Atlanta is headed for a future of congestion, deadly roads, and inequality. How did we get here?
Atlanta is headed for a future of congestion, deadly roads, and inequality. This is what we can do to avoid it.
MARTA is a radial grid that focuses all of its movement to an empty downtown and lacks the vital circle lines that allow people to get to the activity centers spread around the city. To the vast majority of City of Atlanta residents, MARTA is only useful if you work downtown and the occasional event or jury summons. With More MARTA’s current prioritization, MARTA will continue to offer no relief to Atlanta’s overloaded roads as the city’s population grows by 140% in the next 20 years.
Scooters have the ability to significantly change the way we move around our cities but we will need to cross the chasm of adoption first. Segway-Ninebot (Xiaomi) has the ability to push micromobility over the chasm of adoption sooner than later
We recently sold data to Sonoma Technology to study the health impacts of living and working near high traffic
Sonoma Technology (STI) has selected Citilabs Streetlytics to help investigate the health impacts of living and working around high traffic areas in Southern California. STI is an environmental consulting firm that provides high-quality, innovative, science-based solutions for environmental challenges worldwide, including study design, measurements, analyses, modeling, and software development services. STI is best known for the development and operation of the U.S. EPA’s AirNow Program, which provides the public with easy access to national ambient air quality information using a health-based scale (the AQI) seen everywhere from the Weather Company to the Apple Watch. STI is also known for its work with universities to assess the air pollution and traffic exposure of pregnant mothers, children, and adults in air pollution health effects studies. These research studies contribute to the scientific knowledge base that is used by the US EPA and World Health Organization (WHO) to establish air quality standards and guidelines. Evaluation of the consequences of exposure to traffic-related air pollution on human health has become an increasingly important area of public health research in the last 20 years.
Citilabs combines its understanding of transportation simulation with mobile phone data and traffic counts to create Streetlytics. Streetlytics is the measurement of hourly vehicle and pedestrian volumes, speeds and demographics on every segment of road in the US, Puerto Rico, Vancouver and Toronto. Streetlytics also includes the origin-destination pairs for every vehicular trip and the turn by turn path in-between.
How is Streetlytics being used
Until recently, most air pollution health assessments were conducted using only data from regional air monitoring networks, which have sensors located 10 to 30 miles apart in urban areas. Selected studies that included traffic assessment mostly relied on proximity to roadways and roadway density as traffic exposure indicators. These studies ignored the huge variation in vehicle activity on nearby roadways. Studies that incorporated traffic volumes had to rely on annual average data traffic counts which were only available for major roads. Because local motor vehicle activity and emissions are very often the primary factors explaining air pollution variation within communities, most exposure assessments failed to capture the local-scale granularity of pollution within each community. With Streetlytics’ hourly traffic volumes and vehicle speeds on all size roads, and with HERE’s accurate roadway geometry, STI is able to apply air quality dispersion models to more accurately estimate the motor vehicle emissions’ contributions to air quality at the community scale. With this approach, concentrations of traffic-related pollutants such as NOx, CO, PM2.5, elemental carbon, organic carbon, and trace metals are estimated at the residences, schools, and work places of health study participants to more accurately characterize their differences in exposures on the relevant time-scales (weekly, monthly, yearly). The rich spatial and temporal coverage of Streetlytics’s traffic data provided the key to much more accurate maps of community air quality levels.
STI’s community level air quality modeling is being used to evaluate the health effects of living and working near high traffic areas. The current and proposed studies involve the effect of vehicle emission exposure on the cardiovascular, respiratory, neurological, and birth outcomes.
STI is using Streetlytics volumes, speeds, and road network in modeling exposure in Southern California for NIH-funded research with the University of Southern California’s Keck School of Medicine on:
Maternal And Developmental Risks from Environmental and Social Stressors, which is a 5-year study designed to untangle the causes of childhood obesity in low income, urban minority communities.
