Passenger Flow Scale Prediction of Urban Rail Transit Stations Based on Multilayer Perceptron (MLP)

Table 7

Accuracy analysis of passenger flow forecast at key stations—average daily passenger flow.

		Working days				Nonworking days
		Real value	Predictive value	AE	RE (%)	Real value	Predictive value	AE	RE (%)
Wangjing station	MLP	52278.86	55241.35	2962.49	5.67	34324.33	30859.36	3464.97	10.09
	RBF		42518.02	9760.83	18.67		58468.51	24144.18	70.34
	KNN		80979.51	28700.65	54.90		53692.67	19368.33	56.43
	Multiple linear regression		78608.62	26329.76	50.36		49270.28	14945.94	43.54
Xuanwu gate station	MLP	48186.43	52131.55	3945.12	8.19	24992.00	32609.65	7617.65	30.48
	RBF		66599.10	18412.67	38.21		27668.02	2676.02	10.71
	KNN		46616.37	1570.06	3.26		37529.43	12537.43	50.17
	Multiple linear regression		60622.19	12435.76	25.81		37243.07	12251.07	49.02
Wangfujing station	MLP	54176.57	44590.78	9585.79	17.69	48212.33	45665.49	2546.84	5.28
	RBF		25790.44	28386.13	52.40		35479.34	12733.00	26.41
	KNN		38219.10	15957.47	29.45		34991.91	13220.43	27.42
	Multiple linear regression		38979.40	15197.17	28.05		22537.04	25675.30	53.25
Sanyuanqiao station	MLP	102299.14	73384.90	28914.24	28.26	45860.33	44970.05	890.28	1.94
	RBF		65641.31	36657.83	35.83		36802.97	9057.36	19.75
	KNN		56235.18	46063.96	45.03		36105.05	9755.29	21.27
	Multiple linear regression		69438.94	32860.21	32.12		41644.00	4216.33	9.19
Sun palace station	MLP	50452.57	46294.36	4158.21	8.24	26563.67	23033.66	3530.00	13.29
	RBF		38063.89	12388.68	24.56		30234.18	3670.51	13.82
	KNN		44542.43	5910.14	11.71		24633.10	1930.57	7.27
	Multiple linear regression		44584.70	5867.87	11.63		26859.95	296.28	1.12