Maiden Application of a Structural Regression Model for Consumer Goods Demand Sales Forecasting Based on Consumer Behavior
Main Article Content
Abstract
Big data analytics has become an important new factor field of study for both researchers and professionals, demonstrating the significant demand for solutions for companies' Problems in an economy based on impulse knowledge. Today, the customer's behavior when buying consumer goods has been expressly transformed. The Big Data associated with the buyer's behavior is volatile, cumbersome, and truthful. Offering customized services to its customers through the consumer goods market is a big challenge. In this paper, we are developing a structural regression model that combines classifiers and predictors to perform sales forecasting for the future demand for an FMCG (Fast Moving Consumer Goods) company’s products or services. The novelty in this paper is designing the structural machine learning model which does not exist. The objective is to predict and forecast the sales which are the fast-moving goods for future decision-making. Here we choose a few FMCG hair care products with huge data to compare services, products, and sales based on consumer behavior of different characteristics. The proposed model considers first, the coincidence of the items purchased by the customer systematically together or one after the other. Secondly, Frequency: customer purchases at certain times of the year. Third, recurrence: sequential purchase during each period. By using the regression algorithm and moving average method, we can predict data analysis on training data sets. Our results show that variables from online reviews of customer behavior supported the demand prediction of marketing strategies. This can help manufacturers enhance ways to progress competitiveness and productivity by reducing prices and improving delivery and responsiveness to customer needs. This proposed model incorporates customer information like past web activities, reviews, comments, and tweets online.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.