Coffee with a Researcher 2: Shared Mobility Data and Usage Trends in the U.S. with a Deep-Dive Analysis of Chicago (Text Version)

This is a text version of the video for Coffee with a Researcher 2: Shared Mobility Data and Usage Trends in the U.S. with a Deep-Dive Analysis of Chicago presented on Feb. 23, 2023.

Lauren Reichelt: Thanks for joining– oh, recording in progress. Thanks for joining us for the second webinar in this EEMS' Coffee With a Researcher webinar series. This series is part of a concerted effort to better connect Department of Energy EEMS Research with Clean Cities Coalitions and technology integration, more broadly.
We really want to open up lines of communication between coalitions, between your stakeholders, and researchers working on EEMS efforts through DoE. So in addition to this webinar series, we are planning to send Clean Cities' directors to the annual merit review this year to gather insights, to better understand the opportunities and the challenges that coalitions face around EEMS projects.
If you are interested, as a coalition director, in being a part of that group that attends AMR on behalf of EEMS, please reach out to me. We don't have a plan fully fleshed out yet, but we are in the process, and I am looking for volunteers. We are also developing educational materials and a Clean Cities University course on EEMS topics for coalition staff, and a brochure about EEMS that coalitions can provide to their local stakeholders. And all of those materials should be published and available for coalition use fairly soon. Next slide, please.
And then, of course, there's this EEMS Coffee With a Researcher webinar series. So this is the second session of a 5-part webinar series. These sessions will highlight available EEMS tools and insights, and help coordinators identify local and regional partners and projects that might benefit most from these capabilities. So we want these sessions to be really conversational, and provide an opportunity for EEMS researchers to ask for input from coalitions, and for coalitions to engage with EEMS. Clean Cities Coalitions are in a really unique position to describe the mobility priorities and the challenges that your own communities face. And researchers can use those insights to enhance EEMS research as well.
We will, as Cass mentioned, have ample time for discussion. So please collect your questions throughout the presentation, suggestions, thoughts, ideas for how this might apply to your work, or even how this work could be improved in future iterations to be more applicable to community needs. You can put comments in the chat. You can raise your hand. We will, like I said, have plenty of time to discuss later. Next slide.
So we want to start off with a question to get a feel for who's in the room and what you're interested in. So I'm going to ask that you put your responses in the chat to this question. So why are you interested in shared mobility data and usage trends? Are there specific angles there that you're really interested in? Do you have projects that this might apply to? Are you looking for transferable lessons? Really interested in what you all are interested in.

All right, who's going to be brave and share? There we go. Looking for an overview. Trying to figure out how their coalition might be supportive of local EEMS projects. Last mile use cases. Shared mobility charging, or corrals and electric vehicle charging. So aligning shared mobility with EVSC, EV infrastructure. What's actually being used versus the hype. I think Joann can share a lot on that topic, actually.
Yeah, these are great. If others have other reasons they're here, please keep sharing in the chat. This is really helpful. And it can help Joann make sure she's sort of touching on the points that you're all looking at. So with that, I should probably introduce Joann.
Dr. Yan, or Joann Zhou, is a principal analyst and Interim Director of the System Assessment Center at Argonne National Laboratory. She leads several Department of Energy and industry-sponsored research projects on energy and emission impacts, analysis of alternative fuel and vehicle technologies, electric vehicle charging behavior, infrastructure modeling, and transportation energy burden analysis.
Dr. Zhou is a member of the Transportation Research board's Alternative Transportation Fuel and Technologies Committee. She received her master's and PhD in transportation engineering from Clemson University in South Carolina. And with that, I will pass it off, so that we can learn a little bit about the work that she's been doing.

JOANN ZHOU: Great. Thanks, Lauren for the introduction. You can hear me, right?

Lauren Reichelt: Yes

JOANN ZHOU: Good. I'll start sharing the slides. And thanks, everyone, for spending the time today. As we talked in the beginning that this is supposed to be a conversational, not just one-way communication, so please feel free to stop me any time for questions, or feel free to put it in the chat. And I'll also try to pause during my presentation just to take questions. I cannot see the chat directly. But, hopefully, the host or my co-presenter, Dave Gilkey, can help me with the questions, if any. We put it in the partition mode as well.
