queues – The OM Blog by Heizer, Render, & Munson

Guest Post: Southwest Airlines’ Boarding Overhaul–When Queuing Theory Takes Flight

Providence College Professor Jon Jackson discusses a topic that every flier understands.

For over 50 years, Southwest Airlines has stood out with its open seating policy—passengers lined up, boarded in order, and picked any available seat. That system will soon end. Starting in January, Southwest will introduce assigned seating and its first premium seats, as part of a major redesign of its boarding process. Internally called “Project USA,” this initiative is more than a marketing shift—it’s a deep operational rethink based on one principle: if queuing isn’t good, boarding isn’t good.

Southwest’s new boarding plan is a live experiment in queuing theory and process design, balancing efficiency with customer experience. Boarding is a significant driver of turnaround time—a key metric for airlines. Every minute saved at the gate means higher aircraft utilization, lower fuel costs, and better schedule reliability, writes The Wall Street Journal (Oct 13, 2025).

Astrophysicist Jason Steffen’s boarding process, which boards passengers in a diagonal pattern, minimizes blocking and maximizes parallel activity. Simulations suggest it could reduce boarding times by 30–50%, but it relies on strict compliance and passenger discipline—hard to guarantee in practice.

Southwest’s new system will use a variation of the WILMA method—boarding Window, then Middle, then Aisle seats—to reduce aisle interference and speed up boarding. This approach is validated by queuing research, though the Steffen process is even more efficient in theory.

Boarding an aircraft is fundamentally a queuing problem—a test of bottlenecks, flow efficiency, and human behavior. The new plan introduces nine boarding groups and two parallel lines to create a smoother, more predictable flow.

Classroom Discussion Questions

1. How do the WILMA and Steffen boarding processes differ in terms of efficiency, complexity, and customer experience? Why might Southwest choose a simplified version rather than the most efficient theoretical model?

2. What behavioral factors might reduce the effectiveness of Southwest’s new boarding plan, and how could the airline design around them?

Guest Post: Fast Food Restaurants

Prof. Howard Weiss shares his insights with our readers monthly. We all spend time in fast food restaurants, so today’s topic should be of broad interest.

There are nearly 200,000 fast food restaurants, also known as quick serve restaurants (QSR), in the U.S. Obviously a key to these restaurants is short waits and fast service. Module D of your Heizer/Render/Munson textbook lists important measures for waiting line situations including:

 Average time that each customer spends in the queue
 Average queue length
 Average time that each customer spends in the system (waiting time plus service time)
 Average number of customers in the system

These measures have been increasing at many QSRs. However, the increase is not due to reduced productivity but rather to changes that have occurred in QSRs over the past several years. One change is that menus at fast food restaurants have expanded to include meals that take more time to prepare. The first fast food restaurant was White Castle, which opened in 1921 and had a very limited menu with only four items – a slider hamburger (which cost 5 cents), Coca Cola, coffee and apple pie. Today, QSRs have much more varied menus. Another reason that service times take longer is that patrons are becoming more sophisticated in placing custom orders, which take more time to prepare.

Recently, there has been an increase in the number of patrons who use the drive-thru lane. Much of this increase is due to COVID. All fast food restaurants reported an increase in 2021 of drive thru traffic with the percentage of patrons using drive-thrus being reported as 37% in one report and 52% in another. Because more customers are using drive-thrus the number of customers in line increases and therefore so does the waiting time.

There are steps fast food restaurants have taken to reduce the customer time in the system. Many have installed kiosks inside the restaurant so that ordering and payment is self-service. Just as ATMs increase the service capacity in a bank these kiosks increase the capacity at fast food restaurants. Another step is encouraging mobile ordering and payment so that the order will be ready when the customer arrives to pick it up. Some updated restaurant designs have increased the number of drive thru lanes.

Classroom discussion questions:
1. How have apps and kiosks changed how you receive service at QSRs?
2. What could be the downsides of QSRs using apps or kiosks?

