基于拉丁超方抽样的全渠道供应链网络优化

全渠道零售模式下需求的大样本、差异性、相关性特征将改变传统供应链的网络结构,文章构建离散拉丁超方抽样的免疫遗传算法进行全渠道供应链网络优化.先通过离散拉丁超方抽样对需求数据进行处理,保留样本特征的基础上减少样本容量,然后通过免疫遗传算法进行供应链网络多目标多机制优化,在保证较快收敛的同时避免局部极值,以此解决全渠道零售模式下供应链网络优化问题.实例证明该模型与算法的有效性以及在实践中的可行性.

The optimization of omni-channel supply chain based on Latin hypersquare sampling

The demand data in omni-channel retail is big, different, and relevent. It will change the network structure of the traditional supply chain. In this paper, a discrete Latin hypercube sampling immune genetic algorithm is constructed to optimize the omni-channel supply chain network. Through the data processing by discrete Latin hypercube sampling, we can reduce the sample size while preserving the data characteristics. On this basis, we adopt the immune genetic algorithm to solve the multi-objective and multi-mechanism supply chain network optimization problem. This method will ensure fast convergence and avoid local extreme. The example indicates the effectiveness of the algorithm, and its feasibility in practice.