So today's topic, we're here to present the shared mobility data and a usage trend in the US. In a base of the data we have collected and we did some analysis of Chicago, which is one of the cities we found out as a very good data on the different issue of mobility technologies. So this is a project founded by Vehicle Technology Office of the DoE. This is a part of the data program ended with your analysis program, so it's a data project.
So with that said, the original objective really was to summarize the data availability, what is out there, what kind of information is out there, how accessible, available to us or to researchers or decision-makers. So that was the original purpose anyways. The data we collected, we found out that, yeah, we could do some analysis to address certain questions. And during that time, we can also talk about the pros and cons of the data availability out there.
So we had a report published, feel free to read it, which has much more details. But today, we just give you some highlights of our research result and findings. So in, general, I see a lot of chats that talk about why you're interested in mobility data and usage trend. For us, for DoE, we originally were thinking whether the data is available for us to quantify the energy emission or economic impact of those technology deployment, right? Whether that can give us some kind of guidance when we electrify them or replace certain type of private vehicle usage.
However, we found out in general that the data available in the public domain is quite limited to support that kind of purpose. And the troop-level information, like who take what trip, and from where to where, for what purpose, those information, critical information, are missing. So it's really challenging for us to do energy or emission analysis because you don't really know what vehicle was taken for which trip, for what purpose, and who took it.
So then we thought about what we can do with the available information. So then we found out that, yeah, we could do analysis like, for example, what are the general usage trend in a new mobility technology. And here, we include not only in the TNC, the Transportation Network Company such as Uber, Lyft. We also include shared bikes and scooters, so those are the three shared mobility technology that are covered in this presentation.
So we want to see what are the general trend over time, over different regions. We also want to understand what other demographic factors contribute to the ridership and the usage. We want to see how equitable the current distribution of the shared mobility infrastructure and the resource, especially for the bike, right? Because some of the docked bike do rely on some kind of infrastructure, using a bike lane or the stations, to support their usage.
And also, for example, going back to the bike usage, what are the effect of the low stress versus the high stress bike lanes near transit station would affect the usage. If people have used the bike– a shared bike, you would have notice that some cities have some kind of dedicated lane. some city don't have that. So those certainly played a factor in the usage trend.
So again, this is the typical questions we found out that we can answer, to some extent, with the data available in the public domain. That indicator were important as a factor that– critical thing is, that we need a more shared mobility usage data that are more detailed information in order to support in-depth analysis, especially when you want to look at energy, emission, or economic impact.
So we with that, I will just report some of our results. Again, feel free to let me know your questions, and the host please, [INAUDIBLE] to stop me because I don't see the chat now. So let's start with the data availability. So we went through all the data in the public domain that we can find.
We not only summarized for this three shared mobility technology, which city had what data available for the public that you can easily assess, or you can assess based on a request. We also went through the database to see what kind of data is available from them.
Basically, the detailed summary include that whether they have where, when, what, and who information. Again, this is helpful for us to understand any kind of impact of those technology, usage trends, and also energy emission implications. So where is basically the location information, which sensor tract is located, what is the community this is located, or where the trip was started.
The when is the time, right? What time of the day, what time of the year. And also what, like what type of vehicle is used for that. And what type of device, for example, a scooter and a bike was used for that trip. And the who means that what kind of a people used that.
So you can see that here is a very simple aggregated table to show what information is available for which city at a trip level or an aggregate level. But we have a very detailed table in our appendix on the back end of our report to document for each of the city, each of them are the shared mobility technology.
Whether they have where, when, what, and who information for what time period, so those are important. As you can see from some of my example results later that they are important for us to understand the impact and also guide us to– for future deployment projects.
So some of the general trend. Here I'm using the TNC as the example. But in our report, we also examined the scooter usage by city, by year, and also the bike share by year and by city. So here I'm showing two charts. Both are the TNC, so that's Uber on the left side of the technology.
