Volume 20 No 22 (2022)
 Download PDF
Demand and Supply Forecast using Hybridization Machine and Deep Learning
Monika Saini, Dr. Vinti Dhaka
As the level of competition among retailers in the market is growing on a daily basis, businesses are placing a greater emphasis on the use of predictive analytics methods in order to lower their expenses while simultaneously raising their levels of productivity and profit. Both having an excessive amount of inventory on hand (also known as overstock) and not having enough available (also known as stockouts) are highly important issues for merchants. A decrease in revenue may result from an excessive stock level since there is a direct correlation between stock surplus and firm capital. Depending on the kind of goods being sold, having an excessive amount of inventory may also result in higher expenses associated with warehousing, labor, and insurance, as well as a decline in product quality. Products that are out of stock may result in lost revenue, decreased customer happiness, and decreased consumer loyalty to the shop. If a client is unable to locate the goods that they are seeking for on the shelves, they may either switch to a different rival or purchase things that are a replacement. It may be challenging for merchants to maintain their customers' loyalty, particularly in the middle and low end of the market. Forecasting customer demand is one of the most important challenges facing supply networks today. Its goals were to maximize profits, decrease expenses, and improve client retention while simultaneously increasing sales and profits. To achieve this goal, historical data may be evaluated to enhance demand forecasting via the use of a variety of methodologies including machine learning techniques, time series analysis, and deep learning models. The development of an intelligent demand forecasting system is the focus of this effort. The improved model is based on the analysis and interpretation of the historical data using various forecasting methods such as time series analysis techniques, support vector regression algorithm, and deep learning models. This analysis and interpretation of the historical data is what makes up the basis of the improved model. To the best of our knowledge, this is the first study to combine the deep learning methodology, the support vector regression algorithm, and various time series analysis models by a novel decision integration strategy for demand forecasting approach. This was accomplished in order to provide an improved method for predicting future demand. Improve MSE and RMSE error with the help of the suggested method, which combines a CNN-based regression methodology with a random forest.
Retail demand, Supply demand, Deep learning
Copyright © Neuroquantology

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Articles published in the Neuroquantology are available under Creative Commons Attribution Non-Commercial No Derivatives Licence (CC BY-NC-ND 4.0). Authors retain copyright in their work and grant IJECSE right of first publication under CC BY-NC-ND 4.0. Users have the right to read, download, copy, distribute, print, search, or link to the full texts of articles in this journal, and to use them for any other lawful purpose.