Abstract
The jumbo iced tea business is growing rapidly, but one partner in Solo faced challenges in manual transaction recording and sub-optimal stock management due to sales fluctuations. This problem has an impact on operational efficiency and customer satisfaction. This research aims to design and build a point of sales application with sales prediction features using random forest regression algorithms that are able to cope with non-linear data and have high accuracy. The method used is waterfall, which includes the stages of identification and planning, analysis, system design, implementation, and maintenance. The dataset for prediction is taken from historical daily sales data, with variables used including day of the week, weekend status, previous day's sales, and average sales of the previous 7 days. The test results show that the random forest regression model with a total dataset of 30 data has a good level of accuracy, with an average absolute error percentage of 2.85%. The application system developed is able to improve the operational efficiency of the jumbo iced tea business, support better decision making, increase profitability, and improve customer satisfaction through more structured stock and transaction management
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call