So the TNC actually grow in all the cities that we have our data public available. And so they grow from 2015 to 2019, in general. However, they all decreased in 2020, especially in a big city like New York City, Chicago, and San Francisco. But it started to slowly come back, starting from 2021 and to now, to the present.
There's a wide range of per capita ridership in those cities. In some of the cities, they could be in less than five ride per capita per year, such as Austin. But in some cities, like especially those cities that attract a lot of tourists, in general, like San Francisco, DC, or Boston, the downtown of those cities, they could be nearly 100 rides per year per capita. So those has the most ridership, especially in the downtown area. But we realize, on those, a lot of ridership may not be taken by the residents there, more by the tourists visiting the city.
We also found out that, on the right side, that chart is showing the TNC growth is highly correlated with the number of vehicles on the road. So this chart showing you the number of drivers on the horizon and also the annual ridership per capita on the y-axis. So each color stands for different city, and New York City is the right-ish bubble and the Chicago is the bluish bubble. And the bubble size stand for their population size. So that's why you can see that New York and Chicago has the biggest bubble, especially in New York City.
And so when the driver– does more driver exist in a system that you can see, in general, there's more ridership there as well. However, you also see the trend I was talking about earlier that in 2021, the COVID-19 hit. The trend, the ridership decrease. Using New York City as an example, I labeled the year on each of the bubble that you can see a very steady increase from 2015 to 2019.
But suddenly, it jumped to that two faded bubble which one stand for 2020, the others standard for 2021. That you can see the ridership decreased a lot, even go back to the 2016 level. A similar trend you can find for that blue dots, which speak for– stand for Chicago and that two faded bubble, same thing for 2020. And 2019– 2020 and 2021, that go back even close to 2015 level of the ridership in Chicago City.
So we see a similar trend in scooters and bike share. Although the decrease is less than the TNC. It seems like the COVID hit them, the impact is less compared to the TNC ridership. And we have that result documented in our report. We later on in our project, we found out that Chicago– besides Chicago and New York City, which are the two city that we found out has the best available public available information on their website.
We also found out that California has the data available, for example, TNC ridership, upon request. Because the California Public Utility commissions require Uber and Lyft to share that information. And that information is available upon request to them. So we got this information from California in the later phase of our project. We also connected ridership with per capita, the population density per sensor tract and the zip code for California.
So here, you can see that the red color shows the more ridership, and the purple shows less ridership. And the gray area means that either we couldn't find data for that, or there's no zip code or sensor tract information associated with that area.
So as you can see that San Francisco, Los Angeles, and San Diego has a high usage per capita, TNC ridership. Some of them could be hit– some sensor tract in the downtown area could hit like 100– more than a hundred rides per year per capita. But some other area that could be low, that close to 0, or less than 10 rides per year.
So San Francisco and also– is the number one spot for the ridership, followed by Los Angeles and San Diego in California. Should I stop? Any questions so far? Or should I just move on?
LAUREN REICHELT: I don't see any questions in the chat. But if anyone wants to jump in with any questions at this point, it's a good break. All right, if not, yeah, let's keep going.
JOANN ZHOU: Yeah, some general trend that we mentioned a report I just want to point out here. So for Chicago and New York City that we found out, in average, there's about 1,600 rides per year per driver. So yes, usage trend like this is reported. And if you're interested to see in a city this big, like what is the typical size of rides that a driver would take on an annual basis, and all on a per capita basis, there's information available in our report. Hopefully, that helps you to plan for the future projects.
So, as mentioned. that Chicago and New York City has the best shared mobility usage information. Although, there's still limitations there, but compared to other city, they provide the most information available, and they regularly updating their database for public to assess. So with the data we have, especially for Chicago, we analyze some of the usage trend by demographic factors.
So if you know the city of Chicago, then the map here is showing the city next to the Michigan Lake, and the lake is on the East side of the city. And then the city, northern side or the middle part, that has a higher income. So that's showing on the chart on the right side. So the darker color showing the sensor tract where it's relatively higher household annual income. And the lighter color shows the less annual incomes.
