This is an assignment of my school class and I’d do my best to give out my opinion to existed design and would try to learn from them.
A photo of people using e-gate in the airport. Via https://eznewlife.com/168711
For my first assignment, I decided to make design critique on e-gate in my country. The e-gate system is an automatic process of speedy customs clearance. However, the system often being complained by first-time users. It’s kind of an interesting truth that interns at the airport told me that younger people rarely have any problem using the machine instead of feeling inpatient because older people often stuck at the gate.
In this case, user need is pretty simple: they want to go pass the customs as soon as possible. The e-gate follows exactly the same process that human counter process: Validate the passport, recognize face and fingerprint. That’s all. So what are the problems of the design? Why are people being confused at the e-gates and stuck there?
The navigation design works pretty nice at the first step in the first gate, most people are able to understand they have to put the passport on the scanner. There are sign and illustration telling people to do so. Even without the illustration, the layout itself looks clear enough. However, many people had left their passport there and entered the next gate and cause troubles.
I believe that there are reasons for designers to split the process into two gates (such as do not want irrelevant people having access to the camera and fingerprint reader) The problems here is that the gate opens immediately after the passport being validated. I also understand “open the gate” is a clear signal to tell users they did everything right. The problem here is the designer forget that people probably would be too eager to walk forward to remember their passports. Never to mention it is understandable for common users to believe they would get their passports after the next gate. Although the process sound perfectly corresponds to the traditional process, the mapping methods might not able to be applied by all the users. There are users traveling abroad for the first time in their life. Even for those who have travel several times, they might not understand what is happening in the traditional customs counter. All that they realize could be “give out the passport, stand there for a while, wait for the customs say I could go and get my passport back.” In this situation, users and designers analyzing the task in different ways. While the designers thinking bottom-to-top, the users only observe the pattern they could observe.
There are more troubles at the second gate, which meant for face and fingerprint recognition. While entering the gate, the monitor would give out clear instruction telling the user to stand in front of the camera and put the finger on the fingerprint reader in both animation, text, and recorded voice. There are footprints on the floor suggest the best place for the user to stand (though probably could not be applied to every user) The navigation looks perfect, so what’s wrong with the design? The interaction design doesn’t work nicely with the face recognition. Although users could see their own face at the monitor so they would not confuse about are they in the camera or is the camera still working, however, the machine doesn’t give out advice very often. For those who have enough experiences, they know that frontier hair, glasses, different angles or decorations could confuse the face recognition program ( so would these elements confuse human eyes) so they’d try to adjust the way they are to help the recognition system. Unfortunately, the system couldn’t handle this error in an efficient and valid way. It also does not tend to suggest users take off the glasses (if needed) as the people in human counter would ask travelers once they feel like it’s needed.
To improve the design, I think that the programmer could give another signal for finishing the validation step and only open the gate while nothing being left on the scanner. Some animation hint could be added in the face recognition step on the monitor to help users fix the problem that the machine could not recognize their face.