What I am calling the "dead man" switch is a buzzer that goes off in the cab of a train locomotive at infrequent intervals. The locomotive engineer has only a short time (seconds) to respond (slap the switch) or else his locomotive controls will be safely disabled. This is to prevent a "runaway train". The locomotive will come to a safe controlled stop, while simultaneously alerting the railway system of the problem
...more »
What I am calling the "dead man" switch is a buzzer that goes off in the cab of a train locomotive at infrequent intervals. The locomotive engineer has only a short time (seconds) to respond (slap the switch) or else his locomotive controls will be safely disabled. This is to prevent a "runaway train". The locomotive will come to a safe controlled stop, while simultaneously alerting the railway system of the problem on that train and that section of track, in the event the operator has a heart attack or is otherwise incapacitated. Hitting the “dead man” switch within the designated period of time allows the train to continue. All pertinent train-operating data, to include the "dead man" switch, is digitally recorded. If the operator is in the process of actually working some control, e.g. throttle or brake, the "dead man" switch resets itself to occur after some period of inactivity of the controls.
It should be feasible to put a similar “dead man” switch system in air traffic control towers. The controller would have to respond to the "dead man" switch at periodic and infrequent intervals during the work shift.
NOTE: We should not blindly adopt the “buzzer” that is used in the railroad industry. That would just serve as an alarm clock. Use a mixture of visual and auditory signals that an alert air traffic controller would be expected to notice during the course of his duties when not actively supporting the take off/landing of aircraft. When the “system” is actively supporting the take off/landing of aircraft, the “dead man” switch is disabled.
Initiate an escalating series of actions if an air traffic controller "failed" to respond to the "dead man" switch.
I have only listed two such actions below to develop some metrics (measureable standards - the air traffic controllers should help write these).
First "failure" – Establish phone contact with the controller. Alert controller(s) in adjacent area(s) to be prepared to assist with air traffic control for the affected area.
Second "failure" – Establish physical contact with the controller. Controller(s) in adjacent area(s) take over air traffic control for the affected area.
The use of the “dead man” switch in all air traffic control towers manned by just one controller, day or night, should be cheaper and more effective than requiring a second controller on each shift.
« less
full details »