Lifecourse Approach to Developmental Repercussions of Environmental Agents on Metabolic and Respiratory Health (LA DREAMERs) Study which is part of the NIH program on Environmental influences on Child Health Outcomes (ECHO). This study involves subjects from prenatal to age 40 for the study of metabolic and respiratory outcomes.
The Interstate Highway System was one of America's most revolutionary infrastructure projects. Unfortunately, highways sliced through the center of many American cities destroying neighborhoods and brought concentrated pollution. Highways did allow Americans to live further from their jobs in the city center but as more people moved to the suburbs, lanes were added to highways to increase capacity and reduce congestion.
Today, Americans have repopulated urban centers and the neighborhoods divided by highways. More Americans are living next to highly traveled roads than ever before. All metro areas across the country are expected to grow even more over the next few decades.
America recognized negative effects of vehicular travel a couple of decades after the highway system was built. The Clean Air Act was passed to combat these negative effects and since then:
New passenger vehicles are 98% cleaner for most tailpipe pollutants compared to the 1960s
Fuels are cleaner. Lead has been eliminated and sulfur levels are more than 90% lower than they were prior to regulation
American cities have much improved air quality, despite ever increasing population and increasing vehicle miles traveled
EPA Standards have sparked technology innovation from industry
Invisible is still dangerous
An MIT study in 2013 found emissions from road transportation are the most significant contributors to air pollution. Vehicular emissions caused 53,000 premature deaths in 2013, almost double the number of people killed in traffic crashes that year.
Scientific studies, like the ones conducted by Sonoma Technology, show that some pollutants can harm public health and welfare even at very low levels. As a result of these studies, the EPA has progressively lowered light and heavy-duty vehicle emissions limits. For the last 10 years, California has provided large financial incentives for public and private owners to replace older polluting cars and trucks with modern, clean technology vehicles.
Vehicular traffic causes particularly elevated risks to public health in communities near large roadways. As the traffic increases, vehicle emissions flow linearly in to nearby neighborhoods. Pollution is greatest on the road and diminishes with distance from the road. Public health officials have long warned that traffic pollution can drift well over 1,000 feet from traffic and more recent research suggests that it may drift more than a mile.
What is emitted
Vehicles emit pollutants such as NOx, CO, PM2.5, elemental carbon, organic carbon, and trace metals. These pollutants cause asthma, cancer, heart attacks, strokes, reduced lung function, and pre-term births. Recent research has added more health risks to the list, including childhood obesity, autism and dementia.
Today’s cars emit 98% less pollution per mile driven than they did in 1960. Electric cars and alternative fuels will continue to help, but the sheer number of cars on the roads offset these improvements. People can help by driving less. Combine trips, walk, bike, scoot, carpool or use public transportation.
Avoid homes and work places within 1,000 feet of any major road
Avoid opening windows and use air filtration with MERV 13+ rating during times with moderate and high pollution levels
Find physical barriers like sound walls and vegetation
Avoid exercising near traffic
Avoid driving on major roads for long periods of time and always recirculate air in moderate and heavy traffic
Avoid truck routes
Don’t count on electric cars to solve the problem (it will take a long time and electric vehicles still emit air pollutants from brake wear, tire wear, and road dust resuspension).
Today, Streetlytics is used by governments to build safer roads and by insurance companies to create more robust pricing models and acquire new customers. This blog explains how governments and insurance companies are using Streetlytics to reduce risk and increase safety.
Vehicles that spend less time on the road have a lower risk of being involved in an accident. Citilabs works with insurers to identify neighborhoods with low average Vehicle Miles Traveled (VMT). VMT is calculated to understand the average daily distance traveled, as well as the weekday and weekend VMT. Understanding the driving habits of a neighborhood helps insurers more accurately price a current customer and market to new ones.
Traffic Accident Exposure
A range of factors influence risk of traffic accidents. Until recently, these factors have been limited to information about the drivers involved in a collision and the permanent attributes of the road like lighting, signage, and lane marking. Today, Streetlytics helps governments and insurers understand the other people using the road when a crash happens.
