Differential Evolution with Novel Mutation and Adaptive Crossover Strategies for Solving Large Scale Global Optimization Problems

Table 4

Results of multiple-problem Wilcoxon’s test for ANDE, ANDE-1, and ANDE-2 versus SDENS, jDElsgo, DECC-DML, DASA, EOEA, MA-SW-CHAINS, and DMS-PSO-SHS over all functions at a significance level 0.05 (with 1.25E + 05 FEs).

Algorithm			value	Better	Equal	Worse	Dec.
ANDE versus SDENS	203	7	0.000	19	0	11	+
ANDE versus jDElsgo	210	0	0.000	20	0	0	+
ANDE versus DECC-DML	162	48	0.033	14	0	6	+
ANDE versus DASA	44	166	0.023	6	0	14	−
ANDE versus EOEA	26	184	0.003	2	0	18	−
ANDE versus MA-SW-CHAINS	30	180	0.005	2	0	18	−
ANDE versus DMS-PSO-SHS	104	106	0.970	12	0	8	≈
ANDE-1 versus SDENS	210	0	0.000	20	0	0	+
ANDE-1 versus jDElsgo	210	0	0.000	20	0	0	+
ANDE-1 versus DECC-DML	149	61	0.100	13	0	7	≈
ANDE-1 versus DASA	44	166	0.023	6	0	14	−
ANDE-1 versus EOEA	61	149	0.100	4	0	16	≈
ANDE-1 versus MA-SW-CHAINS	19	191	0.001	2	0	18	−
ANDE-1 versus DMS-PSO-SHS	110	100	0.852	12	0	8	≈
ANDE-2 versus SDENS	185	25	0.003	17	0	3	+
ANDE-2 versus jDElsgo	194	16	0.001	18	0	2	+
ANDE-2 versus DECC-DML	157	53	0.052	15	0	5	≈
ANDE-2 versus DASA	45	165	0.025	6	0	14	−
ANDE-2 versus EOEA	37	173	0.011	3	0	17	−
ANDE-2 versus MA-SW-CHAINS	31	179	0.006	4	0	16	−
ANDE-2 versus DMS-PSO-SHS	105	105	1	13	0	7	≈