A Novel Output Prediction Method in Production Management Based on Parameter Evaluation Using DHNN

Table 9

Comparison of output predictions and actual output.

Seasons	Actual output	Output prediction of Model A	Output prediction of Model B	Output prediction of Model B′	Error of Model A	Error of Model B	Error of Model B′
2008	101	98	99	100	2.97%	1.98%	0.99%
2009	119	117	118	118	1.68%	0.84%	0.84%
2010	135	139	140	137	−2.96%	−3.70%	−1.48%
2011	150	160	157	156	−6.67%	−4.67%	−4.00%
2012	160	165	164	162	−3.13%	−2.50%	−1.25%
Mean absolute percentage error (MAPE)					3.48%	2.74%	1.71%