Most dams are designed with traffic lanes that could be used by regular vehicules or mobile cranes for example. Why isn't this type of loading considered in the stability analysis of dams?
This is an interest question. The dam design must withstand all the stresses from lowermost to uppermost (I.e. bottom to top). The maximum stresses at the bottom of the dam would dictate the required safety factors for the entire dam, including the top of the dam. Moving (dynamic) loads of traffic at the top will be well inside the safety parameters of a properly designed large dam capable of supporting traffic. The top construction of the dam roadway must of course comply with the maximum load requirements of a conventional roadway. The properly designed dam will support it.
It is of course an error not considering traffic loads. However, a typical distribute load will be of 10 to 20 KPa acting on the width of the road (7 to 10 m wide, normally), which is very small, when compared to the selfweight of the dam. Therefore, this load probably is often neglected for overall stability.