基于蚁群算法的多规格货箱装车模型研究

Research on multi-specification cargo-box loading model based on ant colony algorithm

为满足仓储自动化物流系统中对货箱自动装车的需求,利用启发式思维,设定摆放、稳定性等约束条件建立货物装车的模型框架,运用蚁群算法进行多目标优化,以货箱体积占用率和重心范围为目标函数,创新性地将各个货物的相似度和目的地路径作为启发因子,动态地将货箱的体积占比作为信息素的参数,自适应改变信息素的挥发程度来加快算法收敛速度,得出一种能提高货车空间利用率以及货物稳定性装载方案。并通过实例验证得出,在货箱种类和数量较多的情况下,由本算法模型得到的货箱装载方案的满载率,相较于人工,提升了6%~7%。

Based on the urgent need for automatic loading of cargo boxes in the warehouse automation logistics system,this paper uses heuristic thinking to set constraints such as placement and stability to establish a model framework for cargo loading,and uses the ant colony algorithm to perform multi-objective optimization.The volume occupancy rate of the cargo box and the range of the center of gravity are the objective functions.The similarity of each cargo and the destination path are innovatively used as the heuristic factors,and the volume ratio of the cargo box is dynamically used as the parameter of the pheromone to adaptively change the pheromone.The volatilization degree is used to speed up the convergence speed of the algorithm,and a scheme that can improve the space utilization rate of the truck and the stability of the cargo is obtained.And through example verification,when there are many types and quantities of containers,the full load rate of the container loading scheme obtained by this algorithm model is increased by 6 to 7 percentage points compared with manual work.