OM in the News: TSA, Thanksgiving, and OM Come Together

TSA’s airport operations center in Arlington, Va., opened this summer and coordinates checkpoint operations at the 30 largest airports. Here, screens show projected and actual passenger volume by hour, plus projected wait times, staffing levels and number of lanes open for terminals at Chicago’s O’Hare and New York’s LaGuardia airports — TSA’s airport operations center coordinates checkpoint operations at the 30 largest airports.

With the busiest travel day of the year approaching the Sunday after Thanksgiving (2.8 million people will pass through U.S. airports that day), the Transportation Security Administration says it’s better prepared to handle record numbers of travelers and should make it through Christmas without gridlock.

“This marks the first winter test of a new command center in Virginia, built to react more quickly to trouble spots at airports,” reports The Wall Street Journal (Nov. 18, 2016). The command center has screens on 3 walls displaying data from the 30 largest airports. They monitor wait times, equipment failures, number of lanes open, and projected and actual passenger volume hour-by-hour at security checkpoints. Airports with waits of under 30 minutes for standard screening and 10 minutes for trusted-traveler PreCheck lines are green, but turn yellow if waits jump higher in either category. A morning conference call with airports and airlines takes place 7 days a week. From there, TSA has shifted screeners and canine teams and used overtime to nip major delays.

On a recent day, a checkpoint here in Orlando, showed a 38-minute wait around noon, grabbing TSA’s attention. Extra officers were added, and by 2 p.m. the wait dropped to 5 minutes. TSA staffing is also up, with 44,800 uniformed officers. TSA says it will have enough officers and overtime dollars to keep all lanes open at peak holiday periods.

And automated lanes, commonly used in Europe and Canada, have just opened at Chicago’s O’Hare Airport and are in the works at other big hubs. They have multiple places for travelers to load trays for X-ray screening, so one slow passenger doesn’t delay everyone. If a suspicious item is detected, bins are automatically diverted to secondary screening instead of stopping the X-ray belt and waiting for an officer. The approach shows a 20-30% boost in throughput.

Classroom discussion questions:

1. What OM approaches is TSA employing?

2. What other ideas do students have to streamline the screening process?

Video Tip and OM in the News: Why the Other Line Always Moves Faster

Queuing (see Business Analytics Module D) is always a popular topic with students–and evidently with readers of the Wall Street Journal (Oct. 7, 2016) as well. This Journal article is a very basic tutorial on the history (Erlang discussion) and logic of waiting line modeling.

The piece writes: “You’ve probably participated in this familiar dance: Given a choice of checkout lines, you’ve somehow picked the slowest. You could wait it out. You could chassé to another queue. Or you could bail out altogether. After all, no one likes to wait. But are the other lines really faster? When parallel lines feed multiple cashiers, you may not be in the slowest one, but chances are, you also are not in the fastest.”

Prof. Bill Hammack, at the U. of Illinois (YouTube’s “Engineer Guy”), explains it like this in his 4-minute video: “Imagine three lines feeding three cash registers. Some shoppers will have more items than others, or there may be a delay for something like a price check. The rate of service in the different lines will tend to vary. If the delays are random, there are six ways three lines could be ordered from fastest to slowest—1-2-3, 1-3-2, 2-1-3, 2-3-1, 3-1-2 or 3-2-1. Any one of the three (including the one you picked) is quickest in only two of the permutations, or one-third of the time.”

Classroom discussion questions:

1. Why doesn’t every service provider use the multiple-server, single line approach?

2. Explain Erlang’s theory.

OM in the News: Queuing Up for Airport Security

Queuing up at the Orlando airport last week, en route to giving a lecture at Tulane University, I encountered a 40+ minute wait for TSA security checks. Returning from New Orleans a few days later, the TSA walk through took less than 2 minutes. Reading the Wall Street Journal’s article “Airport Security Lines Grow” that day (March 3, 2016), leads me to believe this is a good topic for class when you cover Waiting Line Models in Module D. Here is what the Journal writes: “Longer lines are the result of a collision of three changes: reduced staffing from federal budget cuts, a surge in travelers at some airports, and efforts to fix significant screening lapses.”

Current staffing is about 41,000 screeners, below the congressional cap of 42,500, and 5,600 fewer screeners than in 2011. At the same time, TSA has been intentionally slowing down security screening to tighten it up. This comes after some failures to identify weapons and other mistakes during covert government testing.

