.png?width=556&height=152&name=Abley_MasterLogo-Website%20(1).png "logo")
.png?width=200&height=55&name=Abley_MasterLogoWhite-Website%20(1).png)
A key question you generally ask when doing road safety work in low and middle income countries (LMIC), is where do I start? There are road safety issues everywhere. Another challenge is how much weight do you place on the crash history data that is available, as reporting rates for non-fatal crashes are often low and there are issues with the accuracy of crash locations and many of the reports have missing information (even for fatal crashes). The answer is that often you can make use of the crash history data, but you want to validate the high crash locations using predictive analysis tools.
Safety Performance Functions developed for 47,000km highway network
During 2021, volume-based Safety Performance Functions (SPF, which are also called crash prediction models) were developed for the national highway network, just over 47,000km. The national network makes up just over 10% of all roads in Indonesia. The vehicle models consider road stereotype, roadside land-use (4 types) and province groups (4 types) and volumes of four-wheel vehicles (as this data was more accurate than the volume of 2/3 wheelers). Pedestrian crash rates were also developed using the same factors but using population density as a surrogate for pedestrian activity levels.
Risk maps for vehicles and pedestrians
Using the SPFs, traffic volumes and population density we produced risk maps that showed low to high risk (in five levels) road sections for vehicles and pedestrians (these were typically 2km or more long). The high-risk category includes the highest 10% of road sections by crash risk. The medium high sections where a further 20% of the network. Example of these risk maps for both vehicles and pedestrians are shown below.
Figure 1 – High-risk corridors for vehicle-only crashes
Figure 2 – High-risk corridors for pedestrian crashes
Black routes and black spots identified across Indonesia
In addition, 900+ high crash risk locations (blackspots) were identified across Indonesia. By targeting improvements to these high-risk black routes and black spots we can more effectively target high-risk crash routes across the country.
I will be presenting on the road safety work our team have been doing in Indonesia (over the last two years) at the upcoming Australasian Road Safety Conference in Christchurch. Hope to see you there either face-to-face or virtually! https://australasianroadsafetyconference.com.au/