Streetlytics understands the hourly average directional movement on every block of US road. This includes:
Number of vehicles on the road (and turning)
Number of pedestrians
Driver and pedestrian home locations
Driver and pedestrian demographics
How long each vehicle has been traveling
How much further each vehicle is going
How often drivers and pedestrians pass by this location
The other people using the road can influence the risk of traffic accidents. Streetlytics helps governments and insurers to identify other potentially risky locations.
Not all trips are equally safe. Streetlytics understands the average turn by turn path of every vehicular trip on US roads. Governments and insurers can identify neighborhoods with high risk trips by combining a neighborhood’s vehicular paths with a map of high-risk locations. This combination creates an understanding of the risk level for traffic accidents for any household in America.
Location Health Risk
People who spend significant amounts of time near high volume roads are at risk for multiple health conditions due to poor air quality. Streetlytics helps governments and insurers understand locations with potentially poor air quality due to high traffic volumes.
Location Safety Risk
Roads with a high percentage of people who do not live nearby see decreased safety. Streetlytics helps insurers understand how many people on the road in front of a specific location are cutting through rather than living in that neighborhood. This information can help the insurer understand the property damage risk due to vehicles from the road.
Change over time
Mobility will see significant changes over the next few years. These changes must be measured to be understood. Streetlytics measures the average week of movement every month. Comparing monthly measurements can show the changes in seasonality and the gradual change over time.
Environmental exposure for automobile crashes is vital to understanding who is on the road around every incident.
LEVs provide a solution to a limited transportation system. 2019 will be the beginning of a wave of new LEV manufactures and forms that will last for years to come. Location will be extremely important for which LEV forms become popular, how they are used and how they develop. The rise of LEV use will lead to an unbundling of the use of the personal automobile followed by a change in transportation infrastructure. Just like mobile phones, there is an opportunity for multiple winners over the next few years.
Types of Location Data
At Hugecity, we initially tried to start a ‘things to do’ company based on where people were right now using cellular location. In 2010, cellular location data proved to be prohibitively expensive for our free social media startup so we were pivoted. Today, is a completely different story. The internet of things has created a deluge of location data exhaust. Many of these companies believe they can create new revenue streams by selling their data while other companies aggregate multiple sources together for resale. The number of companies selling location data combined with an uneducated market has created a race to the bottom price.
Types of Location Data
This chart details the four major sources of consumer level location data – apps, always on apps, cellular location, and indoor positioning systems. All four vary in terms of precision (ranging from 5 meters to 100 meters) and persistence (pings per user). Precision is good for confirming that a person is in a specific place while persistence is good for understanding patterns of movement.
Indoor positioning systems are primarily used in retail to understand how people move within an enclosed environment. These WiFi/beacon systems are used by most large retail companies to capture the movement of everyone who walk through the door.
Mobile phone applications share a user’s location with digital advertising markets to allow companies to better target advertisements to an app’s users while they are in specific locations. A user’s location is shared with these markets only while the user is using that specific app. Most downloaded apps are rarely used which makes it difficult for an app to share multiple locations from a specific user. This limits the use most location data from apps to targeted advertising.
A select few mobile phone applications are only useful if users constantly share their location. These always on apps cover a variety of categories like safety, lifestyle, dating, travel, e-commerce, navigation, and weather. Multiple pings per user per day are capable of showing a user’s pattern of movement. When this data is aggregated and anonymized with other always on apps it is possible to understand population patterns of movement. Always on apps have both precision and persistence but limited sample size compared with the other technologies.
Similarly, cellular location data is also good to understand population movement. Cellular location data comes from mobile carriers with millions of users by constantly triangulating their users’ location between cell towers. What cellular location lacks in precision it makes up in sample size. Cellular location data cannot understand which specific stores are most visited, but it is precise enough to understand how neighborhoods move.
Location intelligence is the process of using location data to derive meaningful insight to solve a particular problem. It is possible to layer the location data on a map to identify relationships between different sets of geospatial data. For example, retailers combine their transactional data with their indoor positioning data to understand the how many people enter each store, what they buy, where they live, and their demographics.