It is not just Orlando International. Chicago O’Hare has had lines this week snaking through concourses, delaying hundreds of flights. Atlanta has seen peak-time security screening waits of an hour recently because checkpoints are “woefully understaffed.” TSA and airlines have started advising travelers to arrive up to 2 hours before a domestic departure and 3 hours for international flights.

American Airlines says it has had to delay hundreds of flights in January. At Delta’s New York Kennedy terminal, the PreCheck expedited screening line stretched almost out the building door on a recent Friday morning. Airports in Minneapolis, Las Vegas, Denver, Seattle and Miami all say they have seen longer lines at checkpoints over the past two months.

Classroom discussion questions:

What tools in the text could be used to analyze this queuing problem?
Why the vast disparity between Orlando and New Orleans TSA?

OM in the News: Perception vs. Reality in the Supermarket Checkout Queue

Standing in the supermarket queue, you note that other customers are seamlessly drifting forward in their appealingly shorter lines. Should you should stay put, switch lanes, or just head home empty-handed? One research study found that reduction in wait times for express-lane customers didn’t offset the overall increase in wait times for everyone (Five Thirty Eight, Oct. 16, 2014). So would it be better if the supermarket didn’t have an express lane — or, better yet, if it got rid of multiple lines altogether and had all customers join a single long line where there were no winners and losers. Our math in Module D shows that the single line is the best approach. But is it really?

If single lines reduce wait times by so much, why do stores queue us in separate lines for each cashier? One reason is that models overestimate the difference between single and multiple lines because they don’t take into account some human behaviors. Maybe you know who the fastest cashier is. Maybe you switch lanes (“jockeying”) or simply ditch your items and leave (“reneging”). Those behaviors reduce the average wait time in a multiple-line queueing system and bring it a little closer to a single-line system.

In addition, perceptions don’t always match up with reality. The longer we stand in line, the more the gap between perceived and actual wait time grows. By the time we’ve been in line for 5 minutes, we think we’ve been waiting for 10. (See “Why We Buy,” by P. Underhill). So rather than simply shoving us all into one line, supermarkets are exploring three alternatives to reduce both our actual and perceived wait times. First, customers waiting in line can have their items scanned by roaming tellers with hand-held machines, to reduce their service time once they finally reach the cashier. Second, customers can register their place in line, go away, and come back once it’s their turn (like grabbing a ticket from the deli counter). The 3rd strategy: distraction. You can try to entertain customers with videos and in-line merchandising.

Classroom discussion questions:

1. What is the main reason supermarkets do not use a single checkout line?

2. What other alternatives can stores use to speed up the lines?

OM in the News: Wal-Mart’s “Checkout Promise” to Speed Queues

Wal-Mart's "check-out promise" aims to alleviate chronic long lines — Wal-Mart’s “check-out promise” aims to alleviate chronic long lines

My mother-in-law recently commented that she won’t shop at Wal-Mart anymore, primarily because the checkout lines are too long. It turns out she is not alone. The Wall Street Journal (Aug.15, 2014) writes that “to lure more customers this holiday season, Wal-Mart is promising to staff each of its cash registers from the day after Thanksgiving through Christmas during peak shopping times.” The move, called the “checkout promise,” is aimed at addressing my mother-in-law’s very complaint.

“Taking the possibility of waiting in long lines off the table will attract more people into stores,” says the chief merchandising officer. The move comes as Wal-Mart has struggled to win back U.S. shoppers after 7 straight quarters of falling traffic. Many customers have ditched the chain in favor of quicker trips to smaller rivals. The company also has battled with complaints about too many out of stock items and empty shelves. Refilling shelves alone could bring back $3 billion in sales.

Wal-Mart’s supercenters typically have about 30 traditional checkout lanes—giving it more than 100,000 across the U.S.—but the number that are staffed varies throughout the day. It has made aggressive use of technology to cut back on labor costs and more precisely schedule checkout lanes based on real-time demand. But the drop in traffic and customer complaints have forced it to reassess the economics of that approach. After increasing the number of self-checkout systems across its 4,000 U.S. stores, longer lines began forming at its staffed checkouts to deal with customers with more complicated and time-consuming transactions, such as shoppers who use coupons.

