Evidently, the final report is online. The hogger's medical and vision state was indeed quite a mess.
" onclick="window.open(this.href);return false;
One thing I noticed, and which the NTSB should be thinking about, is whether these new designs of wayside signals currently being installed that have only one sun-reflecting hood are contributing to the failure to correctly identify signals. In the report, there is an exemplary photo of a wayside signal on the Pratt Sub. The sun glare on the face of the signal, combined with the lack of individual sun hoods that would block the sunlight over each color aspect, makes it more difficult to interpret the signals than it should be. The NTSB needs to track accident frequency on corridors equipped with these new darth vader signal heads.
Unless you go to robots, the health of employees will vary, and some employees will have vision that is more impaired than others. You cannot correct this very cheaply, but you can correct the signal designs, restoring them to the practices of old, for relatively little costs. I mean, just how many pennies would it cost to apply individual sun hoods to each aspect? A few bolts and piece if metal?
Signal engineering seems to be going backwards in the past few decades, driven by cost reduction. Old style engineering practices that were adopted as a result of past safety issues seem to be ignored nowadays with new signal installations. The signals at Georgetown Jct on the B&O, where the collision happened between Amtrak and MARC, is one example. There is a left hand curve as you come into the interlocking. The original EB signal for track 2 was much taller than the EB signal for track 1, to allow a stop aspect on the signal to be seen above the tops of the rail cars on any train moving around the approach curve on the adjacent track. After the accident, when the interlocking was rebuilt, the signals were replaced and the new EB signal for track 2 is now as short as the EB signal for track 1. The result is a shortened sight distance, where you have to be fully around the curve before you can read the signal aspect. This is as good as we can expect in 2013?