We’ve been doing some R&D work on the application of routing algorithms and wanted to share our progress.
We’re interested in using routing algorithms to help identify the safest journey along a given route. Usually routing tools provides the shortest or quickest route from a starting point to a destination. To create a route, a journey needs to have a minimum of two points, multiple waypoints can also be used.
We’ve been exploring how we can add contextual information to a route. The video below shows an example in Iraq. Here we show how incident data can be added to a routing algorithm. We have a start point at Baghdad International Airport and a finish point in Baghdad. The incident data displayed is for 2007.
In this example, we selected two different route options. We can then see the incidents that happened within 500 metres of the roads chosen. Some routes may provide the quickest journey between two points, but additional data can add important context. The number of nearby incidents may be significant as it could suggest that an area may see a higher number of security related incidents over time. Security managers may want to plan journeys that avoid areas or roads with a higher number of incidents.
General routing algorithms work really well for users living in busy cities like New York, London or Dubai. However, if a company is operating in a frontier region, like Iraq, Afghanistan or Libya, routing can become more complicated.
Journey managers operating in frontier regions have different priorities when planning a journey from one point to another. Safety and security of personnel becomes paramount when figuring out the best route to take to get to a destination. We decided to develop a routing algorithm to try and make managers’ jobs quicker and easier.
We have several years’ worth of incident data relating to security issues in Iraq within an existing application (our IMS). We combined this data with more traditional routing information such as quickest or shortest route. This way we can show routing options in relation to security incidents within a buffer of 300 or 500 metres from the road. This allows those responsible for planning journeys in high risk environments to assess which route will be the quickest and is most likely to be the safest.
Combining routing algorithms with additional contextual information is not just useful in frontier regions. For example, we can see applications in developing safest routes for bike journeys. You could combine routing data with information about accidents involving cyclists so you could more easily identify dangerous sections of your journey.
We used a number of open source software packages to explore routing algorithms including:
If you are interested in routing or have any comments, drop us an email.