Modeling vehicle interior noise exposure dose on freeways: Considering weaving segment designs and engine operation

Figures & data

Figure 1. Weaving segment configuration (a) Type A, (b) Type B, and (c) Type C (source: Highway Capacity Manual [HCM 2000; Transportation Research Board, 2000]).

Table 1. Specification of the weaving segments and basic segment on the test route.

Segment^a	Scope^b	Length^c	No. of lane changes^d	No. of lanes	Traffic flow
WA1	Southmore / N MacGregor	370 m (1214 ft)	1	4	Towards SH 288 and I610, crowded in the morning, sometimes congested
WA2	Holcombe Blvd / Southmore	560 m (1837 ft)	1	3	Towards downtown, crowded in the evening, sometimes congested
WB1	N MacGregor / Holcombe Blvd	680 m (2230 ft)	0	4	Two-lane off ramp is 187 ft and immediately connected to a signalized intersection on feed road, often congested
WB2	Reed Rd / Bellfort St	220 m (722 ft)	0	3	Two-lane off ramp is 432 ft and connected to a signalized intersection, rare congestion
WC1	Yellowstone / SH 288	1090 m (3576 ft)	≥2	4	Diverging segment from one four-lane to two two-lane towards SH 288 and I610, often congested
WC2	Bellfort St / Yellowstone Blvd	920 m (3018 ft)	≥2	4	Merging segment from two two-lane to one four-lane towards downtown, seldom congested
Basic	Airport off-ramp / Airport on-ramp	9,656,064 m (31,680 ft)	0	3	Towards Pearland, seldom congested

Notes. ^aThe first two letters indicate the type of weaving segments, e.g., WA is Type A weaving segment; the digit is the number of the specific weaving segment in the study, e.g., WA2 is the second Type A weaving segment.

^bThe entrance and the exit, indicating the scope of each segment.

^cThe length from the point on the entry gore to the point on the exit gore.

^dNumber of required minimum lane changes for weaving vehicles.

Table 2. Traffic conditions in groups.

Group of traffic conditions	LOS	Density range (car/1.6 km/lane)*
Free-flow mode	A and B	0–24
Mild traffic	C and D	24–36
Heavy traffic	E and F	>36

Note. *Source: Highway Capacity Manual (HCM 2000; Transportation Research Board, 2000); compiled from Exhibits 24–2 and 24–3 and Exhibits 25–4 and 25–5.

Figure 2. Observed interior noise levels on segments.

Figure 3. (a) Probability distribution of daily interior noise exposure level. (b) Daily interior noise exposure levels on segments.

Figure 4. (a) Average hazard equivalent exposure time factors (EETFs). (b) Total EETFs under different traffic conditions.

Figure 5. (a) Hazard quotient values for the threshold 85 dB(A) on segments. (b) Hazard quotient values for the threshold 90 dB(A) on segments.

Figure 6. Importance of the 11 predictor variables for the bagged decision tree–based in-vehicle noise model.

Figure 7. Out-of-bag regression error for 300 deep grown trees.

Table 3. Text description of the first 10 splits in decision tree for regression.

Node	Text description
1.	if speed <62.7239 then node 2 elseif speed ≥62.7239 then node 3 else 33.9469%
2.	if speed <23.3969 then node 4 elseif speed ≥23.3969 then node 5 else 29.8813%
3.	if type <0.5 then node 6 elseif type ≥0.5 then node 7 else 37.0308%
4.	if speed <0.403899 then node 8 elseif speed ≥0.403899 then node 9 else 28.6712%
5.	if cLoad <10.5874% then node 10 elseif cLoad ≥10.5874% then node 11 else 31.9281%
6.	if speed <91.1465 then node 12 elseif speed ≥91.1465 then node 13 else 39.0558%
7.	if type <2.5 then node 14 elseif type ≥2.5 then node 15 else 35.7611%
8.	if IRI <3.66 then node 16 elseif IRI ≥3.66 then node 17 else 26.6343%
9.	if speed <16.3229 then node 18 elseif speed ≥16.3229 then node 19 else 28.9359%
10.	fit = 11.2171%
11.	if acc <0.906592 then node 20 elseif acc ≥0.906592 then node 21 else 31.9589%
12.	…

Note. *The type is defined by the number of minimum lane changes required for weaving.

Figure 8. Fitted regression lines for (a) training data set and (b) validation data set.

Transportation Research Board. 2000. Highway Capacity Manual, 4th ed. HCM 2000. Washington, DC: Transportation Research Board.

 Google Scholar