And the left side, we put all the three mobility data usage on the same map, that including the TNC and the bike and the scooters. Scooter is only available in the pilot area. The data is only available in the pilot area. But even with that limitation, you could still see that the high-income communities they have a high mobility– shared mobility usage across the three modes. And regardless the good accessibility to the public transit lines.
As you can see, the red lines here that shows the metro lines in the city. So those area with higher income, they also have a very good access to the transit line. But still they also show a very high usage of TNC, bike, and scooters. And the lower income area has a lower usage across the three modes.
So we also look into that to see whether the household or vehicle ownership play a factor in that as well, besides the income. So this is another way to look at the same information than when we put the sensor tract income, so all the sensor tract into different income brackets, OK. So this is average annual household income, and all the sensor tract will be grouped into different brackets. And now we look at the TNC per capita that is started from the sensor tract.
So one general trend, again, is high-income communities, they have a much higher mobility usage. This is example for TNC, but we found a similar trend for scooters and bike share. So they will have a higher usage with TNC. And within each of the income brackets, the household own less vehicle has much higher TNC usage.
So this is across the board effect no matter what income bracket you're looking at. But in the higher income group, you see the most significant difference of the household that own less than 1 vehicle in average compared to the household own at least one vehicle. So that means those households are using TNC as a travel mode, a important travel mode for them.
And remember. In earlier charts show those communities. They also have a very good access to the public transit systems. But they still use the TNC as one important travel mode. Often we got a question and people ask this question like, what is the correlation relationship between the TNC usage and the transit ridership, TNC taking away the transit ridership.
So here, I'm not going to read the chart or read all the words here. I know it's kind of a complicated figure for people to understand a short time. But what we are trying to see here is– trying to show here is the correlation between the transit ridership, which is on the vertical line, with the TNC ridership, which is on the x-axis, the horizontal line.
And each color here shows a different year. So starting from 2015 to 2019, before the COVID hit, when the data become abnormal. And within each of the color band, each dot shows a month. So then you will see there will be 12 dots in each of the color band.
So we look at a general trend annually from 2015 to 2019. So the correlation between the transit ridership and TNC ridership looks like it's a negative. Means there's more TNC ridership and the less public transit ridership, if you look at this black line going a five years trend. So it looks like that the TNC is taking away the transit ridership in Chicago. And we see a little bit of similar trend in New York City as well.
However, when we look at each year, so that's when you go to each year of the color band, and especially look at each dot which represent the month data, and you start to see that there's actually a positive correlation between the transit and TNC ridership, especially for Chicago is a cold city that in the spring, summer, and the early fall time that you start to see people going out, do more travel.
And there's actually a positive connection between the TNC and the public transit. So that means that we could promote more transit ridership by providing the TNC as the first mile, last mile mode to connect people with the public transit. And that actually, in particular, true for low-income communities that I will show a result in the next slides.
So, again, the correlation between TNC ridership and the transit ridership is not necessarily a negative relationship. It's actually depends on the time scale. And that actually time scale means that in the time that people need to go out, spring and summer time, they like to take TNC and the public transit at the same time, there's opportunity there that we can promote growth of both.
So seasonal variation really show played a positive relationship here between those two. So what so we look at the census data of the taxi ridership in Chicago so this is going back to what I said about that we particularly found low-income areas using TNC as a first-mile last-mile mode.
Because when we look at the TNC ridership per capita in Chicago by census tract, remember, the high-income area is the center of the city and also the northern part of the city, so the darker color shows more ridership. So that's more higher-income, more ridership in Chicago. And here is the airport. So that's why you see also high TNC ridership in that sensor tract because that's the people go in the airport and come out of airport.
However, we do see there's true sensor tract in the low-income community also have a relatively high, or even very high ridership– TNC ridership. And then we look at those trip starting time and also trip end the time, that we found out in general that they have a high arriving trip in the morning, and a high leaving– departure trip in the afternoon.
So that indicates– and this is at the end of the metro line, Red Line. And that kind of indicates that local community– that low-income community is likely using the TNC as the first-mile last-mile mode in order to get to the transit stations.
