Impact on Network Performance of Probe Vehicle Data Usage: An Experimental Design for Simulation Assessment

Table 1

Relative (%) and absolute errors (min) for long trips (Manhattan distance origin to destination centroid over 3 km) when 5 replicas are considered in the base scenario at 90%, 95%, and 99% confidence levels.

Mean Travel Time	Number of Replicas, 5
min	TG	TNG	TOURIST	RG	RNG	REGULAR	EG	ENG	EXPERT	GUIDED	NON-G
Mean	9.54	9.88	9.80	9.45	9.74	9.67	9.53	9.98	9.88	9.49	9.85
Abs error (min)
90	0.28	0.22	0.22	0.11	0.17	0.15	0.11	0.17	0.14	0.11	0.17
95	0.36	0.29	0.29	0.14	0.23	0.20	0.14	0.22	0.19	0.15	0.22
99	0.60	0.48	0.47	0.23	0.38	0.33	0.23	0.37	0.31	0.25	0.37
Rel error (%)
90	2.90%	2.23%	2.24%	1.12%	1.79%	1.57%	1.12%	1.71%	1.46%	1.20%	1.73%
95	3.77%	2.91%	2.92%	1.45%	2.34%	2.04%	1.46%	2.23%	1.91%	1.56%	2.26%
99	6.26%	4.82%	4.84%	2.41%	3.88%	3.38%	2.42%	3.69%	3.16%	2.59%	3.74%

TG: tourist guided; TNG: tourist nonguided; RG: regular guided; RNG: regular nonguided; EG: expert guided; E-NG: expert nonguided. Tourist, regular, and expert columns indicate driver classes (averaging guided/nonguided cars) and guided/non-G columns indicate mean over driver classes.