Abstract
In this thesis, we design an Ant Colony Optimization Algorithm (ACO Algorithm) for the zero-one Multidimensional Knapsack Problem. The zero-one Multidimensional Knapsack Problem is the problem of choosing some of n items such that the corresponding profit sum is maximized without violating m constraints.ACO algorithm can be discussed with three aspects: heuristic value, solution construction, and pheromone update. In chapter 1, we introduce some different ACO algorithms that have been discussed. In chapter 2, we propose 5 kinds of ACO algorithm for the zero-one Multidimensional Knapsack Problem. In chapter 3, we try to set parameters using two different methods. In chapter 4, we discuss the computational results between the algorithms we proposed in chapter 2. Compared with the past algorithm, the ACO algorithm we designed can solve the zero-one Multidimensional Knapsack Problem with less computing time, and the solution qualities are the same.