So because this is the line that goes into the deep south of Chicago, and that this looks like an only line goes very south of Chicago. And you can see that people go into using the TNC in order to go to that sensor tract in the morning time to get on transit. And in afternoon time, departure from the transit lines.
So switching the gear. So earlier slides was about TNC, and switching a gear to the bike share. So we also did a similar study to see how the bike ridership changed by demographic factors, such as income. The animation here, you can see how the bike station and the bike ridership evolve in Chicago from 2013 to 2020.
If you pay attention to the 2018 year– oh, sorry. 2020. In August 2020, the city made a significant effort to extend the bike stations to a large area of Southern Chicago, where the low-income community sits. So when you see the animation jump to 2020, you will see, yeah, suddenly, it cover a large area in the South.
Again, each dot indicate one bike station, and the color shows the bike ridership. So, again, we put an income in a public transit line map here that you can correlate to see where those communities are, so the blue circle here. And then we also look at the ridership and indicate them using the number of dockless E-bike trips.
And however, we found out, even though the bike infrastructure, the bike stations, is available and extended to those area, we still see low bike usage in the low-income community. So if you know the Chicago, you know this is the downtown here. And there's the University of Chicago and this large area of the Southern Chicago here, that you see that even though you have the dots here which is the bike station. However, the usage is still low compared to a lot of other areas in the city.
So again, this is a data available by the end of 2021. And we need to continue to monitor the usage trend, and to see how that would change over time. But when we spoke with some of the people who know the communities there, One reason they told us is that the crime also– the local crime also played a factor there that prevent the people to use the shared bike even in the daytime.
So that shows that a different community have a different problems or concerns when it comes to transportation accessibility. Providing the infrastructure or the vehicle available to them may not be the only thing that we need to do in order to improve the accessibility, improve the transportation convenience.
So that's something to think about, of course, when you design a shared mobility program. Work with your local stakeholders to understand how we can help the local community in order to promote the shared mobility usage.
So I alluded earlier that even the station available, the ridership is still low in low-income area. However, the chart I showed earlier only– that the city made effort basically in August 2020, and this is still a little bit fairly new. So we have a two-year data. We need to continue to monitor whether those low-income communities continue to show either low ridership or start to take off in terms of bike ridership.
So we had another side to dig into that kind of perspective. So we look at all the bike stations in Chicago, and put it into the same timeline, the years since the bike station opened. Here, I say that year since the first ride from a station means, OK, when the station opened. And it has the first people taking a ride.
So basically you can understand is when we look at all the bike stations in Chicago as the time since it opened. So and then each of the line, the different color shows the income, the different income of the community where the station was located at.
So as you can see that a high-income, which is a darker blue, sensor track, they have accumulated ridership increase faster, much faster than the sensor tract has a lower income. So lower income, for example, those two light lines they– even up to eight years since the station opened, they still have a relatively low ridership compared to the sensor tract has a high income.
So this is going back to the point earlier that, in general, the low-income area has a much less ridership across the mode. And even takes years for them still stay at a relatively lower level. However, when we did a regression analysis to look at other factor, other demographic factors that may correlated with ridership, we found out that actually population density, college student percentage, employment rates, those has a higher correlation. Meaning, they are more important factors affecting the bike ridership than income.
You could argue that the sensor tract has a high employment density or college graduates, they tend to be higher income to start with. So there's definitely this indication there. But what are we trying to do is to see whether we can see which sensor tract or which community that need more bike lanes or bike stations.
So when you deploy a project in the city, you want to understand which sensor track actually meet this, need the bike station, or need a bike lane to promote the usage. And that's what are we trying to do, use the data we collected. So first, we have to understand how many different bike lane, the bike infrastructure is there. So we kind of arbitrarily separate our bike lane types into low traffic stress and the high traffic stress.
So there's a four we grouped in as a lower traffic stress, that the bike has a very separate space, the rider has a very separate space to ride their bikes. And when it come to high traffic stress, means the rider– the bike rider has to compete with the vehicle, the models, model vehicles for their space. So there's no separation, either physically or marked space, like a buffered bike lane that there's a marked space to separate them.
