Research on Vegetable Sales Strategies in Supermarkets Based on MLP Optimisation Algorithm and Random Forests

Abstract

The aim of this study is to propose a comprehensive solution to the challenges in vegetable merchandising in superstores. For the sales volume pattern study, data processing and correlation analyses reveal the association between vegetable categories and explore possible merchandising strategies for merchandise mix. For the pricing problem, a cost-plus pricing method and a machine learning model are introduced to develop a price strategy for superstores to maximise sales profit, and a prediction model for future pricing is proposed. For the replenishment plan, the MLP optimisation algorithm and regression model are used to design the replenishment plan to maximise the revenue within the limited sales space. Finally, the importance of data is emphasised and multifaceted data collection and analysis methods are proposed to help superstores better understand market demand, optimise inventory and pricing strategies, and ultimately maximise revenue and customer satisfaction. In summary, this study provides a systematic set of vegetable merchandising strategies that provide useful reference and guidance for superstores to cope with market challenges.