Online Semisupervised Learning Approach for Quality Monitoring of Complex Manufacturing Process

Table 4

t-test result on injection molding dataset.

Scenario	Model 1	Model 2	T value	value
Sporadic access (current batch prediction)	ParsNet++	ParsNet	15.7305	2.66e−07
	ParsNet++	NADINE	13.6995	7.77e−07
	ParsNet++	ODL	40.1705	1.62e−10
	ParsNet++	ResNet18	10.3871	6.39e−06
	ParsNet++	VGG11	13.9605	6.72e−07
Sporadic access (next batch prediction)	ParsNet++	ParsNet	16.7425	1.64e−07
	ParsNet++	NADINE	26.3042	4.69e−09
	ParsNet++	ODL	11.328168	3.32E−06
Infinity delay (current batch prediction)	ParsNet++	ParsNet	33.8308	6.37e−10
	ParsNet++	SCARGC-SVM	44.5193	7.15e−11
	ParsNet++	SCARGC-1NN	32.4618	8.84e−10
Infinity delay (next batch prediction)	ParsNet++	ParsNet	18.9618	6.19e−08
	ParsNet++	SCARGC-SVM	23.5353	1.13e−08
	ParsNet++	SCARGC-1NN	28.3756	2.57e−09