Investigation of ANN Model Containing One Hidden Layer for Predicting Compressive Strength of Concrete with Blast-Furnace Slag and Fly Ash

Table 4

Summary of different quality assessment criteria over 100 simulations with different ANN structures.

Criteria	ANN-16N		ANN-24N		ANN-33N
	Train	Test	Train	Test	Train	Test
R^2
Min	0.9077	0.6947	0.8462	0.6882	0.8542	0.7766
Average	0.9365	0.8886	0.9418	0.8895	0.9486	0.885
Max	0.9482	0.9228	0.9574	0.9285	0.9611	0.9201
Std	0.0062	0.028	0.0173	0.0295	0.0145	0.0221
RMSE
Min	3.8059	4.5031	3.4086	4.4266	3.3328	4.8466
Average	4.1591	5.5521	3.954	5.5317	3.7171	5.6654
Max	4.9525	10.6746	6.4868	10.9525	6.3695	7.8546
Std	0.1925	0.7268	0.4915	0.7873	0.4261	0.5431
MAE
Min	2.7629	3.3248	2.5026	3.2971	2.4531	3.3841
Average	3.0742	3.8967	2.9071	3.8506	2.7212	3.8979
Max	3.7582	4.7447	4.9218	5.1684	4.6705	5.8684
Std	0.1541	0.2493	0.3774	0.318	0.323	0.3243