That is basically in the high stress traffic bike lanes, you basically only have a one line to separate them. So you physically compete the space with other vehicles. So this– is important for us to show that later analysis is because the bike lane structure played a factor in the bike usage. And the bike lane infrastructure we separate into two, low traffic stress and high traffic stress.
So this map shows you how the different bike lane available in Chicago. Again, this is going back to what I showed in the last slide. So there's a different type of bike lanes, some of the low traffic stress, some high traffic stress. And this is how they lay out in the inner city.
And if you have ever used a bike share in the city, you may know that along the Lake Shore Drive, there is very separated bike lane on the lake for bike riders to use. And that is considered is a low-traffic stress bike lanes. So going back to the general usage first. So in general, we see the bike share bike usage increased from 2014 to 2019.
The COVID, again reduced the ridership. However, only a fact about 10% of the decrease. So in general, there's, in 2021, which is now shown in the chart here, there's over 4 million riders rise in the city of Chicago. And tens of millions of dollars into the new bike lane dedicated by the city of Chicago.
So again, which sensor track has a more or less bike lane availability based on the factors we identified that are important? So this is the map showing you, again consider all the factor I mentioned, either the bike infrastructure, whether it is a high stress or low stress, the employment density, the quality percentage, since all the factors that we identified could affect the bike lane– sorry, bike usage.
And we apply the regression analysis trying to see whether the sensor track has more bike lane available than needed. Or it has less bike lane available as needed. So we do want you– deploy our investment, our future project focus on the community that need more bike lanes than they deserve than they currently have. Sorry.
So the map here shows that the light color indicates that area has a freer bike lane present and predicted, means those are the places that they can use more bike lane infrastructure, because of the current demographic factor, current usage. And the sensor track, again on the head as a darker color means the opposite.
This means that probably there's already efficient– sufficient bike lane available for the current usage and the current demographic factors. So this kind of inform the decision in the future that when you try to deploy a project, for example, a bike lane infrastructure, which community that need a more attention?
LAUREN REICHELT: We do have a few questions in the chat. So some of these were from like 15 minutes ago, but we'll just start from there. So in your analysis, if someone lives in an entertainment or tourist area, where there's lots of shared mobility activity, are you assigning shared mobility rides in that area to the residents or to the visitors actually using it? So how do you distinguish?
JOANN ZHOU: Yeah, so that's a very good question, right? So we only know the rider trip information, where to start, when it ends. So when we do it anonymously, for example, the rides per capita, we only can consider the population live in that sensor tract. We don't know how many visitors are tourists visiting that sensor tract on annual basis.
So it could be a bias or even underrepresented because we recognize those riders could be tourists or visitors coming from different sensor tract or even different cities. So that's going back to the data limitation, as I said. We don't know who took the trip. Those information is just not available in the public domain.
DAVE GILKEY: Joann, if I may note. This is Dave Gilkey. [INAUDIBLE]
JOANN ZHOU: Yeah, go ahead
DAVE GILKEY: –the co-author of the story. And I'll drop this into the chat. We did a regression analysis, or at least for the bike share, to determine what might be driving this. And we included employment density and– kind of as an explicit variable there.
We did find, for the bike share, that the highest use areas are downtown, where there's a high employment and population density, and near the University of Chicago. And the single most frequented one is at Navy Pier, which is a tourist attraction here in town. So there definitely is a correlation there.
But that's really– we did want to see what we can do with a regression analysis to say what else can we tease out of it. And we do find that both employment density and population density matter in different ways. But when we're really focused on equity, then we don't really know who, as Joann just noted, is riding the bike or taking the system. So we're going based on what that population is. If it's a heavily commercial area, maybe the results are slightly less relevant for that particular access.
LAUREN REICHELT: Thomas, did you have any follow-up questions there? Feel free to jump in.
SPEAKER 3: I do not. Thank you.
