Let me explain my overall objective and then the specific problem that led me to post on the forum.
I have a line feature of bus routes and the number of riders on each route for a given time period. The routes overlap frequently, but no two routes are exactly the same. I would like to "union" the overlapping route lines and then run a dissolve that sums up the total number of riders on route segments where there are 2+ routes that overlap. Unfortunately, the Union tool does not work for line features, so I have been hunting for a work-around. One option could be to use the "Split Line at Vertices" tool (to break the routes down into small segments) and then take the resulting layer and run the "Unsplit Line" tool in order to aggregate the line segments (and sum ridership) in the places where the bus routes overlap. Unfortunately, I can't seem to advance past the second step - which brings me to the specific problem:
When I run the Unsplit Line tool on the layer of split lines... nothing happens. Well, the tool successfully runs and the number of fields is reduced to the number of statistical summary fields I have chosen. However, the line segments are not aggregated and the number of attributes remains the same. Any suggestions on what I am doing wrong? Or alternatively, any ideas about how to achieve my objective through alternative means?
Unsplit works on pairs of lines termianting at a singe point (that is, at what was onced called, long ago and in a workspace far away, a pseduo node). If three or more lines come together, nothing happens. Thus if you have two lines that coincide or overlap in the middle (like this: >-<), and you properly intersect them with nodes at the intersections, Unsplit won't do anything.
You likely want to build routes: routes can share segments so the same bit of line (the central straight segment in the diagram above) can participate in both route features.