Zahri-Edu University

Equation Solver › Optimisation

solver.zahri-edu.net

Linear programming

max cᵀx subject to Ax ≤ b, x ≥ 0

Simplex method for linear programs with ≤, ≥ and = constraints; detects infeasible and unbounded problems.

Problem size
Objective and constraints

Objective: c₁x₁ + … + cₙxₙ. Each constraint: coefficients, a relation, and the right-hand side. All variables are ≥ 0.

Examples:

Press Solve.

Method

Constraints are made to have non-negative right-hand sides and turned into equalities with slack and surplus variables. Artificial variables, penalised by a large constant (Big-M), supply the starting basis, and the simplex method pivots with Bland’s rule until no reduced cost is positive. A positive artificial variable at the end means the problem is infeasible; an entering column with no positive entry means it is unbounded. For two variables the feasible lines and the optimal objective line are drawn.

Computed in your browser with double-precision arithmetic; nothing is sent to a server. Check results by substituting them back.

Other solvers