kint

How it Works

From Data to Decision

Five steps. Your team stays in control at every stage. kint handles the math.

01

Your Data

Bring what you have.

Databases, APIs, CSV, JSON, Excel, or a plain text description. kint connects to whatever your team already uses. No special format needed.

New Problem
Type
Objective

Minimize total cost

Constraints
47 trucks available
6-hour delivery windows
EU driver regulations
Vehicle capacity: 12 tons
Variables

2,847

deliveries

143

time windows

12

vehicle types

Complexity
NP-hard
Objective

Minimize idle time

Constraints
28 machines available
Shift patterns (3 shifts)
Setup times between jobs
Due date priorities
Variables

340

jobs

28

machines

96

time slots

Complexity
NP-hard
Objective

Maximize risk-adjusted return

Constraints
Max 5% per asset
Sector diversification
ESG score ≥ 7.0
Liquidity requirements
Variables

200

assets

48

constraints

12

risk factors

Complexity
QP
02

Your Model

kint learns from your data.

kint identifies the structure: what are the variables, what are the constraints, what's the goal. Your team reviews and adjusts. The model reflects your business, not a generic template.

How would you like to start?
or describe it

“I need to schedule 340 jobs across 28 machines minimizing idle time...”

routes.xlsxuploaded
vehicles.csvuploaded
timewindows.jsonuploaded
postgresql://prod-db:5432/logisticsconnected
tables: routes, vehicles, depotssynced
2,847 rows loadedready
POST /api/v1/optimizeconfigured
payload: 2.4 MBvalidated
schema: 12 fields matchedready
03

Simulate

Test ideas before committing.

Change a parameter. Add a constraint. Remove one. See what happens to the outcome in seconds. Your team explores options without risk. Every scenario is calculated, not estimated.

Problem Analysis

“minimize cost for 47 trucks across 6 countries with time windows”

Parsing
Objective

min total cost

Variables

x_ij routes, t_k times

Constraints

capacity, time, EU regs

Data

2,847 pts, 47 trucks

Model typeMixed-Integer Program
Size12,483 variables · 847 constraints
Data Quality Report98.2% Score
Completeness98.2%
ConsistencyNo conflicts
Duplicates3 found, removed
Missing values47 / 2,847 (1.6%)
Outliers2 flagged

Data ready for optimization

Compatibility Check5/5 Passed
Problem typeVehicle Routing (CVRPTW)
SolverGurobi MIP
Data coverageAll required fields present
Constraint typesLinear + integer
Estimated solve time< 3 minutes

Problem fully compatible with kint optimization engine

04

Optimize

kint finds the best answer.

Mathematical solvers search for the optimal solution. The optimality gap tells you exactly how close to perfect the answer is. 0.00% means there is no better option. Mathematically proven.

Solving
Running
Objective Valueoptimal€18K€14K€10K
Iteration184 / ~200
Best found€9,847
Bound€9,812
Gap0.36%
Time1m 42s
LP relaxation
Branch & bound
Cutting planes...
05

Decide

Your team makes the call.

kint presents the optimal solution with full transparency. Your team sees why it's optimal, what the tradeoffs are, and what would change if conditions shift. Then they decide.

Optimal Solution
Verified

€9,847

total cost

-21%

38

routes

-19%

96%

on-time

+14%

Why this is optimal
All 2,847 points served
No capacity constraint violated
EU driver hours respected
Mathematically proven: gap 0.00%
47
Before
38
After
Routes4738-19%
Distance12,400 km9,100 km-27%
Cost€12,480€9,847-21%
RouteStopsCostStatus
R-0112€247✓ optimal
R-028€189✓ optimal
R-0315€312✓ optimal
R-0411€203~ near
Total: 4 routes€951 total cost

Integration

Works With Your Existing Systems

Standalone for your team. API for your workflows. OEM for your product.

Your Software
Sends problem & data
kint
Optimizes
Optimal Result
Verified & explained
Your Software
Receives solution

The math is ours. The decisions are yours.

Book a Demo