شبکه های عصبی و رگرسیون خطی چندگانه برای پیش بینی ابعاد مدرسه کودکان برای طراحی ارگونومی مبلمان مدرسه

The current study investigates the possibility of obtaining the anthropometric dimensions, critical to school furniture design, without measuring all of them. The study first selects some anthropometric dimensions that are easy to measure. Two methods are then used to check if these easy-to-measure dimensions can predict the dimensions critical to the furniture design. These methods are multiple linear regression and neural networks. Each dimension that is deemed necessary to ergonomically design school furniture is expressed as a function of some other measured anthropometric dimensions. Results show that out of the five dimensions needed for chair design, four can be related to other dimensions that can be measured while children are standing. Therefore, the method suggested here would definitely save time and effort and avoid the difficulty of dealing with students while measuring these dimensions. In general, it was found that neural networks perform better than multiple linear regression in the current study

Primary school students spend much of their time sitting on chairs daily. Bad design of furniture may lead to health and learning problems. Therefore, design of furniture with proper dimensions is critical to encourage appropriate postures (Straker et al., 2010). So many studies were conducted to ergonomically design the school furniture using anthropometric measurements which vary according to many factors. Most of these studies showed school children frequently use furniture that is not suited to their anthropometry (Straker et al., 2010). Anthropometric measurements are not easy to perform and they need a large sample size and a lot of dimensions. Some of these dimensions are necessary to ergonomically design the chair and yet they are not easy to measure. This study attempts to find some easy-to-measure dimensions that are capable of predicting the difficult-to-measure ones used in designing school furniture for primary school students.

To ergonomically design chairs, five anthropometric dimensions are necessary. These dimensions are usually not easy to measure. However, they can be predicted through models such as linear regression and neural network using easy-to-measure dimensions. Four anthropometric measures were used as inputs due to their ease to be measured where they can all be measured while the person is standing. A comparison was made between neural network and linear regression for predicting the anthropometric measurements. By considering (R2) only, the two models are almost the same; and by considering (S) value, all the measurements are better predicted by neural network. Generally, neural network is better than linear regression in predicting anthropometric measurements needed for ergonomic chair design. In this study, only four anthropometric dimensions were chosen as inputs. Future studies may consider other input dimensions which may lead to better performance in predicting output dimensions. Future studies may focus on prediction of female body dimensions which may lead to models with different parameters.