Cluster Analysis for Two HoReCa Channels of a Large Grocery Retail Company

This work was developed within the scope of industrial mathematics, in partnership with a large grocery retail company in the Portuguese market, Jerónimo Martins, in particular with Recheio, the largest chain of Cash &Carry stores in Portugal. The objective was to segment the Families (groups of products in the marketing structure) in Recheio’s HoReCa channel, into disjoint sets that can be interpreted by the business. Initially, database manipulation, preprocessing and choice of variables to use were carried out. Then followed the segmentation phase carried out in two ways: through the total HoReCa channel and through the online HoReCa channel. A clear separation of the company’s products was achieved using various Cluster Analysis methods such as K-Means, K-Medoids and Hierarchical. Different groups were obtained, based on common characteristics of the Families, with the aim of better understanding the company’s market, helping to make strategic decisions and develop more effective marketing and production plans. The work involved the use of several types of software, namely Excel, SQL, Python and R.