Proceedings of the
The Nineteenth International Conference on Computational Intelligence and Security (CIS 2023)
December 1 – 4, 2023, Haikou, China
Constrained Single-Objective Optimization Model for Vegetable Automatic Pricing and Replenishment
1School of Management, Guangdong University of Technology, China.
2School of Mathematics and applied mathematics, Guangdong University of Technology, China.
ABSTRACT
Due to the short shelf life of vegetables and limited shelf space in supermarkets, the need for well-designed replenishment and pricing strategies is imperative to maximize the sales revenue of vegetables in retail settings. However, existing research on dynamic pricing and replenishment strategies often focuses on single factors, such as vegetable spoilage rates and seasonality. Consequently, this paper integrates these factors into a comprehensive model. By utilizing wholesale prices of vegetables from the training dataset as independent variables, we derive fitted equations for the total sales quantity of various vegetables. Subsequently, we deduce predictive equations for the total sales quantity of vegetables by incorporating the fitted equations and discount sales data for vegetables. Finally, we enhance the model through fitness-based optimization using genetic algorithms, rendering it more interpretable and persuasive. Experimental comparisons are conducted using a well-segmented testing dataset before and after model enhancement. The results demonstrate that the model improved through genetic algorithms is less prone to local optima and provides more realistic outcomes compared to the original model.
Keywords: Goal programming, Improved genetic algorithm, Automatic pricing.
Download PDF
1School of Management, Guangdong University of Technology, China.
2School of Mathematics and applied mathematics, Guangdong University of Technology, China.
ABSTRACT
Due to the short shelf life of vegetables and limited shelf space in supermarkets, the need for well-designed replenishment and pricing strategies is imperative to maximize the sales revenue of vegetables in retail settings. However, existing research on dynamic pricing and replenishment strategies often focuses on single factors, such as vegetable spoilage rates and seasonality. Consequently, this paper integrates these factors into a comprehensive model. By utilizing wholesale prices of vegetables from the training dataset as independent variables, we derive fitted equations for the total sales quantity of various vegetables. Subsequently, we deduce predictive equations for the total sales quantity of vegetables by incorporating the fitted equations and discount sales data for vegetables. Finally, we enhance the model through fitness-based optimization using genetic algorithms, rendering it more interpretable and persuasive. Experimental comparisons are conducted using a well-segmented testing dataset before and after model enhancement. The results demonstrate that the model improved through genetic algorithms is less prone to local optima and provides more realistic outcomes compared to the original model.
Keywords: Goal programming, Improved genetic algorithm, Automatic pricing.
Download PDF