JuMPModelPlotting.jl

Plotting functions for linear JuMP models with exactly two variables.

The main function is plot_model(model), which takes a linear JuMP model with exactly two variables and returns a graphical representation of the model. This graphical representation consists of:

• The LP feasible region, using the Polyhedra package
• The affine constraints $Ax=b$
• The integer feasible region, if one or both of the variables are integer/binary
• The objective level curve corresponding to the optimal solution, if the model has been solved and has a solution available

The implementation of the package has been split into functions corresponding to the four bullet points above, in order to make the code somewhat modular. The plot_model(model) -function calls these four functions, resulting in the final plot.

Example:

Given a model

$$ \begin{align*} \max \ & x_1 + 2x_2 \\ \text{s.t. } & 6x_1 + 4x_2 && \leq 24 \\ & x_1 + 2x_2 && \leq 6 \\ & -x_1 + x_2 && \leq 1 \\ & x_2 && \leq 2 \\ & x_1 \geq 0 \\ & x_2 \text{ integer}, \end{align*} $$

the resulting plot is