An Improved Artificial Intelligence Based on Gray Wolf Optimization and Cultural Algorithm to Predict Demand for Dairy Products: A Case Study

Alireza Goli,Ahmad Sadeghieh,Hassan Khademi Zare,Reza Tavakkoli-Moghaddam

doi:10.9781/ijimai.2019.03.003

Abstract

This paper provides an integrated framework based on statistical tests, time series neural network and improved multi-layer perceptron neural network (MLP) with novel meta-heuristic algorithms in order to obtain best prediction of dairy product demand (DPD) in Iran. At first, a series of economic and social indicators that seemed to be effective in the demand for dairy products is identified. Then, the ineffective indices are eliminated by using Pearson correlation coefficient, and statistically significant variables are determined. Then, MLP is improved with the help of novel meta-heuristic algorithms such as gray wolf optimization and cultural algorithm. The designed hybrid method is used to predict the DPD in Iran by using data from 2013 to 2017. The results show that the MLP offers 71.9% of the coefficient of determination, which is better compared to the other two methods if no improvement is achieved.

Highlights

Demand prediction includes determining the volume of sales or market trends for a particular product and determining its sales in the entire market
It is necessary to analyze the set of indexes that could affect the demand for dairy products
If the Pearson correlation coefficient of X and Y is close to 1 or -1, respectively, it means that the relationship between X and Y is significant, and as close as 0, it means the insignificance of the relationship between these two variables

Summary

Introduction

Demand prediction includes determining the volume of sales or market trends for a particular product and determining its sales in the entire market. By forecasting demand in a company, it is possible to determine the company's market share and the required planning for performance improvement. One of the earliest stages of budget planning is demand prediction. If sales forecasting is not highly accurate, it will affect other monetary and financial variables [1]. Since prediction is one of the key tools in manufacturing system planning, various researchers have well attempted to develop this area. Demand prediction is focused on behavioral patterns of demand in the past. In these methods, the use of historical data is important [2]

Methods

Results

Conclusion