Originally Posted by
xzh445
There are multiple safeguards built into the system, to avoid boarding a person to the wrong flight. When operating in fully automated (scan and process) mode, I can't think of a scenario that allow a the wrong person to avoid. If any of the manual processes are used...the failure would come from the operator, not the application.
I think that the core problem is that the computer's behavior is the same, and the same sound is emitted, for a trivial issue (wrong boarding group when companions travel on separate PNRs), a routine issue of moderate importance (validate exit row criteria), a different routine issue of moderate importance (seat assignment changed), and a critical issue involving operator intervention (wrong boarding pass scanned). In my opinion, this should have been obvious -- there's a reason that your fire alarm doesn't sound like a telephone, and vice versa. :-) Then again, hindsight is always 20/20...
The human brain is poorly equipped to deal with this kind of repetitive task, which normally requires very little attention but infrequently requires a great deal. As [MENTION=677629]ajGoes[/MENTION] said, this is an area of no small amount of study.
The same principle is behind regular reports of "TSA missed XYZ." Some tasks are
truly mind-numbing.