivity. Find the few inputs that move the result most.
  3. Ask what-if on those inputs. Put a number on each trade-off.
  4. Define two or three strategies. Compare the results.
  5. Stress test the favourite. Check it against the futures that worry you.
  6. Release and record. A person releases the plan; the record keeps who and when. When conditions change, run it again.

bring the plan that worries you.

We put it through the scenarios your team fears most and show where it holds and where it breaks, on real data.

start with one decision
Tools
What-if · scenarios · sensitivity · trade-offs · learned outcomes · stress tests
Where
On every optimization run
Record
Every scenario and every release kept with the run
Contact
hello@kint.pro