RESEARCH PROPOSAL A MACHINE LEARNING APPROACH FOR QUALITY ASSURANCE ASSESSMENT IN HQCF PRODUCTION INDUSTRY High Quality Cassava Flour (HQCF) is the most viable product among the cassava products in term of business and productivity. From its name, it is expected that the final product is of high quality. However, as a result of the technology involved in the processing of this product, which is an upgrade of the traditional cassava processing technology, the final product usually deviates significantly from the expected quality. These deviations are majorly due to contaminations which normally get into product during processing, thereby demeaning the potentials and productivity that can obtained from this industry. The need for engineers to re-arrange the various machines and equipment in HQCF industry to enhance automation has been suggested by Kolawole and Agbetoye (2007) but most researches in this area focus more on just one or two operations rather than a holistic research that considered all the processes involved. As a result of this, the quality of HQCF obtained is usually compromised. In most cases the extent of the quality deviation results into the production of low quality HQCF that poses health related challenges to the final consumer. Hence, this research aims at using machine learning to control the quality of HQCF production. Methodology The methodology that will be applied for this research is as follow: (1) Survey: Medium and Small Scale Enterprises (MSMEs) with standard (HACCP/QACCP) system for controlling input materials and processing procedures to meet pre-determined quality characteristics will be surveyed (2) Data Collection: These data must contain data of different levels of HQCF contaminations using non-destructive spectral imaging technology such as Rapid and Atomic Absorption Spectrometry (AAS) analysis (Tao, et al., 2021). (3) Data Integration and Model Development: Discriminating features will be extracted from each category of the pure and contaminated flour. (4) Training of the Model: Calibrated models will be built using state-of-art machine learning algorithms including artificial neural network, support vector machine (SVM) and random forest (RF) (5) Validation of Developed Model: Calibrated models will be validated using unknown pure and contaminated samples (6) Implementation: Final developed robust model will be deployed in an automated HQCF system. Benefits HQCF production industry will be significantly improved in terms of processing, production and quality of product. The research will also help in eliminating associated problems such as loss of production time, poor quality products, poor maintenance practice, poor management and high rate of post-harvest losses from the industry. More importantly, the success of this innovative research in HQCF industry will definitely develop and enhance food processing industry as many food and Agro-Allied will most likely adopt the system to improve productivity and quality of products from the various sectors. References (1) Kolawole, O.P. and Agbetoye, L., (2007). Engineering Research to Improve Cassava Processing Technology, International Journal of Food Engineering, Volume 3 Issue 6, Article 9. (2) Tao, F., Liu, L., Kucha, C., and Ngadi, M.O. (2021) Rapid and Non-Destructive Detection of Cassava Flour Adulterants in Wheat Flour Using a Handheld MicroNIR Spectometer, Biosystem Engineering, March 2021, 203(10):34-43.