Data can help inform understanding of the potential impacts and mitigating strategies for potential changes to Metrorail’s span of service.
Metrorail ridership after 10 PM on Sundays, by month, since August 2010. The black line is the average, the yellow band is one standard deviation +/- average. The dashed line is the maximum.
In June of 2016, Metro began closing at midnight on weekend nights to allow more time for track work over the weekends, in support of the SafeTrack program. The Metro GM/CEO has recently announced he will pursue making this service change permanent. To meet the necessary track work requirements to get the system in a state of good repair – recommendations which the FTA and others have also made – the current budget proposal for FY 18 includes various additional early closure options, including ending Metrorail service at 10 PM on Sundays.
Staff from across the agency are currently working to assess the impact of such closures on customers and determine what strategies we might employ to replace some or all of that rail service with alternatives.
Similar to our data release related to SafeTrack, we are glad to provide ridership data to assist with analyses by our local partners and members of the transit advocacy community.
First, average, standard deviation and maximum ridership after midnight on Friday and Saturday, and after 10 PM on Sunday by month, station, and hour, from 2010 to 2016. (Excel, 13 MB) The graph above illustrates one dimension of this data: the average, usual range and maximum system ridership on Sundays by month since August 2010. The biggest spike there is during Obama’s second inauguration weekend. There is a lot to learn from this data set.
Secondly, we’ve put together Metrorail entries by half hour by day type. The full dataset is available for download, but the relevant data is presented in the image below, showing the half-hour segments that have the lowest ridership.
We have also assembled additional visualizations of ridership during the potentially impacted periods: Read more…
We want your feedback on a new online tool, Metrobus Explorer, which allows visualization of Metrobus service frequency and geography.
One of the biggest challenges facing bus transit is making the service extremely easy to understand. Metrorail stations are filled with customer information, including system maps, fare and travel time tables, station-ahead lists, and passenger information display systems (PIDS) screens, leaving little guesswork for the savvy traveler. Moreover, with limited real estate available for customer information, Metrobus stops are often at a disadvantage. While Metro continues to improve bus stops around the region — including the design and installation of new diagrammatic bus system maps — information technology is also playing an important role in filling the bus customer information gap, including BusETA, information displays, and trip planning sites and apps.
Metro’s Office of Planning is developing a new online tool called Metrobus Explorer that is geared to answer two questions about the Metrobus network: “How often do buses arrive at a given stop (or set of stops), and where do they go from there?”
Screen shot of the bus frequency and spider map tool. Click the image to access the live tool online.
Metro Data Day 1 will bring together Metro staff, the app developer community, riders and advocates.
Metro has some great data feeds, and app developers — from Google and Apple to Metro Hero — are consuming them to provide great tools for transit riders in the Washington DC region.
That’s great news, and yet there are opportunities to do even more! For example, we’d love to see app developers help customers plan an accessible trip – one that routes a user to station entrances where elevators are present. And this isn’t likely the only unmet need from the Metro transit rider community. There are loads of great ways to make this data more useful to you. That’s where you come in. Read more…
If you are great with data and love cities and transit, we have a job for you.
We were excited to announce that a job listing for a Business Intelligence Analyst position within Planning’s Applied Planning Intelligence unit has just been posted. We are looking for a healthy overlap between a data scientist and a transit nerd. For the full job description, head over to the wmata.com/careers site, scroll down and click on View all jobs. A short description is posted below.
Currently a team of two, we work to convert Metro’s many data sources into information that can be used to inform plans, policies and procedures. Many of our projects have been featured here on PlanItMetro, including:
Customers are saying great things about SelectPass, Metro’s new unlimited monthly pass program.
Word Cloud representing what SelectPass customers have said about the new pass program. Click for a larger version.
This post was submitted by Metro’s Director of Customer Research.
Metro’s limited time offer SelectPass has early adopters talking. In April, Metro sought to provide new payment methods by introducing SelectPass—a multi-tiered pass option allowing customers to ride as much as they want on Metrorail (and Metrobus). Currently two price points are available. Read more…
We have developed a customer impact analysis that shows how and where customers may be impacted in the SafeTrack safety surges, to help guide regional partners plan mitigation and alternatives
The first SafeTrack project began on June 4, meaning that Monday, June 6th, was the first day with peak-period service disruptions. This first safety surge is to accommodate the track improvement project planned for the Orange and Silver lines between East Falls Church and Ballston. With only one track available for revenue service, we are cutting back to just over 3 trains per hour on each line through the work zone. When service levels decrease without a decrease in demand, we see an increase in passenger loading on rail cars. We measure that with a metric called passengers per car, or PPC.
Our rail cars are designed to comfortably transport around 100 passengers each, with most sitting and a few customers standing. After special events (or any given weekday on some lines) we often find rail cars with much higher passenger loads. From a planning perspective, an average PPC of greater than 120 is considered crowded. Also, we know that customers don’t evenly distribute themselves across rail cars, so an average PPC of 120 means some cars are much more crowded. Read more…
The new Metro SelectPass is structured to to maximize pilot participation while minimizing the risks. Making that happen involves overcommitting to truth in advertising – and we’re fine with that!
