Data analysis in a "dark store" environment

Abstract

The scope of this thesis is to analyze multidimensional data from a dark store called PockeeMart and derive deep business insights in order to achieve a complete data-driven strategy. The main objective of the dissertation concerns the application of machine learning to the data with purpose to elicit association rules, segment the customers based on their shopping behavior and create a predictive scenario, which classifies the order’s placement recency of the customers. PockeeMart is active in the FMCG sector, which is very demanding and competitive. The number of issues, that need to be taken into consideration and the number of tasks that must executed, makes it really challenging. During the pandemic period dark stores became more popular, as they do not require physical presence being part of online retailing and offer various other benefits for the users.