Page 43 - Proceeding of Atrans Young Researcher's Forum 2019_Neat

“Transportation for A Better Life:
Smart Mobility for Now and Then”

23 August 2019, Bangkok, Thailand

Table 4 Effect of vehicle type on driver response. other 1.000
Go decision Distance from the stop -0.079 0.000** 0.924
variables line
coef. p-value odds ratio
Speed 0.085 0.000** 1.089
Type of vehicle
Constant -1.028 0.197 0.358
PC -1.469 0.000** 0.230 -2LL (287.854), R (0.786), N (593)
2
MC 1.000 Front vehicle same type go (FVSG), Front vehicle in adjacent
Distance from -0.075 0.000** 0.928 lane same type stop (FVASS), * is statistically significant at 5%
the stop line level, ** is statistically significant at 1% level
Speed 0.071 0.000** 1.073
Constant 1.220 0.048* 3.388 For the passenger car driver, significant
2
-2LL (468.823), R (0.742), N (824) factors are “a front vehicle, the same type, go
* is statistically significant at 5% level, ** is statistically (FVSG)”, and “a front vehicle in adjacent lane, same
significant at 1% level type, stop (FVASS)”. FVSG makes a driver
probability to go 3.631 times and FVASS makes a
A distance from stopping line make a driver driver probability to go 0.099 times when compare
probability to go 0.928 times, with estimated with other situations. That means a driver tend to
coefficient -0.075. Speed makes a driver's follow their heading vehicle that decide to go
probability to go 1.073 times, with estimated through the intersection and tend to stop when their
coefficient 0.071. The odd-ratio and of coefficient’s front vehicle in the adjacent lane stop. These
sign imply that when the distance from stopping line findings is in-lined with Elmitiny et al. (2010) and
increase, a decision to go decreases. In addition, Bao et al. (2018) whose found that a driver is more
when the approaching speed increase, a decision to likely to go when the heading driver chooses to pass
go increase. This result is the same study in the intersection.
Pathivada and Perumal (2019) Chen et al. (2018)
Haque et al. (2016) and Elmitiny et al. (2010). Table 6 Effect of adjacent vehicle on motorcycle
For the effect of vehicle type, the table show rider response.
that a passenger car driver has a probability to go go decision
0.230 times less than those of the motorcyclist. The variables odds
findings imply that at the onset of yellow, the coef. p-value ratio
passenger car’s drivers are likely to stop than a Front vehicle
motorcyclist. This result is in lined with Pathivada et FVDG 2.228 0.021* 9.492
al. (2019) whose found that a passenger car has a other 1.000
probability to stop more than a motorcycle. These Front vehicle in adjacent
findings also statistically confirm the descriptive lane
analysis results shown in the previous section. FVADG 1.245 0.038* 3.593
other 1.000
3.4 Effect of adjacent vehicle on driver Distance from the stop
response line -0.060 0.000** 0.942
By controlling distance from the stop line Speed 0.062 0.000** 1.068
and vehicle speed, effects of adjacent vehicle on Constant 0.286 0.793 1.332
passenger car driver and motorcyclists’ decisions are -2LL (139.205), R (0.723), N (231)
2
as shown in Table 5 and Table 6, respectively. Front vehicle different type go (FVDG), Front vehicle in
adjacent lane different type go (FVADG), * is statistically
Table 5 Effect of adjacent vehicle on passenger car significant at 5% level, ** is statistically significant at 1% level
driver response
Go decision For the motorcycle riders, significant factors
Variables are “a front vehicle, different type, go (FVDG)”, “a
coef. p-value odds ratio
Front vehicle front vehicle in adjacent lane, different type, go
FVSG 1.290 0.004** 3.631 (FVADG)”. FVDG makes a rider probability to go
other 1.000 9.492 times and FVADG makes a driver probability
Front vehicle in to go 3.593 times. That's mean a motorcyclist who
adjacent lane decided to go have influence from difference type in
FVASS -2.361 0.000** 0.099 front and front in the adjacent lane when they choose

18