The company also recently nixed “Scan & Go,” a program which allowed shoppers to use their mobile phones to scan items as they walked through stores and pay at self-service kiosks, skipping the cashiers’ lines. Wal-Mart said the process was too complicated for customers.

Classroom discussion questions:

1. How has technology complicated Wal-Mart’s queues?

2. What other approaches could the company try to speed up lines?

OM in the News: Can You Help TSA Shorten Security Queues?

In anticipation that more fliers will be eager to pay for expedited checkpoint screening, the Transportation Security Administration has promised to award $15,000 in cash prizes to whomever can design a faster waiting line system, reports Nextgov (July 18, 2014). The competition is on InnoCentive, a website for crowdsourcing solutions to problems, which enables individuals and teams to submit proposals. “There is a guaranteed award,” the contest overview states. “The total payout will be $15,000, with at least one award being no smaller than $5,000.”

The challenge aims to solve expected problems with TSA PreCheck, a program where passengers who undergo a background check and pay $85 get access to fast lanes that don’t require removing shoes, coats, liquids and laptops. “Current queue layouts at TSA Pre✓ airports will need to adapt to support the increasing population of TSA Pre✓ passengers,” the competition states. “TSA is looking for the Next Generation Checkpoint Queue Design Model to apply a scientific and simulation modeling approach to meet queue design and configuration needs of the dynamic security screening environment with TSA Pre✓.” TSA also is asking for approaches that would help speed standard, “free” waiting lines.

Competitors must supply a proposal that considers physical logistics, peak hours and staffing schedules, among other constraints. The “line” extends from the point where a passenger joins the end of the queue to the metal detector or body scan machine. Wait times cannot be more than 5 minutes for PreCheck and 10 minutes for standard lines. Also, the model should enable TSA to apply a “Computer Aided Design drawing to define the physical space available for queuing.”

Competitors are required to provide an animation of a computer screen that shows passengers flowing through lines. It must display real-time reporting during the animation, and allow a user to pause a simulation run when necessary for analysis or evaluation. What a great example of a complex, real-world queuing problem to ask your students to discuss!

Classroom discussion questions:

1. Why is TSA turning to crowdsourcing?

2. What ideas do you have for speeding the lines?

OM in the News: Queuing Up For TSA’s Fast Security Line

The TSA wants to speed up this line. Really.

Fliers gripe that getting through airport security lines can be too slow. Now, it may be fliers who are slow to sign up for a program to speed them through the lines. The Transportation Security Administration is aggressively trying to encourage more people to sign up for TSA Precheck, reports The Wall Street Journal (April 17, 2014).

Precheck, launched in 2011, is much-loved among travelers because they don’t have to take off their shoes and jackets, don’t have to pull liquids and laptops out of baggage, and can walk through metal detectors without a full-body scan. By doing background checks on Precheck enrollees and scanning law-enforcement databases, TSA offers what is essentially pre-9/11 screening to “trusted travelers.”

TSA wants lots more people enrolled in Precheck to make better use of its designated security lanes, which currently number 590 at 118 U.S. airports. “It’s one of the last great bargains the U.S. government is offering,” TSA Administrator John Pistole has joked. To entice travelers into Precheck and test TSA’s ability to handle more people, the agency has been selecting regular passengers to go through Precheck security lanes and get it printed on their boarding passes. Selection is based on criteria like passengers’ travel history and the route being flown. TSA officers trained in behavior detection also can move passengers they deem low risk from regular queues into Precheck lanes.

Pistole said he has heard the complaints about Precheck lanes getting clogged, and TSA has already decided to stop moving travelers 75 years of age and older into Precheck service, unless they are enrolled, because they sometimes can take 10 minutes to move through. “It used to be great, but recently the Precheck lines have been the slowest of all the lines,” said Northeastern University OM Professor Fred Van Bennekom, who has timed TSA lines. “Sometimes there’s almost no one in regular lines and we’re all backed up at Precheck.”

