Discrete Particle Swarm Optimization with Scout Particles for Library Materials Acquisition

Table 4

Results of different initialization strategies on two test cases.

Case	Measure	Improvement		Execution time
		Random	Greedy	Random	Greedy
I	Average	52.46%	52.11%	1.6455	1.5956
	Stdev	0.2805	0.2795	1.5258	1.3455
	Observations	480	480	480	480
	𝑧-score	0.1942		0.5362
	𝑃 value	0.8460		0.5918
II	Average	71.32%	73.01%	779.9824	800.9922
	Stdev	0.3675	0.3722	318.31	324.77
	Observations	160	160	160	160
	𝑧-score	−0.4090		−0.5843
	𝑃 value	0.6825		0.5590