The two fare levels most likely to be popular for the SelectPass are $2.25 and $3.75.
We are excited about the launch of the new SelectPass pilot. As we have begun to roll out this new pass product, we are listening to your questions (via twitter, comments posted to articles, etc.) and we hope to address as many of them through the proper venues. PlanItMetro seems to be the best forum to answer the persistent question, “Is this really only for two fare levels, and why don’t you tell everyone that they can probably save money?”
Testing the capacity of the Fare System
When we roll out new features, we want to eliminate as many risks as possible before committing to them. In this case, the primary risk Metro faces is that our aging fare technology might not be able to accommodate a very different fare product such as SelectPass. So we developed a program to test the pass at two individual “levels” as a proof of concept and not push any limits of our fare collection technology. Read more…
New data download features rail ridership by origin, destination, day of week, and quarter-hour intervals.
Subset of the visualization made by BioNrd aka Mike from our October 2014 data download data.
As you’ve probably noticed, it’s been a while since we’ve released a fresh batch of Metrorail ridership data. Continuing the spirit of openness, we have recently uploaded data from October 2015 in CSV format. (The number of rows is too great for Microsoft Excel).
This new dataset includes day of week data, so you can begin to investigate impacts of evolving workplace policies such as compressed work schedules. You can also compare it to October 2014.
In the past, we have seen a lot of innovative analyses of the data we share. Perhaps the best so far was a visualization of Metrorail station entries and exits by station by “BioNrd” aka “Mike.” What else can we learn from this dataset?
Customers showed high levels of interest in a customizable monthly pass.
Metro customer interest in a new unlimited monthly pass concept, by market segment.*
Metro is not raising fares this year, and instead is innovating ways to make it easier and more affordable to use the system. Metro is taking a page from private industry, which has moved away from charging customers for each purchase and towards giving customers the option to “subscribe” to a company in exchange for unlimited access. A Netflix subscription has replaced a membership at the local video store. Amazon Prime offers unlimited shipping rather than shipping on each item. Spotify subscriptions have replaced purchasing individual CDs. Why not a subscription to use Metro?
Fortunately, we found a way to provide this to our customers and we’re really excited to begin testing it out starting this month. The idea is to allow customers the ability to customize an unlimited access pass based on their usual travel patterns. Modeled after Seattle’s Puget Pass and frequently discussed on Greater Greater Washington over the past few years, this pass would allow customers to subscribe to a monthly pass, priced based on their typical trip costs, that offers unlimited travel on rail and the option to add on the same flexibility on bus. We are calling it the Metro SelectPass.
Here’s the basic concept. Customers tell Metro their usual start points and end points. We then figure out how much that trip costs and offer you unlimited travel on rail up to that value in exchange for you buying 18 days worth of trips. For example, if a customer’s “usual” peak trip is $2.25, they can get a pass priced at $81.00 (about $2.25 x 18 x 2) and then all trips valued at $2.25 or less would be free for an entire calendar month. Extra trips for lunch, a night on the town, doctor’s appointment – it’s all included in one low price. If you travel on a more expensive trip for any reason, you only pay the difference for that trip. Most customers may enjoy savings of over 20% off of the pay-as-you-go rate, and they’ll also get the benefit of knowing they can travel as much as they want, whenever they want, all for one price.
For an additional $45 per month, customers can choose to add unlimited bus travel on top of unlimited rail travel. That’s a huge savings compared to pay-as-you-go! Read more…
Adopted from queueing theory, this new method of assessing delay on transit systems with tap-in-tap-out fare systems accounts for natural variations in customer behavior.
As you may have heard, Metro is testing out a new customer-oriented travel time performance indicator. Many departments here at Metro have been collaborating on this effort. Metro has decided to initially pilot a measure where we define delay as anything greater than train run time, a headway, and the 1-3 minutes it takes to travel from the faregates to the platform. However, as we began our research into customer travel time, we got to asking the question, “How do we define customer delay on the rail system?”
As we quickly learned when digging into the data, on good days with no delay on the rail system, there is still a wide variety of “normal” customer travel times. Some variation in travel time is because customers arrive at random to the origin station, but all leave the destination station more or less at once. Additional factors influencing this variation include walking speed, use of elevator vs. stairs, escalator or elevator outages, and customers with suitcases and strollers.
We could start with a threshold for “on time” but by definition we know on a good day there were no rail delays so we would be counting slower customers as “late.”
Additionally, on a day when we know a disruption has occurred, we might count very quick customers as “on time” when in fact we know that everyone experienced some delay.
So we set to determine a method for calculating delay that accommodated for the natural variation in customer speeds. These travel time curves started reminding me of delay calculations from queueing theory from grad school. Read more…