기상 및 대기 환경적 영향 요인을 고려한 커피 소비 예측 연구

A Study of Coffee Consumption Prediction Considering Atmospheric Environment and Meteorological Factors: Application of Machine Learning Techniques using Big data

This study aims to evaluate variable-importance and predict coffee consumption by using a machine learning of artificial neural network. The variable-importance analysis indicated that the location has been shown to be an important factor for success due to easy accessibility to people and it guarantees high profits. In addition, meteorological factors such as high temperature, low humidity, and low wind speed have a positive effect on sales by increasing outdoor activities. It can be explained that the weather changes directly affect consumer behavior and change psychology leading to spending consumption. In terms of atmospheric environment factor such as high levels of particulate matter, people find indoor activities instead of outdoor activities, so it is necessary to understand the consumer behavior and adjust marketing strategies. As a result of predictive modelling has a strong prediction power related to coffee consumption, meteorological and atmospheric environmental factors. Therefore, the results of coffee demand forecast simulation can be used for data-based decision making. For coffee operating companies, predicting sales and customer visits by situation of this study can help to effectively manage food material and staff optimization.