If this is the case, I will happily eat my words. Again, I have seen people do crazy stuff at the airport because they think it will save them a minute, or a dollar, or be more convenient for them. I'm just speaking from experience working in this environment.
I've commissioned and been involved in cost effectiveness studies for airports on interventions involving design, passenger flow and accessibility. One thing that has always surprised me is that when we did studies that interviewed people after experiences and we got an idea why people misinterpreted and misunderstood things it was recognisable that heterogeneity in behaviour was often a function of things we hadn't considered beforehand.
The challenge though is that results are often implementable in a rather dichotomous manner. We can design for pure efficiency knowing that a tiny number of people not able to navigate that system can cause it to implode (just like this), or we can design a system for the lowest common denominator and accept that it makes the system more inefficient for everyone. How do we consider these trade-offs? In practice, they're also not binary, but more a continuous function, and also not likely linear.
For example, an airport wants to use bidirectional travelators and escalators. Which side gets the forward direction? In a domestic airport, we know that we can place them in the same manner that road traffic travels (left side forward in countries who drive on the left and vice versa) - this is more intuitive for most airport users. In international terminals in big connecting hubs, does this intuitively make sense for most people? Think somewhere like Dubai where you have people coming from everywhere!
One project we did was testing the placing and design of signs to try and nudge more people in the right directions. Guess what? The signs caused disruption by slowing/stoping people, and more slowing/stoping than it saved at the travelator or escalator. Net impact was particularly poor since the people who the sign benefited were not the people who needed the nudge! What that led to was a more discrete separation of the flow of passengers so that the flow was only in one direction. This seems simple but is more more difficult to integrate into design. This is a total abstraction from the case, but just highlights that this may be as much of a design issue, and we cannot design for infinite redundancy, and cannot always anticipate behaviour of an individual who may be unfamiliar, inexperienced, or have some other exogenous factor that exposes a design feature.