> A disproportionate number of them occurred at Ashland, VA. Which makes me think it's a bug in the traffic design. Without knowledge of the crossing, it's def

> A disproportionate number of them occurred at Ashland, VA. Which makes me think it's a bug in the traffic design. Without knowledge of the crossing, it's definitely possible. Some crossing designs are better than others, and changing the design of a crossing can be very difficult depending on the site details. Changing the path of rails is always difficult due to the constraints of trains, so a poor crossing will almost certainly need to have the alignment of the road changed, but sometimes there's not enough buildable space around the crossing to build an over or undercrossing (grade separation) and it is expensive and disruptive. Sometimes it's possible to close the road if the crossing is bad and the road is deemed unnecessary, but it's pretty rare to deem roads unnecessary.