Commodity multi-target purchase order dispatching method and system, equipment and storage medium

Abstract

The invention relates to a commodity multi-target purchase order dispatching method and system, equipment and a storage medium. According to the invention, a two-stage scheme is designed for a purchase order dispatching task, wherein each target is firstly independently optimized to obtain an optimal value of each target, weighted summation is carried out on a deviation proportion of each target and the optimal value of each target to obtain a total target, and the total target is finally optimized, so that good support can be provided for multi-target optimization. After adopting the scheme provided by the invention, compared with a traditional sorting scheme, under the condition that the purchase cost and other indexes are basically kept unchanged, a reasonable order dispatching mode obtained through calculation can greatly reduce the freight cost.

Description

technical field [0001] The invention relates to the field of e-commerce data processing, in particular to a method for dispatching orders for multi-target procurement of commodities. Background technique [0002] Purchasing order dispatch refers to assigning the procurement requirements of multiple commodities to different suppliers (hereinafter referred to as merchants) under certain constraints in a single procurement, so as to achieve the optimization of certain goals (such as cost, aging, etc.). The relationship between goods and merchants is as follows figure 1 As shown, a product can be supplied by multiple merchants, and a merchant can also supply multiple commodities, and the two form a bipartite graph. In addition, the connection between the commodity and the merchant has constraints, indicating the merchant's supply inventory for the commodity. [0003] For the above task of dispatching orders, the traditional method is to sort the suppliers of each commodity, a...

AI Technical Summary

Problems solved by technology

The traditional method is to dispatch each product independently, so it is difficult to deal with some global goals, such as freight (different products purchased from the same merchant will be packaged and sent together, and the calculation of freight often distinguishes between the first weight and the second) Heavy, and the additional weight is cheaper than the first weight, in addition, the package that is less than the first weight is charged according to the first weight, so generally, the shipping cost is lower when purchasing from fewer merchants), etc.

In addition, traditional methods are also difficult to provide good support for the needs of multi-objective optimization.

Images