It seems impossible to increase efficiencies in a manufacturing setting without first pinpointing the causes of waste. The goals of this research were to use the quality management data collected to determine which aspects were significantly correlated with manufacturing waste, and then to use machine learning techniques to create a model that could predict manufacturing waste. The case study research was used for this analysis. The data range from 2019 to 2021 and includes 215924 rows and 13 features. Five predictors and eight terminal nodes were used to create the final CART model, which achieved an R² of 0.389 and an accuracy of 97.31%. Inventory management, customer wait times, meetings, and unscheduled breaks all contributed to 80% of the time that was wasted. High levels of inventory management problems occurred in months 6-11. The waiting times caused by customers averaged 2.97 hours per month between months 7 and 11. Meetings that started at 8:00 lasted, on average, 7.5 hours. The majority of the time spent waiting is caused by customer complaints that come in between the hours of midnight and 5 a.m. At 7:00, with a 4.3-hour lag, most reports of inventory problems were made. Only quantitative secondary data were used in this analysis. Qualitative approaches should be prioritized for future studies. Models for predicting waste in the future could benefit from using random forests and TreeNet.