LAUREN REICHELT: Yeah. And then Thomas had one other question, too. So I'll just read that for you. Have you considered studying if and to what extent there's a correlation between neighborhood crime rates and shared mobility usage? Is crime a strong predictive variable? It's a really interesting question.
JOANN ZHOU: Yeah, it's definitely interesting question. That was not a part of our study scope. And yeah, definitely. From what we hear from the local community that the low usage, low bike usage in Southern Chicago, even with the available bike stations there, is because of the crime rate. So I would assume in cities like this. In sensor tract like that, that could be a strong indicator. But that wasn't part of our study. And we didn't have that data available back then.
LAUREN REICHELT: And then, Eric asked, how did you define access routes as a low stress bike way?
JOANN ZHOU: Yeah, see. Oh. Dave, you want to take this? My understanding was we consider .25– like quarter miles buffering range close to the bike lane is the route that is low stress or high stress based on this more subjective definition.
DAVE GILKEY: And specifically for this, if I recall correctly from the raw data, the access routes are basically the on-ramps to the off-street trails, if I recall. They're not even listed here in this little graphic put together by the city because those are kind of almost a rounding error in the grand scheme of things.
But looking at the– there is one that is kind of debatable, as far as our assignment. We call neighborhood greenways as a little bit higher stress for our analysis. These really could go either way, in our opinion. Talking to the city, they say that these are meant to be kind of side-street thoroughfares. I could definitely, though, as a mediocre bike rider, see that they could be more stressful than just the standard street, even if you do have your kind of designated lane, because you are going against oncoming traffic.
So while they're easy enough, once you're used to them, maybe a little more difficult. So we decided to mark that as a higher traffic stress for our analysis. It doesn't change the overall analysis much. It's really about if you have those unprotected versus protected or separated bike lanes there. That's the bigger driver.
LAUREN REICHELT: Thanks, Dave. That is all the questions we had in the chat, so I'll let you I'll let you finish your presentation.
JOANN ZHOU: I think that was the end. This is really the summary. Again, the report we have available summarize all the data availability we have found on the public domain. And the more importantly, is what kind of data is available. It's not just a yes or no. It's whether you have the when, where, how, and what type of information. So those, hopefully, that could be useful for people who want to dig into the data availability, see what is out there.
And with the data available, we found out the city of Chicago and New York City has the most– the best available information and also annual– I mean, regularly updated. And those are two big cities that you can take a look at. You see the general usage trend compared to other smaller-sized cities.
And also how, in Chicago, those usage trend varies by demographic factors. Some of the takeaway is, yeah, not a surprise. High-income areas, even though they have very good public transit accessibility, they still use all the shared mobility modes much more than– much higher usage than the low-income community.
However, the low-income community are using TNC as the first-mile last-mile mode. And they could indicate action can be taken to promote such a usage to connect them with the public transit. Either you can incentivize the usage or provide more drivers available in the traffic peak time. Encourage driver to go there during the traffic hours to connect those people to the transit lines.
And also the bike lane study, just to try to use the example to inform the decision-maker to see where you need more infrastructure. How equitable your current infrastructure is distributed in your city? What actions can be taken to improve that and promote more usage?
The crime is one thing. But I know we didn't have a data back then to do the analysis. And also sometime is– it is a longer term problem for us to tackle. Yeah, that's basically the summary of the presentation. Thank you.
LAUREN REICHELT: Thank you so much, Joann. We have a few more minutes. If anyone has additional questions or thoughts, anything they'd like to share.
THOMAS LAMONTE: I have a quick question. Why do you think E-bikes will make a difference? And I'm curious, since e-bikes face higher charges, will the higher cost of e-bike usage in the shared mobility mode cancel out the benefits that come from e-bikes, in terms of more extended range and perhaps easier travel?
JOANN ZHOU: Yeah, that's a good point. I think, just in Chicago, the bike is the dockless bike, bike share. So that dockless bike doesn't limit people to get at a bike or return the bike, especially when you return. So that we see that that's a way to improve the ridership. Dave, you want to add anything here?
