Skip to main content

Table 5 Regression parameters of developed accident prediction models

From: Safety assessment of Czech motorways and national roads

M01 \( \widehat{N}={\beta}_0\bullet {\left({AADT}_{major}\right)}^{\beta_1}\bullet {\left({AADT}_{minor}\right)}^{\beta_2}\bullet \exp (Type)\bullet \exp (Control) \)
  ALL INJ FAT+SEV SEV+SLI FAT SEV SLI PDO
β 0 7.185
E-05
1.365
E-05
2.534
E-06
1.286
E-05
7.885
E-07
1.746
E-06
1.112
E-05
5.820
E-05
β 1 0.671
β 2 0.337
Type = crossing 1.354
= roundabout 1.307
= merging 0.195
= diverging 0
Control = no –0.761
= yes 0
M02 \( \widehat{N}={\beta}_0\bullet {AADT}^{\beta_1}\bullet {L}^{\beta_2}\bullet \exp (Curve) \)
  ALL INJ FAT+SEV SEV+SLI FAT SEV SLI PDO
β 0 1.326
E-03
2.519
E-04
4.676
E-05
2.373
E-04
1.455
E-05
3.221
E-05
2.051
E-04
1.074
E-03
β 1 0.772
β 2 0.612
Curve = curved 0.624
= straight 0
M03 \( \widehat{N}={\beta}_0\bullet {AADT}^{\beta_1}\bullet {L}^{\beta_2} \)
  ALL INJ FAT+SEV SEV+SLI FAT SEV SLI PDO
β 0 2.702
E-04
5.133
E-05
9.529
E-06
4.837
E-05
2.965
E-06
6.564
E-06
4.180
E-05
2.188
E-04
β 1 0.967
β 2 0.699
M04 \( \widehat{N}={\beta}_0\bullet {\left({AADT}_{major}\right)}^{\beta_1}\bullet {\left({AADT}_{minor}\right)}^{\beta_2}\bullet \exp (Turn) \)
   ALL INJ FAT+SEV SEV+SLI FAT SEV SLI PDO
Rural β 0 4.982
E-04
2.392
E-04
4.994
E-05
2.284
E-04
1.075
E-05
3.919
E-05
1.892
E-04
2.591
E-04
β 1 0.481
β 2 0.476
Turn = yes –0.267
= no 0
M05 \( \widehat{N}={\beta}_0\bullet {\left({AADT}_{major}\right)}^{\beta_1}\bullet {\left({AADT}_{proportion}\right)}^{\beta_2}\bullet \exp (Turn) \)
   ALL INJ FAT+SEV SEV+SLI FAT SEV SLI PDO
Rural β 0 2.777
E-03
1.722
E-03
4.235
E-04
1.615
E-03
1.067
E-04
3.168
E-04
1.298
E-03
1.055
E-03
β 1 0.907
β 2 0.772
Turn = yes –0.558
= no 0
M06 \( \widehat{N}={\beta}_0\bullet {\left({AADT}_{sum}\right)}^{\beta_1}\bullet \exp (Legs) \)
  ALL INJ FAT+SEV SEV+SLI FAT SEV SLI PDO
β 0 1.273
E-05
4.329
E-06
3.516
E-07
4.304
E-06
2.598
E-08
3.256
E-07
3.978
E-06
8.404
E-06
β 1 1.220
Legs = 3 –0.464
= 4 0
M07 \( \widehat{N}={\beta}_0\bullet {AADT}^{\beta_1}\bullet {L}^{\beta_2} \)
   ALL INJ FAT+SEV SEV+SLI FAT SEV SLI PDO
Rural β 0 4.780
E-02
1.721
E-02
3.421
E-03
1.598
E-02
1.227
E-03
2.194
E-03
1.379
E-02
3.059
E-02
β 1 0.434
β 2 0.584
Urban β 0 9.142
E-03
3.382
E-03
4.687
E-04
3.274
E-03
1.081
E-04
3.606
E-04
2.914
E-03
5.759
E-03
β 1 0.648
β 2 0.713
M08 \( \widehat{N}={\beta}_0\bullet {AADT}^{\beta_1}\bullet {L}^{\beta_2}\bullet \exp \left({\beta}_3\bullet Minor\right) \)
  ALL INJ FAT+SEV SEV+SLI FAT SEV SLI PDO
β 0 6.607
E-04
2.378
E-04
4.728
E-05
2.209
E-04
1.696
E-05
3.032
E-05
1.906
E-04
4.228
E-04
β 1 0.842        
β 2 1.094        
β 3 0.216        
M09+10 \( \widehat{N}={\beta}_0\bullet {\left({AADT}_{major}\right)}^{\beta_1}\bullet {\left({AADT}_{proportion}\right)}^{\beta_2} \)
   ALL INJ FAT+SEV SEV+SLI FAT SEV SLI PDO
3-leg β 0 3.842
E-05
2.612
E-05
4.706
E-06
2.463
E-05
1.492
E-06
3.214
E-06
2.142
E-05
1.229
E-05
β 1 1.221
β 2 0.507
4-leg β 0 9.272
E-05
6.305
E-05
1.136
E-05
5.945
E-05
3.600
E-06
7.757
E-06
5.169
E-05
2.967
E-05
β 1 1.278
β 2 1.004
M11 \( \widehat{N}={\beta}_0\bullet {AADT}^{\beta_1}\bullet {L}^{\beta_2}\bullet \exp \left({\beta}_3\bullet Minor\right) \)
  ALL INJ FAT+SEV SEV+SLI FAT SEV SLI PDO
β 0 6.543 4.450
E-04
8.015
E-05
4.195
E-04
2.541
E-05
5.475
E-05
3.648
E-04
2.094
E-04
E-04
β 1 0.885
β 2 0.985
β 3 0.091
  1. For interpretation of model types (M01, …, M11) see Tables 1 and 4
  2. As introduced in Section 3.3, severity levels are FAT, SEV, SLI, PDO, plus combinations: FAT+SEV, SEV+SLI, INJ (injury accidents), ALL (all accidents)
  3. Explanatory variables (abbreviations, definitions, units) are listed in Table 2 and Section 3.4
  4. Regression constants (β0) are reported using scientific notation (e.g., 7.185E-05 = 7.18510-5)