Exploring Further Advantages in an Alternative Formulation for the Set Covering Problem

Table 6

RPD metrics for each instance group and algorithm in both studies.

Instance group
	d-ACO (%)	n-ACO (%)	ipv	d-ACO (%)	n-ACO (%)	ipv	d-ACO (%)	n-ACO (%)	ipv
First study
A	0.43	0.08	81.80	0.08	0.03	63.83	0.79	0.43	46.03
B	0.53	0.28	46.60	0.36	0.17	51.89	0.62	0.39	37.24
C	1.86	1.74	6.54	1.22	1.17	3.74	2.49	1.89	24.06
ALL	0.94	0.70	25.69	0.55	0.46	17.20	1.30	0.90	30.60
Second study
Instance group
	d-ABC (%)	n-ABC (%)	ipv	d-ABC (%)	n-ABC (%)	ipv	d-ABC (%)	n-ABC (%)	ipv (%)
A	0.38	0.19	51.25	0.16	0.07	56.65	0.66	0.30	54.25
B	0.74	0.19	74.56	0.50	0.09	82.81	1.11	0.29	73.77
C	10.71	1.07	90.01	9.91	0.90	90.95	12.75	1.48	88.39
ALL	3.94	0.48	87.80	3.52	0.35	90.04	4.84	0.69	85.73