Safe Passage - The road to safer travel

Team Name

The Sippy Skippys

Team Members

Project Description

Safe Passage is an interactive website to be used as a tool to analyse crash data and highlight crash hotspot locations by enabling the use of filters to sort the collated data by location and crash severity.

The website enables the user to input a region, suburb, postcode, or road name and view all relevant crash data. A specific road safety report, containing general area information, the summarised data and possible solutions, can be viewed and downloaded by the user.

The simple format allows for easy visualisation of the Queensland crash data to identify where and how to improve Queensland roads, providing government agencies a tool to assist in future decision-making processes. This will in turn save the Queensland Government time and money.

Safe Passage can also be used as a results-based analysis tool such that current implemented solutions can be assessed and their effectiveness realised in the form of statistics.

Data Story

With over 14 million pieces of individual data related to car crashes which have occurred in Queensland, there exists no user-friendly resources that collate this data and allow for time and money efficient decisions to be made regarding road safety. The Queensland government invest millions of dollars a year into road maintenance and campaigns to reduce accidents only to see a 1% decrease in fatal car crashes over the past 5 years. This illustrates that perhaps crash causing factors are not being considered as a whole, and so there lacks an ability to implement effective targeted, relevant and viable solutions.

Our team has used three different data sets which collectively contain over 350 thousand events, with more than 40 different variables. We analysed the data to determine which of these variables would be required, completing filtering and conditional information via excel to remove those that were not. We determined the following to be of that standard:
- crash severity (medical treatment, minor injury, hospitalisation & fatal)
- geographic locations (map plot)
- crash type (single vehicle, multi vehicle, pedestrian hit)
- age demographic (ranging from 0 to 75+)
- influence of the driver in the crash (defective vehicle, fatigue, speed, drink & none)
- lighting conditions (daylight, darkness-lit, darkness-not lit & dusk/dawn)
- weather conditions (clear, rain & fog)

In creating an appropriate method of displaying the data we decided to assign crash severity as the overarching factor (in the form of a pie chart), and for each of the sub-factors, bar charts be constructed to highlight all other factors listed above. The geographic locations were overlaid onto a map of Queensland using Python.

It should be noted that we did search for government financial data relating to what was being spent in the clean up of accidents and also in preventative measures around the state, however, there did not appear to be any open source data available that fully captured our needs. Therefore, it is recommended that government data which encapsulates this be integrated into the website such that the full capability of data synchronisation is met.