Company X is facing the challenge of remaining competitive in an increasingly challenging economic environment. One way of doing this is to increase operational efficiency in manufacturing, to increase resource utilization and reduce waste. The quality testing process was identified as an opportunity area for semi-automation to improve the utilization of one of Company X’s plants, and the purpose of this research was to investigate if machine learning would be effective in achieving this. Using historic data of batches produced in the plant, machine learning models were developed to predict the viscosity of batches made, to reduce the amount of time spent on testing and adjusting the product to achieve results that are within specification. Regression models were successful in achieving this, and the opportunity exists for further study through gathering more end-to-end data of the manufacturing process to refine the machine learning models even further. 

Keywords

Quality testing, manufacturing, machine learning, regression models, classification models.