Design of a distribution network using data obtained from an application programming interface

Abstract

This article studies the problem in the distribution of consumer products carried out by companies in the food sector. The lack of efficiency in the design of the distribution network, based on reliable information on limitations and transfer distances to customers, are the main factors that affect the planning of the supply chain. One example is a case of a distribution food products problem, that must be delivered through a government program of food safety in school breakfasts, whose target population are students enrolled in basic level schools in the southeast area of ​Hidalgo State. A design of a distribution network in two steps or levels is proposed that identifies the routes for the distribution of these products.

Using technological tools such as Google API for calculating transport distances, cluster techniques and software for data analysis, a distribution proposal is developed which is obtained through a model that is solved by a genetic algorithm applied to the Travelling Salesman Problem (TSP). The results of this proposal include an overview of the location of all customers, the actual distances between them, consumption and the possible routes that should be considered in the planning of this supply chain. This proposal is also useful for making decisions about other aspects such as: the correct supply of products, the selection of possible distribution centers, the design of logistics based on distribution costs and the distance traveled by the transports, generating as a result a significant economic savings in distribution that will be allow to plan a future expansion and coverage in other geographical areas.