Traffic can make your bus ride longer, but a lot of the wait on buses comes from the bus waiting for long lines of people to pay one by one. WMATA is exploring a partnership with a private company or investor to install machines near bus stops to cut down on these delays.

Line to board the X2. Image from WMATA.

A bus can spend 10-20 seconds at a bus stop with few passengers getting on or off, but this time known as “dwell time,” can stretch to be several minutes when a long line of people wants to get on, when people are trying to find exact change, or when they run into problems paying.

The S buses on 16th Street, for example, spend about 22% of the time sitting at the stops with the doors open. People paying contributes to a lot of this time.

Here are the nationwide average times it takes for a passenger to pay the fare in different ways on a bus.

SituationSeconds to board

uncrowded low-floor buses

No fare payment1.75
Visual inspection (passes, mobile phones)2.00
Smart card taps2.75
Ticket or token into farebox3.00
Mechanical ticket validation4.00
Exact change into farebox4.50
Magnetic stripe card5.00
Data from TCRP Report 165 via WMATA.

High-floor buses, where you have to climb multiple steps to get on, require about another ½ second per rider. Most Metrobuses are still high-floor, but the agency is replacing the buses over time with low-floor ones. Also, if there are people standing in the bus, each passenger takes about another ½ second.

Also, for many riders, the most convenient time to add money to a SmarTrip is on the bus. Metro estimates this transaction takes an average of 20 seconds. The above numbers are also national; Metro’s internal rule of thumb is six seconds to pay with cash.

These delays frustrate riders, but they also cost a lot of money since paying the bus drivers is the biggest cost of running bus service. When bus routes take longer than the schedule shows, local governments can add more buses and drivers, which adds to the cost. Or, they can cut service, which pushes riders away and cuts down on revenue from fares (not to mention making transit worse).

Kiosks on New York’s Nostrand Avenue Select Bus Service line. Image from WMATA.

Put loading machines at top bus stops

Metro could speed up boarding considerably if riders could buy SmarTrip cards or load them up with fares at the bus stop. This would still mean boarding would take about 3¼ seconds per rider, but that’s a lot better than 5-6 seconds for paying with change or 29 for loading the SmarTrip.

It costs money to put machines at the bus stops, of course, but starting with the highest-ridership bus stops would yield the most bang for the buck. Therefore, working with DDOT, Metro put together a plan to put 260 kiosks at bus stops on the five highest-traffic bus lines: 14th Street (the 50s line), 16th Street (the S line), Georgia Avenue (70s), Wisconsin and Pennsylvania (30s), and H Street/Benning Road (X line).

These five groups of lines carry over 100,000 passengers a day. (Buses in the top 20 corridors carry half of all Metrobus passengers). Just having machines at ten stops per line could save over $2 million a year in direct operating costs while also making the bus more useful to all riders.

The 16th Street, 14th Street, and Georgia Avenue lines. Red dots show the ten highest-traffic stops which could get kiosks.

Here’s a breakdown of the benefit per line:

Corridor Avg. daily riders Avg. daily cash loads Minutes

saved/day

Weekday cost savings Annual cost savings
14th Street 15,635 1,886 44 $1,070 $355,700
16th Street 19,938 1,298 20 $526 $174,909
H/Benning 14,764 3,284 78 $1,964 $653,138
Georgia Ave 21,880 3,517 56 $2,665 $886,040
Wisc./Penn. 20,306 2,744 43 $1,687 $561,059

WMATA doesn’t have a lot of money lying around right now, but if this analysis is anywhere close to correct, the cost of installing the machines should more than pay for itself. Therefore, WMATA is exploring a public-private partnership where a private partner would install and maintain the machines in exchange for a guaranteed payment out of the money saved on lower bus operation costs, or an “infrastructure bank” investor to underwrite Metro doing the work.

WMATA has been talking with the Federal City Council, an association of business and professional leaders who work together on DC’s big challenges. The group is is working to create an infrastructure trust that could help support this project.

According to an analysis by WMATA planners, each kiosk should cost about $27,500, plus about $50/square foot to the local government for the sidewalk space it will take up. The capital cost would range from about $924,000 to outfit the H Street/Benning Road corridor to $2.7 million on Wisconsin and Pennsylvania.

The H Street/Benning Road line. Red dots show the ten highest-traffic stops which could get kiosks.

Add in about $2 million in software costs and all five corridors would cost about $11.2 million to outfit. But it would save $2.6 million a year, paying back the investment in about seven years, the report estimates.

Plus, these numbers don’t include the value to riders of having a bus which doesn’t take as long to travel.

Are there any drawbacks to this plan?

The kiosks will take up sidewalk space. In some places, there is plenty of room; in others, it might be hard to find the space. Already, in some places in dense areas, the bus shelters cut down on room to walk; the kiosks probably could at least go in line with bus shelters, but still, there could be some negative impacts here.

Rendering of a potential kiosk on 16th Street in DC. Image from WMATA.

Also, this plan only reduces time for people who are paying with cash today. Many other people pay with SmarTrips, and this doesn’t speed up their boarding.

Bus boarding could get much, much faster if people could board without paying, especially if they could use the rear doors as well as front. But (unless the buses are actually free) this would require having machines at more stops, machines which print out receipts, and inspectors to periodically check fares.

Already, there is a pretty high rate of fare evasion on Metrobus, and the drivers can only do so much to enforce payment without putting themselves at risk. How much would it cost to have inspectors? Would fining or arresting people for not paying slow the buses down even more? This report doesn’t look at that alternative.

Moving ahead on WMATA’s SmartTrip loading kiosk plan now doesn’t foreclose a wholly different system, but doing both would just cost a little more.

DC is considering rear-door boarding as one of many strategies in its study of the 16th Street buses. That study should wrap up this year, and that will give a much clearer picture about whether any other fare payment approaches are likely to happen soon.

Update: PlanItMetro has now posted the detailed business plan for their proposal.

David Alpert created Greater Greater Washington in 2008 and was its executive director until 2020. He formerly worked in tech and has lived in the Boston, San Francisco Bay, and New York metro areas in addition to Washington, DC. He lives with his wife and two children in Dupont Circle.