Road construction analysis using regression technique

Estimating cost in construction is important to the city's design and planning management hence, the construction cost estimate must not be overpriced which may cause corruption or underpricing that leads to unreliable or low-quality road projects. The total estimated cost is only valid in the same year it was proposed because of the inflation rate the costs may change. The researchers applied Multiple Linear Regression technique in predicting total estimated cost for road construction analysis. The model is evaluated by the means of R-squared to determine the variables if they are correlated or overfitting. The calculated R-squared is equals to 0.696598 with the predictor variables (x1 & x2) Roadbed width and Net length and it means that the predictors (Xi) explain 69.7% of the variance of Y. The higher the R-squared result, the better fit it is for the Multiple Linear Regression model. It also shows that X1 and X2 are significant predictor variables. The coefficient of multiple correlation (R) is equals to 0.834624 and it means that there is a very strong correlation between the predicted data and the observed data whereas the dependent variable (y) is the Estimated cost.
CCS CONCEPTS: Multiple Linear Regression • Construction Estimation • Engineering