Seaweed Demand Production Prediction Using Multiple Linear Regression Algorithm

Abstract

This study examines the problems faced by PT XYZ in predicting the fluctuating demand for dried seaweed, which has resulted in difficulties in determining the optimal production quantity. The objective of this study is to develop a more accurate prediction model using multiple linear regression algorithms, taking into account marketing costs, the number of marketing personnel, and the number of consumers. The methods employed include collecting historical sales data of seaweed from 2020 to 2023, followed by data cleaning and transformation, data exploration, and model development. The results indicate that marketing costs have the greatest influence on market demand, with a coefficient of 34.68, followed by the number of consumers and the number of marketing personnel. The multiple linear regression model built demonstrates very high accuracy, with an R² value of 99.83% on the training data and an RMSE of 0.54 on the test data, indicating a low level of prediction error. This study makes an important contribution to the development of market demand prediction models and provides practical benefits for PT XYZ in optimizing production planning and marketing strategies.