In the transportation industry, there is much debate whether a Linear Referencing System (LRS) or a Spatial Referencing System is best suited to manage road and highway infrastructure. Arif Rafiq, Esri Canada’s Transportation Industry Manager, argues that it’s not necessarily a choice between the two. Rather, it’s about creating systems using methods that accommodate the way different users want to interpret, or see, information.
Just before falling asleep, my five-year-old daughter enjoys a good story from her dad. One of these bonding occasions featured a six-page storybook about two children who wear coloured glasses only to see specific parts of the world around them. Swapping them with different colours revealed different and exciting things to look at, even though they never changed the environment.
A well-designed Linear Referencing System (LRS) works the same way. It can give users the ability to figuratively swap their spectacles to work with only the components of their transportation network that they choose; all the while having a truly simple method of locating features along the route.
Three "methods" of viewing different sets of data for the same stretch of road.
Consider for a moment someone from the safety department at your Ministry of Transportation who identifies a crash on route 12B Eastbound at 45.3km; anyone else should be able to identify where that is with relative certainty. This of course is the standard way of the location referencing business in Provincial Ministries of Transportation; it’s a huge step forward from other agencies who still use convoluted segment IDs to identify the crash location (where the heck is network segment 456B-54ER-01?)
Interestingly though some agencies are looking to skip the Linear referencing concept only to move to a fully Spatial Referencing System (SRS). OK, let’s investigate this: using an SRS we might identify that same crash location as Latitude 52.465010 / Longitude -114.293923, but is this reference any clearer than the cryptic segment ID? I would argue that the LRS reference is simply more natural; I don’t need a GPS device to find Route 12B Eastbound at 45.3km.
Where this crash occurred is plain to see only because there’s a map in the background. Without it, the lat/long coordinates are not immediately obvious
Proponents of Spatial Referencing argue that this method allows a wide range of analysis tools that’s simply not possible with Linear Referencing, such as generating a Maintenance Record Report given an arbitrary polygon drawn on a map. Certainly true, but the same could be said about Linear Referencing: try generating a Road Log Mileage report using a Spatial Referencing System.
So it’s clear that both have advantages. Rather than choosing between the two, it might make sense to define a Location Referencing System that can utilize a Linear or Spatial Referencing “method” to identify the same road feature. In other words, we can examine features on the same transportation network by using different location methods depending on how we want to (or are required to) reference it. Pavement management may reference project sections via a linear stationing method, while the geomatics group references the same feature(s) using a projected 3D coordinate method.
With today’s transportation infrastructure asset management technology, we can build Location Referencing Systems with a whole array of “methods” that enable consumers of the data to work with features regardless of how we want to reference them; kilometre offset, X-Y coordinates, stationing, meters from the bridge pier… the list is endless. We can create environments where large numbers of users can interact with the same data any way they want – and wear whatever colour shades suit their fancy.