Classroom discussion questions:
1. Why is the TSA using the Precheck program?

2. What else can operations managers do to speed up the security screening process?

OM in the News: The Capacity Dilemma at Apple Stores

Apple's store on London's Regent Street — Apple’s store on London’s Regent Street

“It’s only a 2-hour wait,” writes The Guardian (Oct.18, 2013). “An ordinary Thursday afternoon at Apple’s flagship London store and a long line of customers snakes across the first floor.” The technology brand is used to queues for the launch of its latest must-have product, but these people have come carrying faulty iPhones and malfunctioning laptops, desperate for help from one of Apple’s increasingly hard to reach “Genius” experts.

When it opened in Virginia in 2001, the first Apple store was hailed as a retail revolution, allowing shoppers to play with expensive technology without any sales pressure. The emphasis on service, with blue shirted Geniuses on hand to answer queries and fix broken products, has become almost as important to the Apple brand as the aesthetic appeal of its products. But the whole experience is under pressure as a relatively small number of shops struggle to cope with rapidly growing customer numbers. Apple stores–there are only 415 worldwide–have a turnover of $19 billion and the highest sales per square foot of any retail chain.

The Regent Street outlet, for example, employs 120 Geniuses. Each sees up to 30 customers a day but it is impossible to book an appointment less than a week in advance. If the problem is urgent you can turn up and queue, but it could be a very long wait. The problem is not limited to London. In Apple’s Paris flagship store, there were no appointments available for 10 days. There are even reports of scalpers selling Genius Bar reservations in China. Some argue that Apple needs to take itself further upmarket, so it can serve fewer customers more effectively. Or, to cope with growing demand, should Apple open more shops and come up with clever ways of tackling overcrowding?

Classroom discussion questions:

1. What OM options does Apple have to deal with the capacity constraints?

2. To what extent should Apple handle walk-ins?

OM in the News: Queuing Up at the Drive Thru

Chick-fil-A averages a whopping 6.09 cars in its drive-thru queue

Since the advent of the modern quick-service drive thru, restaurant operations managers have tinkered with the nuts and bolts to create a drive thru that is as fast, efficient, and pleasant as possible. Innovations throughout the years, from wireless headsets and order-confirmation boards to dual lanes and pre-sell signage, have created a better drive thru capable of handling the 60–70% of business that loops the exterior of most fast food restaurants. Accuracy has topped out around 90% for most chains. Cleanliness and favorable exteriors are clearly a major investment in the industry. And order-confirmation boards have become a common tool.

Speed of service, though, has slowed at drive thrus, reports QSR Magazine (Oct. 2013). McDonald’s experienced its slowest average speed of service in the past 15 years, at 189.49 seconds; Chick-fil-A’s speed (203.88 seconds) was its slowest showing since 1998, Krystal’s (217.89) since 1999. The decline in speed is a bad thing, but times are changing. Food isn’t so simple anymore. Burgers and fries have become burrito bowls and customizable salad kits.

That could help explain Chick-fil-A’s speed-of-service. The chicken chain had, on average, more than 6 cars in the drive-thru queue, far and away more than competitors and nearly a full vehicle more than last year. The company is investing more in dual-lane drive thrus to better handle capacity issues. But the move toward more specialty sandwiches is also adding seconds to service time.

Speed is still a top priority at fast food restaurants, and they continue to design systems and procedures accordingly. For example, Wendy’s uses a separate grill and sandwich station in the drive thru to maintain its industry-best speeds, and it trains and cross-trains employees repeatedly to ensure the fastest speeds. Taco John’s is revisiting its drive-thru layout to improve little things that might impede the ability to service the drive thru quickly.

Classroom discussion questions:

1. What tools does OM provide to speed drive thru service?

2. How do the various chains differ in handling drive thru customers?

OM in the News: The $1,000,000,000 Queue

The newspaper in your town probably doesn’t run its lead story about queues and wait times. But this is Orlando, and when Disney World (with its 62,000 local employees) announces a $1 billion program to improve wait times with “interactive queues”, it is the headline (Orlando Sentinel, March 26,2011)! Queues alone consume 10-20% of Disney’s capital budget.

