Allocating escalator capacity

I have been teaching the MBA core operations class this quarter. This week we just wrapped up talking about bottlenecks and capacity. I consequently found an article in The Guardian rather timely (The tube at a standstill: why TfL stopped people walking up the escalators, Jan 16). TfL in the article title refers to “Transport for London” which runs the Underground. The article reports on an experiment run at their Holborn station.

The experiment in question gets to a bit of escalator etiquette. Specifically, when using an escalator, should people stand to one side so those in a hurry (or in need of a work out) can walk up the escalator or should people patiently stand two abreast? Now if you prefer to chug up the stairs, you clearly lose under the second scenario. However, can it be the case that accommodating the walkers cost the system as a whole an unacceptable amount of capacity?

It’s all very well keeping one side of the escalator clear for people in a rush, but in stations with long, steep walkways, only a small proportion are likely to be willing to climb. In lots of places, with short escalators or minimal congestion, this doesn’t much matter. But a 2002 study of escalator capacity on the Underground found that on machines such as those at Holborn, with a vertical height of 24 metres, only 40% would even contemplate it. By encouraging their preference, TfL effectively halves the capacity of the escalator in question, and creates significantly more crowding below, slowing everyone down.

Read Full Post »

Banning restaurant tips

The big news in the restaurant world this past week is that Danny Meyer, one of New York City’s most prominent restaurateurs, is going to be abolishing tipping at all his establishments. He discussed the move on CBS.

(He also had an interview on CNBC that covered a lot of the same ground plus a few other points that I will mention below.)

Tipping — at restaurants and in hotels — is something we have covered before. As much as I like the idea of linking pay to performance, I think that tipping is a pretty miserable custom. Meyer touches on some of these points in explaining why he is banning the pourboire. But he also highlights a completely different issue: Attracting and retaining talent.

Read Full Post »

Scheduling workers at Starbucks

We have written in the past about some of the challenges of staffing retailer operations. Given competitive markets and an ample supply of labor, many firms have employed staffing models that may be kindly described as aggressive — although some might prefer to call them abusive (see, for example, here, here and here). In essences, firms want to avoid overstaffing but also don’t want customer service to suffer. Employees are caught in the middle of those goals as employers demand more and more flexibility from them.

But to some extent that has been changing. Labor markets have tightened and regulators have begun asking questions. Consequently, firms have backed off some of their more noxious practices (at least in some jurisdictions). Among the leaders here has been Starbucks. Last year it committed  to posting worker’s schedules at least 10 days in advance and to giving workers more consistent schedules. Further it said it would no longer have workers doing “clopenings” — closing the story one night only to have to be there for the opening the following morning. As the New York Times tells it, the transition hasn’t been so smooth (Starbucks Falls Short After Pledging Better Labor Practices, Sep 23).

But Starbucks has fallen short on these promises, according to interviews with five current or recent workers at several locations across the country. Most complained that they often receive their schedules one week or less in advance, and that the schedules vary substantially every few weeks. Two said their stores still practiced clopenings.

Read Full Post »

A little more on running queues in reverse order

A quick update on Wednesday’s post on running queues last-in, first-out. First, the Washington Post had a story on this as well (Researchers have discovered a better way to wait in line, and you’re going to hate it, Sep 9) and to their credit they get the gist of the model right; the fact that customers value getting served early is key to their results.

Second, I was asked to speak about this article on an NPR station out in California (AirTalk, KPCC, Sep 10). You can hear it here.

Read Full Post »

Incentives in scheduling service

When should phone calls go to front-line service personnel and when should they go to a call center? The best arrangement obviously is going to depend on the setting so let’s consider the case of a car dealer considered in a recent Automotive News article (Call center keeps the service bays packed, Jan 23). The dealership in question has two stores — a Honda dealership and an Acura dealership. The status quo had service calls going directly to the service advisers, i.e., the folks who speak to customers when they drop off their cars or call them with news about what problems were found and how much it would cost to fix it. The proposal would route inbound calls looking to schedule appointments etc to a call center instead of the service advisors.

Now, it seems upfront that there are some real benefits of pulling these calls out of the service department. The call center would keep advisors from having to ditch in person customers to take call. It would also allow for some pooling across the stores. However, these efficiency gains are not what sealed the deal for the dealership. Rather, it was the opportunity to gain better control over scheduling.

