Online Traffic Accident Spatial-Temporal Post-Impact Prediction Model on Highways Based on Spiking Neural Networks

Algorithm 3

Data extraction from dataset.

(1)	Input
(2)	the nearest sensor in PeMS to the location of traffic accident
(3)	upstream sensors set of denoted by
(4)	the closest time point in PeMS to the start time of traffic accident
(5)	time point set in PeMS after denoted by , where
(6)	speed of sensor in denoted by
(7)	sum of detection, verification, response, and clearance time of traffic accident
(8)	Initialization
(9)	means if the traffic flow is in dissipating process
(10)	means the vehicle is low-speed driving or waiting
(11)	For in T do
(12)	For in S do
(13)	If
(14)	Then If
(15)	Then update
(16)	Else
(17)
(18)	If
(19)	Then
(20)	End if
(21)	End if
(22)	Update
(23)	Break
(24)	End for
(25)	If
(26)	Then update
(27)	Break
(28)	End for
(29)
(30)	Output
(31)	full-recovery time of traffic accident
(32)	half-recovery time of traffic accident
(33)	maximum accumulative queue length
(34)	average accumulative queue length