Queues are a delicate balance at all of Disney’s theme parks, especially at Magic Kingdom, which hosts more than 45,000 visitors daily. Guests paying $85 to get in have long complained about the lines as their #1 beef. “Where are you going to put all those bodies? Well, some of them have to be in a queue”, says a UCF prof.

Here are some of the ways Disney’s “Next Generation Experience” project is spending its massive budget: interactive queues at Space Mountain, Winnie the Pooh, Haunted Mansion, and Epcot’s Soarin’. After 40 years of slowly shuffling towards the Haunted Mansion, for example, waiting riders now move through a graveyard filled with elaborate crypts. When you touch the tomb of a composer, instruments carved in the stone play music.

Disney has always paid attention to ride queues, with lavishly themed “pre-shows” that help establish the attraction’s story line. Disney added giant video screens in the Soarin’ queue, equipped with sensors that allows big groups of guests to play collaborative games while waiting. Now it has placed 87 video-game stations in the Space Mountain queue and play areas for kids in the Winnie the Pooh line.

“Guests were willing to wait 12% longer because of the interactive experience”, says a Disney exec. That’s about 7 min. in an hour-long line. (To read our 2 earlier blogs about Disney and queues, click here).

Discussion questions:

1. Disney’s “NextGen” will also let guests book ride times from home and by-pass lines entirely. Is this a good OM idea?

2. Why is Disney willing to spend $1 billion to make its lines more fun?

OM in the News: Disney and the Art of Queuing

On Thanksgiving, I blogged that our family spent a day (mostly in queues) at Disney World, here in Orlando. My report was from the perspective of a customer being entertained while in lines and touring the park.

Now the New York Times (Dec.28, 2010) presents the inside view of the same theme park, but from the underground control room. This nerve center sits below the Cinderella Castle and has made the art of queuing into a science.

“There has been a cultural shift towards impatience–fed by video games and smart phones”, says a park manager. Customers are simply demanding more action. One response: at Space Mountain, 87 game stations now line the queue to keep visitors entertained. Each provides 90 seconds of game challenges.

The operations center monitors all 40 rides at Magic Kingdom, and because of its efforts, the average park visitor can now ride 10 of them (up from 9) in a typical day. For example, if a control center light monitoring the Pirates of the Caribbean ride changes from green to yellow, the operations manager can launch more boats…. or may dispatch Capt. John Sparrow or Goofy to entertain people in line. If Fantasyland is swamped, but Tomorrowland less crowded, the ops center can relocate a miniparade to siphon guests in that direction.

Discussion questions:

1. Why does Disney expend such effort on queue management?

2. What other approaches could be attempted to shorten waits?

OM in the News: Waiting Lines in the Doctor’s Office

My internist of many years, Dr. Gulden, never ceased to amaze me before he retired. For every scheduled appointment, I was seen within 5 minutes of my arrival! This led to research I did in 1994, when I found that the average wait time in doctors’ offices in the US was 20.6 minutes, costing about $15 billion per year in lost productivity.

I guess this topic was of interest since the finding made the front page of papers around the country, from the Boston Globe to the Miami Herald. Yesterday, The Wall Street Journal (Oct.19,2010), with the headline “The Doctor Will See You Eventually“, announced that the “average time patients spend waiting to see a health care professional is now 22 minutes, and some waits stretch for hours”. Are any of us who teach OM shocked?

This is a great article to discuss when you cover waiting line models in Module D. But it may also be useful in Supp.7, Capacity and Constraint Management, because the Journal talks about cutting cycle time. In one doctor’s office, patients helped measure their times from arrival until departure. By identifying bottlenecks, the doctor was able to cut 12 minutes from the typical 40 minute stay.

So why was Dr. Gulden so successful in keeping on-schedule? I think there was one main reason: he made all his staff understand that each patient’s time was as valuable as his was.

Discussion questions:

1. Ask your students to rank the seven methods the article discusses in terms of what they think are the best for time savings payoff.

2. Many hospitals now advertise their ER wait times. What have they done to improve their process flows?

3.What kind of queuing models can be used in a doctor’s office?