DAVE GILKEY: Yeah, the dockless, of course, was the main point to highlight there. There's also accessibility in terms of who might be able to use it. I don't know that– basically, when you have the e-bike, there's less concern about you need to be an expert rider to be able to ride with traffic. So there could be a little bit the other way, where people are a little hesitant to ride it for concerns that it's too much for them. But it does, potentially, open up it to a broader populace as well. But really it's the dockless is the main driver.
And then, related to the cost of it, there are different programs that exist in different towns. I don't have a comprehensive database of these. But that do account for– basically, account for income in the cost of the ridership for members, either reduced membership costs or kind of free rides for e-bikes, or reduced fares for those.
So that could work as well as something that a city or a bikeshare operator could handle in order to improve equity there. That could be true for conventional bikes, of course, as well. But that is one route to kind of cancel out the effect that you just suggested.
LAUREN REICHELT: OK, any other questions for our presenters?
JESSE DREYFUS: Hi, I have a question. Sorry, I joined in late, as well. But I don't know if you already covered some of the legislation going on some of the cities that are piloting this kind of shared space on the sidewalks, where they're using mixed-use. Basically, the point is they're trying to reduce the number of cars and try to increase bicycle traffic. Does that factor into any research?
JOANN ZHOU: I don't think the traffic was considered as a factor in our regression analysis. I would use the population employment percentage. And I think that's already positive correlated with the traffic. Even though we didn't include that in the– yeah. But also, the bike lane infrastructure speak to the traffic stress, as well. So even though it could have been a high-traffic location, but it is a buffered lane that we still consider as a low traffic stress infrastructure.
JESSE DREYFUS: OK. This may be less about the traffic, but more about the idea of using the existing streets they have. Repurposing them for pop-up vending, for example. And it kind of ties into the e-bikes, the cargo e-bikes specifically.
JOANN ZHOU: Yeah. I think this title is another question that Chat that I– what the city of Chicago asking the bike share locations. I do see some insights here, but I forgot exactly what they consider when we had this conversation with them. I think they certainly need to consider a lot of zoning requirements and a lot of demographic factors affecting them to put a bike lane available or not in certain areas. But again, yeah, I don't have the answers to those questions.
JESSE DREYFUS: Thank you.
LAUREN REICHELT: Thanks, Jesse. And is Sam still on? Because I wonder if Sam could speak to this? Sam, do you have a response to what role the city of Chicago plays in the location of bike share docks? Put you on the spot?
SAMANTHA BINGHAM: Yeah, so that's not my department. No, just kidding.
[LAUGHTER]
So my boss, Sean, is responsible for micromobility, and although we have a vendor, we have a contractor that operates our system, we are very much engaged in where the locations of stations go. And so there is a whole plan that takes place. There's a planning exercise that takes place when we do expansions.
How those stations are distributed with the lightweight docks. With the e-bike share, right, that is primarily how we are expanding the system. And so it's something I think, as Joann mentioned, that's fairly new to us. But a lot of that is. It is new. I don't think we have any best practices, so far, on what we've learned. Yeah, I hope I answered the question.
JESSE DREYFUS: I understand. It's still kind of an early, emerging, I guess, field of study. But the basic point was, it seems like there's, not so much here in the US, but other parts of the world, they're experimenting the idea of the digital curb, for example, in lieu of parking spaces on the street next to commercial businesses, for example. And so that they would use that as a place for food trucks, or for a loading zone, or for paratransit access. And they would basically dedicate a bike lane and that space. Instead of having four lanes of traffic, they would put on a two-lanes for example.
LAUREN REICHELT: And what I will say is, Jesse, you should attend our EEMS Coffee with the Researchers session on curb space management, because there may be some more clarity there, or maybe not. But definitely more people to have that discussion with who have a different perspective there.
JESSE DREYFUS: We'll do. Thank you.
LAUREN REICHELT: Yeah, and then I think, Eric, I think we did get a chance to answer your question that came in the chat. And it is the top of the hour. So I want to thank our presenters, and thank everyone who attended today. This will be recorded and posted on the Clean Cities website. If you have any additional questions for our presenters today, please reach out to me, or reach out to them. And we can make the connections. Thank you all.