A Data-Driven Approach to Estimate Incident-Induced Delays Using Incomplete Probe Vehicle Data: Application to Safety Service Patrol Program Evaluation
Table 3
Incident-induced delays by incident types.
Incident types
Entire incident dataset (1/1/2019−12/31/2021)
Before SSP (1/1/2019−9/8/2019)
After SSP (9/9/2019−3/15/2020)
All counts
Delay occurrence
All counts ()
Delay occurrence
All count ()
Delay occurrence
Counts
%
Average IIDs (veh-h)
Counts
%
Average IIDs (veh-h)
Counts
%
Average IIDs (veh-h)
1 vehicle crash
229
162
70.7
54.83
78 (6)
57
73.1
89.11
14 (5)
9
64.3% [8.8%]
17.11 [72.00]
2 vehicles crash
116
96
82.8
111.84
24 (3)
23
95.8
157.69
12 (1)
11
91.7% [4.1%]
123.12 [34.57]
3+ vehicles crash
16
13
81.3
129.02
6 (2)
5
83.3
93.81
0 (0)
0
—
—
Debris
198
38
19.2
2.25
14 (2)
4
28.6
6.73
11 (2)
1
9.1% [19.5%]
0.01 [6.72]
Earlier crash
78
66
84.6
56.15
25 (1)
20
80.0
79.33
4 (0)
1
25.0% [55.0%]
7.07 [72.26]
Emergency vehicle
43
26
60.5
25.59
8 (3)
5
62.5
56.34
1 (0)
0
—
—
Slow traffic
42
34
81.0
432.77
7 (0)
7
100.0
118.80
4 (0)
4
100.0% [0%]
46.12 [72.68]
Stalled vehicles
4,446
1,556
35.0
11.15
418 (85)
260
62.2
18.60
606 (156)
221
36.5% [25.7%]
15.85 [2.75]
Towing operation
24
16
66.7
76.95
6 (0)
6
100.0
159.43
0 (0)
0
—
—
Vehicle fire
25
18
72.0
165.97
3 (1)
3
100.0
126.89
1 (0)
1
100.0% [0%]
57.37 [69.52]
Total
5,217
2,025
38.8
30.36
589 (103)
390
66.2
46.35
653 (164)
248
38.0% [28.2%]
21.21 [25.14]
Delay occurrence % = delay occurrence counts/all counts. : the number of incidents whose speed data are unavailable during the after (A) and before (B) SSP. (program benefit) = before value − after value.
