Personal Pages of Fred Van Bennekom     |     home
Airport Security Line Queues
Great Brook's survey products:
Our Survey Design Workshop teaches you how to conduct your own survey projects from questionnaire design to survey administration to data analysis.
August 10, 2006
Please don't misinterpret this rant.  I believe strongly in effective security screening at the airport.  I welcome the checks as they help protect my safety in the air -- and I fly a lot.  However, I wish those in charge of security lines understood a little more about queuing systems.  And I wish just one reporter knew enough to ask an intelligent question.
With today's terrorism plot in London, security screening at airports has increased dramatically.  Screening time has increased, and lines are longer.  The message put out through the media is to arrive at the airport 2 hours before your flight.  That might be fine if you have a 6 am or 2 pm flight, but for the rest of the day, you'd better camp out at the airport.  Why is the 2-hour lead time bogus?  Basic queuing theory.
Let's say an airport has 10 security screening stations, each fully staffed and operational.  For simplicity's sake, let's say that under normal times each passenger takes 1 minute to get screen.  (In reality that screening time would vary, but again this is a simple example.)  That means that each station can process 60 passengers in an hour and the entire screening area can screen 600 passengers per hour.
Let's also assume that the airport has planes taking off each hour with a total of 600 passengers.  In other words, the system is in perfect balance.  The demand for security screening equals the demand for security screening.  Since the passengers won't arrive at exactly the rate of 10 per minute, the length of the queue will ebb and flow.  But when do you ever see the security stations idle?  Only at very slow periods in the airport, such as 6 am (which isn't all that slow!) and mid-afternoon.  The lack of idle capacity is critical to this argument, so keep that in mind.  
Now the terrorist plot gets foiled, and security screening is beefed up.
Again, for simplicity's sake, let's assume that each individual screening now takes 2 minutes, twice as long.  (That's probably an overstatement, but it keeps the analysis simple.)  The capacity of the screening area gets cut in half to 300 passengers per hour (since the time required per passenger has doubled).  If enough flights are cancelled combined with passengers canceling their trips, then the system could be near balance.  But that large a drop in demand is unlikely.  So, word goes out through CBS4Boston, Fox25, WNAC, and WCVB that passengers should arrive 2 hours ahead of their flight times.  How would that help?
Answer: it doesn't.  In fact, the effect is to create congestion that just adds to already frayed nerves.  Here's why.
If everyone shows up 2 hours in advance, then everyone's relative position in the queue stays the same.  Please reread that sentence.  It's key.  Remember, there's no idle capacity.  If demand is still 600 passengers and capacity is 300, then the queue will build by 300 people each hour.  In other words, after one hour, there would be 300 people in queue, after two hours, there would be 600 people in queue, after three hours, there would be 900 people in queue, etc.  That last person -- the 900th -- in line after 3 hours won't be processed for 3 hours!  
So, what's the impact of everyone arriving early?  More people waiting longer in longer lines.  Advancing the arrival of demand doesn't affect the capacity of the processing stations.  That's the key point that seems missed by all, and the airport PR people who put out the "come early" message are doing a public disservice.  Now we passengers aren't stupid.  We recognize the situation intuitively.  So what's our response? We arrive 3 hours ahead of the flight!  Result, longer lines, more congestion.  The people who arrive extra early do benefit.  They advance their position in the expanding queue, and get through earlier -- and might make their flight.  But if everyone follows suit, you gain nothing.  
The only way everyone arriving early would help the situation is if -- IF -- there were occasional idle capacity in the screening area that would now be utilized because the queue is better stocked.  How many times have you ever seen idle security screening during at the airport, especially during peak times?  Exactly.  The gain from advancing the queue is minimal.
So what is the answer?  The capacity must be equal to or greater than the demand.  (We really want capacity to exceed demand, but I'm keeping this analysis simple.)  Either demand must be cut or capacity must be increased.  If screening does take longer -- the 2 minutes versus 1 minute in this example --  then the only real answer is to increase the number of screening stations.  (In this simple example, the number of stations would need to double.)  Putting aside the cost of the equipment and extra staffing, we've seen how the architecture of the airports constrains this expansion.  
I fly out of Manchester, NH (MHT) and Providence, RI (PVD), both "new" airports, but both built before September 11.  The area for security screening wasn't designed to accommodate expansion.  Passengers are all funneled through this narrow opening.  From a bird's eye view, It is a literal bottleneck.  The airports have gotten creative and experimented with different designs, but they are near their limit without radical reconstruction of the areas.  (This is true for most of the airports I visit.)  Regardless, are they able to increase capacity literally overnight?  Unfortunately, no.  
The other option is to work to decrease the screening time, which in effect increases capacity.  Things that slow down the x-ray screenings could be addressed before they get into the system.  We see the agents telling passengers in advance to remove items that will be confiscated.  Without that, the screening times would really spike.  But there's a limit.  This limit is how long it takes the agent to process the bags viewed in the x-ray equipment.  We don't want to shortchange that time.  The only other option is to reduce the number of carry-on bags, as the British did today at Heathrow, but that means more checked luggage that must be screened at other stations and now be handled by the airlines.  
Besides, the airlines don't want us to check bags.  It increases their costs since they have to handle the luggage.  Carry-on luggage keeps the task in the passengers' hands.  You know the airlines won't increase staff to handle more checked bags -- though they may charge you for it.  
~~~
Security is necessary, and increased security is very necessary so long as there are terrorists among us.  But let's not make the situation worse by making recommendations to air travelers that have no impact and really only make matters worse.  Airports and the TSA need to design airport screening areas to bring on added capacity during periods of surging demand.  Having everyone arrive extra early just creates a more chaotic situation.