Key Workflows
...
Detailed metrics (Long Section...
Crossings
2 min
due to the computational complexity of precisely detecting crossings with existing linear infrastructure, the system makes some approximations in the earlier stages of the optimisations although effort has been made to minimise these as much as possible, some effect of these may be seen during in depth analysis representation crossings with existing linear infrastructure are shown on the map as icons the locations of these icons are tied to the resolution of the underlying data, which means that the icons can appear to be off by up to one unit of resolution (e g 30m) \[screenshot here] this is purely an artefact of the data resolution, and improves as the resolution is increased generation and evaluation modes precise crossing detection is not feasible in generation mode and an approximation is made instead this is done by buffering the linestring data and treating it as polygon data, which generally overestimates the number of crossings the buffering also acts as a proxy to the crossing angle, as crossings that are made closer to 90 degrees will be seen as shorter than crossings made at a shallow angle before results are shown on optioneer, or in evaluation mode in general, crossings are calculated using a slow but precise method that relies on the raw vector data therefore, all crossings visible on optioneer are precise and accurate however, as the results were optimised using approximate crossings, some options may appear unoptimised around crossing locations the most common symptom of this is trenchless sections that are too long, as their length was determined using over estimated crossing locations