“We were able to regain control of scheduling appointments in the service drive, and that’s important because we’re only open a certain amount of hours, so we want to load our shop,” says Proctor, managing partner of Metro Honda and Metro Acura in Montclair, Calif. “The service advisers didn’t see it that way.” …

By creating the call center, Proctor took service scheduling away from service advisers. They are often reluctant to book small jobs, such as oil changes and tire rotations, because they earn smaller commissions on those jobs compared with, say, a three-hour brake repair, he says. …

Proctor’s inspiration for the call center came about 21/2 years ago. He was listening to recordings of randomly selected inbound dealership calls, and one especially disturbed him.

“A customer wanted a warranty-repair appointment, and our associate said no appointments were available for three weeks,” Proctor says.

The customer wanted it done sooner. Proctor listened in shock as the service adviser gave the customer a competitor’s phone number to do the work, he says. ..

Read Full Post »

Can changing how passenger act at the gate improve airplane boarding?

How to get people onto planes is an interesting topic. It is a process most of us go through with some regularity and it is hard not to think that there has to be a better way. There are many articles in the popular press explaining that in many ways airlines are doing it wrong (for an example, see this recent Quartz article). Academics like publishing papers on new methodologies that purport to work better — even if their approach is at best whimsical (would an airline assign seats based on who has carry on luggage?). But what if the secret to a smooth boarding process was really in the gate area not in the jet bridge or plane aisle?

That essentially is the premise of some work reported in the Wall Street Journal (In Tests, Scientists Try to Change Behaviors, Jul 28).

At the Copenhagen airport, Dr. Hansen recently deployed a team of three young researchers to mill about a gate in terminal B. The trio was dressed casually in jeans and wore backpacks. They blended in with the passengers, except for the badges they wore displaying airport credentials, and the clipboards and pens they carried to record how the boarding process unfolds. …

The researchers are mapping out gate-seating patterns for a total of about 500 flights. Some early observations: The more people who are standing, the more chaotic boarding tends to be. Copenhagen airport seating areas are designed for groups, even though most travelers come solo or in pairs. Solo flyers like to sit in a corner and put their bag on an adjacent seat. Pairs of travelers tend to perch anywhere as long as they can sit side-by-side.

For the next stage of the project, the airport has given the researchers permission to change seating configurations at some terminal gates to figure out which arrangements are most likely to encourage greater numbers of passengers to sit down and help make the boarding process more orderly. Among possible ideas the team is considering are expanding the number of spots that could encourage single travelers to sit and placing signs with updates about the status of the boarding in key locations.

When people are uncertain about the process, they tend to follow each other, and that can lead to a large group of people clogging up the boarding, Dr. Hansen says.

Read Full Post »

How do smartphones affect restaurant service?

Has the advent of smartphones changed customer behavior in restaurants? According to a piece in PetaPixel, it has and not in a really good way (Restaurant Finds that Smartphone Photos Have Doubled Table Times Since 2004, Jul 14). Here’s the gist of the story, someone at a popular New York City supposedly sat down and looked at security footage from 2004 and 2014 and compared how long customers sat at tables. They measured out how long it took them to peruse the menu, eat their food etc. Here is a sample description of what they found in 2014.

• Customers walk in.
• Customers get seated and is given menus, out of 45 customers 18 requested to be seated elsewhere.
• Before even opening the menu they take their phones out, some are taking photos while others are simply doing something else on their phone (sorry we have no clue what they are doing and do not monitor customer WiFi activity).
• Finally the waiters are walking over to the table to see what the customers would like to order. The majority have not even opened the menu and ask the waiter to wait a bit.
• Customer opens the menu, places their hands holding their phones on top of it and continue doing whatever on their phone.
• Waiter returns to see if they are ready to order or have any questions. The customer asks for more time.
• Finally they are ready to order.
• Total average time from when the customer was seated until they placed their order 21 minutes. [Compared to 8 mins in 2004]

There are similar delays for taking pictures of food or each others over the rest of the meal. The punchline is that they found that the average time a party sat at a table climbed by 50 minutes — from 1:05 to 1:55.

Read Full Post »

If I bring a laptop on my next flight, what does it cost the airline?

It’s been a while since we have posted about airline baggage fees, one of my favorite topics. As I have argued before (see here or here), baggage fees are interesting since they serve as a way to regulate passenger behavior and potentially lower airline costs. Fewer bags means less labor in loading them on and off planes or tracking down lost bags. It also potentially means less weight if passengers actually bring less stuff aboard. But how much is any of that worth? Are we talking about pennies or dollars or thousands of dollars?

