To run the program you must have Java installed. If you don't you can get it from here: http://www.java.com/en/download/manual.jsp
Then you double click the "StationComparison.jar" or the "run.bat" icon
Select Action->go to start the simulation.
Vehicle acceleration: valid values are from 0.5 to 10 meters/second squared. Default value:1
Maximum vehicle speed: valid values are from 1 to 20 meters/second. Default value: 5
Number of berths (station size) : valid values are from 3 to 20 berths. Default value: 5
Embarking time: this is an arbitrarily large list indicating how much time each vehicle must spend on the berth while passengers embark/disembark. This time includes door opening and closing. Valid values are integers from 4 to 300 seconds separated by spaces (not [Enter]!). When the list is depleted it restarts from the first value.
Example: this list (4 8 11 13 9) means:Both stations have an input queue holding enough vehicles to occupy all the berths. The input queues are automatically filled with vehicles. This means that both stations are tested under maximum load to measure their capacities.
Blue vehicles are waiting for passengers to embark/disembark. Red vehicles have finished and are ready to go.
In the serial station a new cycle starts when all the vehicles at the berths have finished embarking/disembarking. At this point the docked vehicles along with a new set of incoming vehicles from the input queue are promoted simultaneously.
In the parallel station a new cycle starts if there is at least one docked vehicle ready to go. At this point the ready to go vehicles back-up into the main station offline line simultaneously. When they stop these ready vehicles, along with an equal number of incoming vehicles from the input queue which will occupy their now empty berths, are promoted simultaneously. When the new vehicles arrive at the berths (i.e. when they dock) a new cycle may begin.
Serial station has a clear advantage if the variability of embarking time is small. This is because in parallel station vehicles lose time by performing extra maneuvers. To observe this run the simulation leaving the default values for other options but setting the embarking time list to this set: (4) (which means that all vehicles will spend exactly 4 seconds docked at the berth)
Parallel station has an advantage if the variability of embarking time is large. This is because in serial station vehicles lose too much time waiting the slow passenger to embark. To observe this run the simulation leaving the default values for other options but setting the embarking time list to this set: (4 4 4 4 80) (which means that vehicles will spend 4 seconds docked at the berth but every fifth vehicle will spend 80 seconds).
However it is the author's impression that variability in embarking times, as observed in daily practice in metro stations for example, is not large enough to give the advantage to parallel stations.