A genetic algorithm implementation for the multidimensional knapsack problem. The multi-constraint (or multidimensional) knapsack problem is a generalization of the 0/1 knapsack problem. The multi-constraint knapsack problem has m constraints and one objective function to be maximized while all the m constraints are satisfied.
The implementation is similar to the one described in Chu and Beasley, but it's significantly different. It uses Lagrangian multipliers as constraint weights and compared to the paper, it finds close to optimum solutions much faster. (Convergence can be controlled using the parameters).