Development of Predictive Models for Estimating Female Students’ Dimensions Essential for Classroom Furniture Production

Abstract

Creating anthropometric databases require considerable resources like workforce, equipment and funds and thus, the design of classroom furniture (CF) is typically not based on anthropometric principles. This study addresses this challenge by developing models for predicting several female students’ dimensions essential for optimal CF production in secondary schools. An aggregate of 240 students participated in the study and brute-force search technique implemented in ANFIS was employed to select the two most influential of the five input measurements. Regression analyses were employed in modelling the anthropometric data obtained. Out of the 18 developed models, 8 were quadratic while 5 each exhibited two factors interactions [  and linear relationships [ . Adjusted R2 values obtained ranged from 0.902-0.999, 0.876-0.997, 0.881-0.999, 0.993-0.998, 0.950-0.999 and 0.983-0.995 for KH (Knee Height), EH, P, SHH (Shoulder Height), PBL and HW (Hip Width) respectively. The ANOVA results show that the models satisfactorily predicted the needed dimensions for optimal production of CF.