Line vs. Kiosk: optimize the time for that burger to get into your stomach
Mcdonalds has been equipped with self serve kiosks for a while now. As hungry student I want my stomach to be filled as soon a possible.
Let’s estimate how much time it takes for a person to order in line.
- “May I get a big mac meal with a medium coke?”
- “For here or to go?”
- “To go.”
- “How would you like to pay?”
- “Credit card.”
- “$x.xx please.”
- “Beep”
- “Here is your receipt, number xxx”
- “Thanks.”
Here is around 40 words. This roughly translates to 40 seconds.
Now comes the kiosk.
- touch to start
- select take away
- select burgers
- select big mac
- select medium meal
- select coke
- select add to order
- select done
- select proceed to checkout
- pay here
- beep
- take receipt
What is the theoretical fastest time we can order from a kiosk? We can safely assume that there is at least 1 second of animation between each step. So if we have memorized the location of each button, and able to move our finger right in front the button just in time when the transition animation finishes, we can finish that order in 12 seconds if we use our non-dominant hand to reach our wallet while our dominant hand is busy performing the order. I don’t think anyone would spend the time to memorize the location of every button, so each person has around 0.2s of reaction time to locate the next button and another second to move to the next button. So for a seasoned veteran who orders a big mac all the time, it will take around 2.2 seconds per step for a total of 27 seconds. Multiplying it by a safety factor of 1.5, we get around 40s. Why the fudge factor? Because not everyone is super fast at ordering. Also keep in mind we are computing the average time people in front of you are taking.
Now we can appreciate the elegance of the kiosk system. As cumbersome as it might seem, it is probably designed to to approximately match cashier performance. But without taking actual data, it is pretty safe to assume a kiosk and a cashier are probably same. I have a bias towards kiosks being slower. Reason being cashiers spent more time and practice on taking orders efficiently and they have a more efficient software interface. The reason I pointed out this is probabably a bias is that I have failed to account for a cashier can also hold up a line by being away from desk to complete other tasks, while a kiosk cannot hold up a line.
Although there is an efficient market hypothesis that probably implies trades are carried out in an optimized manner. Judging by the trade of a mcdonalds meal, we can tell not everyone is making rational decisions when lining up. Sometimes we see huge lines at the cashiers, but each kiosk has only 1 or 2 person. Other times we also see huge lines at the kiosks, but only a few people at the cashier. If the lines sizes are self regularizing, the ratio between the sizes of cashier lines and kiosks lines would have been more stable.
So how to fill our stomach fast? Quickest rule of thumb is to go to the shorter line. A more cumbersome rule is to statistially compute the average time an average person spend on cashier vs kiosk, and go to the line that will get your order the fastest. Yes I know bummer :) Chances are, we are probably not gonna die if we lose a few minutes to eating, water is probably more critical when we are in this situation.