To enhance the picking efficiency in modern warehousing systems, this paper is based on the improved layout design of the Chevron warehousing layout. According to its characteristics, a calculation model for the walking distance of picking paths is constructed. Genetic algorithm (GA), cuckoo search (CS) algorithm, and ant colony algorithm (ACO) are utilized as optimization approaches to improve the path optimization problem in order picking processes. These algorithms are simulated and compared with two heuristic path strategies, namely {S-shape} and {Return}. The experimental results demonstrate that when the number of locations to be accessed in order picking is 10, all three algorithms can optimize both {S-shape} and {Return} strategies. As the number of locations increases to 20 and 40, the optimization of {S-shape} strategy gradually becomes ineffective for all three algorithms. For the {Return} strategy, when the number of locations reaches 30, the GA weakens, while the CS algorithm and ACO achieve good results. However, the former has an advantage in terms of runtime.