This understanding is limited to the entrance of the store. To help retailers go beyond indoor positioning systems, some location data providers derive insights and analytics from their sampled movement outside of a location.
Unfortunately, only studying a single technology sample leaves the sample prone to bias. Sampling bias occurs when a sample is collected in such a way that some members of the intended population are less likely to be included than others.
Understanding total population movement is possible using movement data from related but different sources. Citilabs’ Streetlytics uses a combination of phone movement data, traffic counts and travel demand modeling used in 2,500 cities around the world. The millions of observations from all three of these complementary data sources are input into a complex mathematical process to create the most complete, accurate and consistent national mobility data set ever created. This combination allows Streetlytics to confidently measure the average total population movement on all US roads every month.
It is necessary to match the best location intelligence to its specific use case. If the intended use is understanding movements within a specific store, indoor positioning systems are best. If the intended use is advertising to people who are in a specific store, general app location data is best. If the intended use is understanding the changes in mobility, urban development, out of home advertising, auto risk or anything else with the total population, then Streetlytics is the best.
Citilabs has built the world’s leading transportation modeling and simulation software, Cube, used in 2,500 cities around the world. In 2017, Citilabs combined its understanding of transportation simulation with mobile phone data and traffic counts to create Streetlytics. Streetlytics is the measurement of hourly vehicle and pedestrian volumes, speeds and demographics on every segment of road in the US. Streetlytics also includes the origin destination pairs for every vehicular trip and the turn by turn path in-between. Since 2017, Streetlytics has become the data that powers audience location measurement for Geopath, the nonprofit auditing organization for the US out-of-home media industry.
The $8B American billboard industry relies on understanding the total audience exposed to each of their billboards. This is a specific need which requires a specific type of location data. Location data on the movement of people can be gathered using four different technologies. Data from each of these technologies is aggregated and anonymized to protect the public’s privacy.
Fully autonomous vehicles (SAE Level 5 AVs) have the potential to disrupt city, region, and state transportation and land use plans. Citilabs recommends for all models that predict transportation beyond 2030 to include AVs. Including AVs in regional models will enable the simulation of multiple scenarios of AV adoption. This blog discusses potential changes caused by AVs and provides tips on how to incorporate possible changes into existing models.
Traffic counts are useless if they are old or not directly in front of the specified location. Streetlytics provides average annual and seasonal daily vehicular and pedestrian traffic for every address in the US, Puerto Rico, Vancouver and Toronto (All of Canada available in 2019). This is updated every month with hourly weekday, Friday, Saturday, and Sunday counts. These counts are directional and show how many people turn right and turn left at every intersection.
99 pages of applications, thousands of pages of supporting documents, $10,861 in fees and a lawyer, 9 days off work for appointments, 1,824 days later, and counting ... my wife is still not a permanent resident nor a citizen of the United States.
If MARTA truly wants to stick to its stated vision of providing public transit to boost economic development and enhance the lives of Atlantans, they should use our money to install light rail around the entire BeltLine loop and it should be done now!
This blog is about the devices and their connections that I have experimented with for home efficiency and security.
Citilabs’ Streetlytics combines transportation demand modeling with cellular and app GPS measurement to understand volume, speed, home locations, origins and destinations for every section of the road for every hour of the day.
Who was affected by the 85 bridge collapse
I mapped as many approved and proposed projects as I could find for the city of Atlanta.
I believe Facebook can rival Google Maps and help local news with an extremely powerful discovery tool created by expanding Events by Facebook to include places, articles, and things for sale.
I discuss the content that can be gathered from the Facebook login and the difficulties in doing so. I hypothesize a third party service that handles this difficulty for companies interested in tapping into the largest social network. I brainstorm a few uses of this content and I am curious to hear additional uses of this service.
My attempt to describe my understanding of event discovery and the fundamental difficulties presented by events as a content type.
My story of the blood sweat and tears that went into building Hugecity.