The Multiple Linear Regression Model on Compression Strength of Corrugated Boxes
Using the multiple linear regression model, we analyzed the test results on thousands batches of corrugated boxes samples to seek the relation between compression strength and a series of affecting variables such as bursting strength, edgewise crush strength Puncture resistance, ply adhesive strength, etc. And based on the non-linear regression model results, we discovered the relation between compression strength and other independent variables is more closed to a power function relation. We made the logarithm transformation on best fitted non-linear regression model to establish the predictive formula for the compression strength. To test how well our formula to predict the compression strength based all other factors, we randomly selected several batches lab tested samples both from Guangdong and Hunan Lab, and compared the real test results to the predicted values of our formula and of the traditional formulas such as Kellicutt, Makee and Wolf formula. We found the predicted values of our model are closer to the real tested value than the predicted values of the other three formulas.