Fast genetic algorithm for pick-up path optimization in the large warehouse system

Abstract

The order-picking processes of fixed shelves in a warehouse system require high speed and efficiency. In this study, a novel fast genetic algorithm was proposed for constrained multi-objective optimization problems. The handling of constraint conditions were distributed to the initial population generation and each genetic process. Combine the constraint conditions and objectives, a new partial-order relation was introduced for comparison of individuals. This novel algorithm was used to optimize the stacker picking path in an Automated Storage/Retrieve System (AS/RS) of a large airport. The simulation results indicates that the proposed algorithm reduces the computational complexity of time and space greatly, and meets the needs of practical engineering of AS/RS optimization control.