The folks at FiveThirtyEight have tried to answer part of that question (Why Budget Airlines Could Soon Charge You to Use the Bathroom, Jun 30). More specifically, they look at the negative impact of adding extra weight to the average flight (or the gain to be had from shedding weight). Here is how they described their methodology.

Our analysis takes into account the distance of a flight, the weight carried onboard the aircraft, and the aircraft type itself. It then simulates every phase of the flight, from departure gate to arrival gate, in order to determine the fuel consumed at each moment along the flight path. To get an idea of how adding small amounts of weight can affect fuel burn on a typical flight, we analyzed a flight from Boston to Denver operated by a Boeing 737-700. Southwest Airlines operates this service three times per day.

According to our model, the total cost of fuel for operating this flight with 122 passengers (85 percent of the maximum seating-capacity) is about $7,900. Each marginal pound onboard the aircraft for this flight will result in a marginal fuel cost of a little less than 5 cents. So if every passenger remembered to go to the bathroom before boarding, shedding an average of 0.2 liters of urine, the airline would save $2.66 in fuel on this flight alone. Such tactics are not off limits. Ryanair famously contemplated charging customers to use the bathroom (in an effort to reduce the number of on-board bathrooms and pack on more seats). Company spokesman Stephen McNamara said in 2010, “By charging for the toilets we are hoping to change passenger behavior so that they use the bathroom before or after the flight.”

Read Full Post »

Do people join a line because it is long?

Have you ever heard of VidCon? Turns out it is, in the words of Wikipedia, “a multi-genre online video convention, held annually in Southern California since 2010. Originally conceived by Hank and John Green of the “Vlogbrothers” YouTube channel, the convention is the largest of its kind in the world, gathering thousands of online video viewers, creators, and industry representatives worldwide.”

My wife and I had never heard of it either until our teenage daughter (who for the sake of this post we’ll call Magenta) put on the full-court press to attend. For better or worse, we caved and today Magenta is Anaheim for the start of the conference with her mom in tow. They have in hand tickets that were purchased months ago. However, they also need to get their IDs for the conference. That was the first order of business today and led to a text I received a little before 9:00AM central time (which, allow me to point out, is not quite 7:00AM in Anaheim):

Just what does the longest line that Magenta’s Mom has ever seen look like? Take a gander (and see if you can spot Magenta).

So why bring this up? Beyond being able to publicly thank Magenta’s Mom for falling on this particular parenting grenade, it serves as a nice lead in to a recent Business Insider article on “Why People Wait In Hours-Long Lines For Shake Shack, Cronuts, And iPhones” (Jun 25).

Read Full Post »

What contract should Uber offer its drivers?

Part of the beauty of Uber is that the payment process is all automated. Once your ride is complete, the firm bills the credit card they have on file, minimizing the time it takes to wrap up your trip; there is no fussing over payments and tips with the driver. But how should the driver be paying Uber? The driver after all is dependent on Uber to match them with riders. Currently, the drivers pay (effectively) by sharing their fares with the company. However, the Economist argues that such an arrangement is inefficient (Pricing the surge, Mar 29).

There is some evidence Uber’s surge pricing is improving taxi markets. The firm says drivers are sensitive to price, so that the temptation to earn more is getting more Uber drivers onto the roads at antisocial hours. In San Francisco the number of private cars for hire has shot up, Uber says. This suggests surge pricing has encouraged the number of taxis to vary with demand, with the market getting bigger during peak hours.

However, the inflexibility of Uber’s matchmaking fee, a fixed 20% of the fare, means that it may fail to optimise the matching of demand and supply. In quiet times, when fares are low, it may work well. Suppose it links lots of potential passengers willing to pay $20 for a journey with drivers happy to travel for $15. A 20% ($4) fee leaves both sides content. But now imagine a Friday night, with punters willing to pay $100 for a ride, and drivers happy to take $90: there should be scope for a deal, but Uber’s $20 fee means such journeys won’t happen.

Despite the revenues a matchmaking fee generates, it may not be Uber’s best strategy. A fixed membership charge is often firms’ best option in two-sided markets. By charging drivers a flat monthly fee Uber would generate revenue without creating a price wedge that gets in the way of matches. Since stumping up cash might put infrequent divers off, they could be offered a cheaper category of membership. Uber should keep its surge pricing in place. But to make the market as big as possible, and really revolutionise taxi travel, it might need to retune its fees.

Read Full Post »