2D Linear Programming Visualizer

A client-side graphical solver for two-variable linear programming problems, built with Expo / React Native Web.

Features

• Maximize or minimize z = c₁x + c₂y
• Add and remove linear constraints of the form ax + by ≤ c or ax + by ≥ c
• Draws every constraint line and the feasible region
• Enumerates feasible corner points
• Draws the objective line at the optimum
• Detects infeasible and unbounded problems

Run locally

cd frontend/app
npm install
npm run web

Or with Docker:

docker compose build
docker compose up

Test

cd frontend/app
npm install
npm test -- --runInBand
npm run typecheck

How the solver works

1. Converts all inequalities to half-planes.
2. Intersects pairs of boundary lines, including the non-negativity boundaries.
3. Keeps only feasible intersections as candidate vertices.
4. Evaluates the objective at each vertex.
5. Checks the recession cone to distinguish a bounded optimum from an unbounded objective.
6. Clips a plotting rectangle by all half-planes to display the feasible region.

License

• Apache-2.0