On Monday, 16 February 2015 16:25:47 UTC+1, Adam H. Kerman  wrote:   
   > bob  wrote:   
   > >On 2015-02-13 20:23:01 +0000, Marc Van Dyck said:   
   > >>rcp27g@gmail.com explained on 13-02-15 :   
   >    
   > >>>or putting in positive singalling control on others (ie where the    
   > >>>crossing is protected by railway signals that aren't cleared for the    
   > >>>train until the barriers are down and the crossing positively checked    
   > >>>to be clear).   
   >    
   > >>This is perfectly feasable but requires to order the gates to go down   
   > >>at a distance that is longer than the braking distance of the train.   
   > >>This means there will be a long delay between the gate going down and   
   > >>the train actually passing the grade crossing.   
   >    
   > >Indeed, this is the case.  It is less convenient but allows for    
   > >positive safety.   
   >    
   > >>As it has been mentioned,   
   > >>motorists are unpatient creatures; if the delay is too long, people   
   > >>think the gates are faulty and start turning around them. You might end   
   > >>up this way with a grade crossing that is inherently less safe, because   
   > >>of human nature...   
   >    
   > >Solved by making the barriers block the whole road.  As the crossing is    
   > >positively checked to ensure the barriers are down and the crossing is    
   > >clear before clearing the signals for the trains, the issue of cars    
   > >being trapped within the crossing is avoided.   
   >    
   > Clearing level crossings several minutes before the train arrives mitigates   
   > against non-existant risk, and it's quite labor intensive. How is the   
   > cost of delay justified? How is the personnel cost justified?   
      
   This "non-existant risk" just killed 6 people.  It is not "quite labor   
   intensive".  I've visited signalling control centres where these crossings are   
   operated from.  They form a very small part of the task that the signallers   
   doing other railway control    
   tasks have to deal with.  If the crossings involved were fully automated,   
   there would be no saving in manpower.   
      
   In 2013 there were 2087 collisions on crossings in the US, resulting in 251   
   deaths and 929 injuries.  In the UK, there were 6 collisions and 10 deaths   
   (all on the smaller crossings without full positive protection).  Normalising   
   for population, the US    
   kills 5 times as many people on level crossings than the UK.   
      
   Based on these statistics, this kind of positive safety at level crossings   
   would save somewhere in the region of 200 lives per year in the US.  That's   
   the justification.   
      
   > No lives have been saved except during the 5 or 10 seconds before the   
   > train arrives.   
      
   It is extremely widely documented that a minority of motorists, cyclists and   
   pedestrians routinely pass warning lights, signs, audible alarms and, if   
   possible, barriers, in the hope of avoiding waiting for trains.  Where   
   barriers come down in a way that    
   offers the absolute minimum of delay to road users, this minority still   
   disregards them.  If the only victims of this stupidity were the people   
   disregarding safety measures put in place for their own protection, then might   
   take the darwinian view.  The    
   fact is, as demonstrated here, the results of their stupidity kill people.    
   Even in the case where the only victim is the driver, or even where nobody   
   gets injured, the psychological effects on the train driver are not something   
   anyone should have to go    
   through.  With a full barrier and interlocked crossing, the crossing is fully   
   blocked and positively checked to be safe before the train is given permission   
   to enter it.   
      
   Robin   
      
   --- SoupGate/W32 v1.03   
    * Origin: LiveWire BBS -=*=- UseNet FTN Gateway (1:2320/1)