Ginting, One Meus and Budiarto, Eka and Mahawan, Bagus (2021) Demand Forecasting in FMCG Company Using Machine Learning and Statistical Analysis for Inventory Control Optimization. Masters thesis, Swiss German University.

Preview	Text ONE MEUS GINTING 21951009 TOC.pdf Download (1MB) | Preview
	Text ONE MEUS GINTING 21951009 1.pdf Restricted to Registered users only Download (1MB)
	Text ONE MEUS GINTING 21951009 2.pdf Restricted to Registered users only Download (1MB)
	Text ONE MEUS GINTING 21951009 3.pdf Restricted to Registered users only Download (1MB)
	Text ONE MEUS GINTING 21951009 4.pdf Restricted to Registered users only Download (2MB)
	Text ONE MEUS GINTING 21951009 5.pdf Restricted to Registered users only Download (226kB)
Preview	Text ONE MEUS GINTING 21951009 Ref.pdf Download (677kB) | Preview

Abstract

Challenges exacerbated by companies today are increasingly heavy, in particular, FMCG or Fast Moving Consumer Goods because of market shifting and consumer behavior always changing especially during a COVID 19 pandemic. Knowing how customer demand and behavior is key to success for FMCG company, make inventory cost always low, and produce customer satisfaction and loyalty to the products. Consumer demand will drive a factory to have good production planning and make a company more competitive than the others because have a good supply chain, decreases loss of sales opportunity, and increase productivity. Since ERP implementation many reports were found that the sales and marketing department complained about the often unfulfilled consumer demand due to the absence of stock (out of stock), causing loss of potential sales (lost sales). This phenomenon appears to be suspected due to the ERP system's lack of customer demand forecasting as a production planning module. This research has compared the statistical analysis with machine learning based on time series consumer demand for 10 products to provide the best demand forecasting. The research using the CRISP-DM (Cross-Industry Standard Process for Data Mining) methodology. Feature engineering involves technical variables like quantity demand and seasonality, fundamental economic like IHSG and USD to IDR exchange rates, and new features in demand forecasting at is pandemic awareness level. During the COVID 19 pandemic, the pandemic level of awareness of society affected to demand rate cause this variable is crucial to demand forecast. ANN is the best model of machine learning for demand forecasting tasks in an FMCG company, confirmed by consistently giving the smallest of MAPE. Demand forecasting using machine learning has been successful to increase supply chain performance by providing better production planning than before implementation. Enhancement of supply chain performance confirmed by decreased loses sales and increased inventory turnover ratio.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	FMCG, Demand Forecasting, Production Planning, Statistical Analytic, Machine Learning
Subjects:	H Social Sciences > HD Industries. Land use. Labor > HD40 Inventory control Q Science > Q Science (General) > Q325.5 Machine learning T Technology > T Technology (General) > T58.5 Information technology
Divisions:	Faculty of Engineering and Information Technology > Department of Information Technology
Depositing User:	Faisal Ifzaldi
Date Deposited:	05 Jan 2022 10:01
Last Modified:	05 Jan 2022 10:01
URI:	http://repository.sgu.ac.id/id/eprint/2298

Actions (login required)

View Item