Analysis and Prediction of Overloaded Extra-Heavy Vehicles for Highway Safety Using Machine Learning

Table 8

Results of the NN.

	Neural network	Highway data (hour)	National and provincial highway data (month)	National and provincial highway data (quarter)	National and provincial highway data (year)
Running time (s)	BPNN	2.016	1.938	2.125	2.953
	GRNN	5.156	7.016	12.313	19.031
	WNN	0.484	1.219	0.563	1.656
	BPNN	0.987	0.967	0.962	0.972
	GRNN	0.534	0.558	0.832	0.863
	WNN	0.162	0.184	0.169	0.021
MSE (%^2)	BPNN	0.960	2.960	3.060	1.460
	GRNN	49.860	74.130	0.560	0.290
	WNN	0.810	1.110	0.930	7.684
RMSE (%)	BPNN	2.700	−0.150	3.880	−3.200
	GRNN	70.610	86.100	7.480	5.370
	WNN	9.000	10.520	9.640	12.524
ME (%)	BPNN	5.200	2.120	4.430	3.840
	GRNN	−10.150	−12.150	−3.020	−8.960
	WNN	5.480	−3.790	−5.710	−9.324
MAE (%)	BPNN	0.090	0.060	0.090	0.060
	GRNN	0.240	0.040	0.210	0.100
	WNN	18.250	48.570	41.620	59.050
MAPE (%)	BPNN	0.089	0.124	0.116	1.046
	GRNN	0.542	0.622	0.048	0.102
	WNN	0.079	0.069	0.073	9.413
RMSPE (%)	BPNN	0.059	0.026	0.058	0.894
	GRNN	0.640	0.684	0.403	0.871
	WNN	0.510	0.352	0.388	3.443