Performance Optimization of an Improved Cassava Mash Sifting Machine

Abstract

Mechanized cassava processing is important in improving food quality and reducing manual labour in rural and semi-industrial applications. The optimization of an improved cassava mash sifting machine using a general full factorial design (GFFD) to enhance its operational efficiency was investigated. In this research, the response variable (viz. sifting efficiency) was analyzed with respect to two critical operational factors: cassava mash mass and operational time. A 3² factorial experimental design with three replicates per treatment was used, resulting in 27 experimental runs. The experiments were conducted at the Nigerian Stored Product Research Institute (NSPRI), Port Harcourt Zonal Office. The machine was tested using mash masses of 30 kg, 60 kg, and 90 kg at operational times of 0.2, 0.4, and 0.6 hours. Data were subjected to statistical analysis, including two-way ANOVA, regression modeling, and response optimization using MINITAB 21 software. The results indicated that both mash mass and time of operation had significant effects (p < 0.05) on sifting efficiency. The optimal sifting efficiency was recorded at a mash mass of 30 kg and a time of 0.6 hours. The multiple linear regression model developed showed high predictive accuracy with R² = 99.10%, Adjusted R² = 98.54%, and Predicted R² = 97.45%. The optimization analysis also yielded a high composite desirability value of 0.991, validating the model’s reliability. This study for process of optimization in cassava mash sifting operations and offers a framework for improving mechanized food processing systems in developing regions.