An ongoing issue for local (and national) planning of cycle routes is that an attractive route during the day may become unattractive at night time due to lack of lighting, perceived dangers, the relative lack of other users, etc.
There are a number of routing applications (websites, apps) such as RidewithGPS, Komoot, Strava, Google, … but one that has taken an interesting approach is cycle.travel. Even better, use of cycle.travel is free (you can make a voluntary contribution)!
The app offers a choice of preferences for how to plan the route including biasing towards a paved surface (probably useful for those with expensive road bikes), biasing towards a gravel surface (i.e. generally off road) and, interestingly, one for night time.
I’m not aware of the details of the algorithm used to pick suitable routes but the software makes use of OpenStreetMap (OSM) data and will be using surface, lighting and similar data within the OSM database. I don’t believe any data on “perceived danger” or frequency of reported crimes has been used but biasing the route planning towards well lit and well used routes may address people’s concerns about personal danger.
As a quick example, planning a route between the city centre and the Blue Peter Island in Alvaston:
For day time or general use, the app suggests using the Riverside path but, for a night time ride, it prefers some back streets and London Road.
I’d recommend exploring the capabilities of the software to find routes that might be attractive to those of us concerned about the suitability of certain routes at night.
The data used is from OSM so any routings that might not appear sensible are likely to be due to errors in the underlying data (“garbage in, garbage out”). People are encouraged to join the ranks of contributors to the OSM data and fix or enhance the areas you use, improve the underlying data, and thus improve the quality of the routing.