Page 50 - NGTU_paper_withoutVideo

Modern Geomatics Technologies and Applications

2
− 2 + 1 ≤ 1
{
0 > 1

Fig. 6. Downstream popularity membership function
By calculating the commuting and non-commuting cycling rate for the edges, using a linear regression model, the
coefficients of each of the environmental parameters were calculated. Input data is 213639 rows, representing 71213 streets in
the morning, afternoon, and evening time windows.
According to Table 4, the time parameter plays an essential role in making active travel. Access to the public transport
network has also had a significant impact on commuting trips. Most commuting trips are done in areas where access to the
public transport system is poor but the opposite is true for the non-commuting trips. Given the positive value of the minor
roads, it can be concluded that cyclists prefer the minor roads to the major roads. On the other hand, streets with high land-use
diversity have a negative impact on bicycle trips. According to the value of p-value, the impact of road length on the cycling
trips was not statistically significant.
Using the regression model developed in the previous step, the influential parameters of cycling trips were identified. To
determine the popularity of each street for bicycle trips at each time interval, by using the linear regression, the number of
bicycle trips per street was modeled. To do this, travel data in the Hotspot and Coldspot regions were used. Considering the
popularity of the route as a positive parameter as well as the rate of inhalation of pollution as a negative parameter of bicycle
trips, the proportion of each street for bicycle trips was calculated using the fuzzy rules defined in Table 2 and the fuzzy
inference system.
TABLE 4 RESULTS OF LINEAR REGRESSION MODEL IN ESTIMATING THE IMPACT OF ENVIRONMENTAL PARAMETERS ON THE
NUMBER OF COMMUTING AND NON-COMMUTING TRIPS
commuting non-commuting
Dependent Variables
Coef. p-value Coef. p-value
Constant 0.374 0.672
morning 0.868 < 0.01 -0.8 < 0.01
afternoon 0.825 < 0.01 -0.746 < 0.01
evening 0.284 < 0.01 -0.073 < 0.01

6