What is mobility data?
Transportation data, and the analysis of that data, is necessary in order to keep the flow of traffic moving smoothly, keep the community safe, and help make transportation more accessible to more people, among other benefits.
No one knows that better than the people who work within or support the transportation industry – a.k.a. you.
“Mobility data,” which has traditionally meant any kind of data related to transportation, has been collected and used forever by many types of organizations, but lately, the term is trending in a new way. It’s time to establish a modern definition.
What is mobility?
To get to the heart of the definition of mobility data, we first need to define what we mean by “mobility” within the transportation industry. Traditionally, in a broad sense, according to the Oxford dictionary,“mobility” means “the ability to move or be moved freely and easily.”
But these days, especially within any industry that is involved in improving or learning from transportation, it has come to also mean how people navigate the world around them. Mobility, in a transportation sense, can refer to:
- Bicycles, scooters, and skateboards
- Motorcycles, vehicles, and trucks
- Boats, ferries, buses, trains, and airplanes
Mobility also refers to the streets, sidewalks, bridges, and other constructed paths that we use to get from point A to point B. It can also include a contextual layer; for example, is the vehicle or bicycle owned or rented, for play or work?
Our expanded definition: Mobility is how people traverse the world.
What is mobility data?
From the expanded definition of mobility, it’s easy to infer the meaning of “mobility data” to be the facts derived from how people traverse the world. But in reality, it means so much more than that.
Mobility data is contextually enhanced (or enriched) telematics data. It means gathering and keeping track of how people move around places like points of interest (POI) or neighborhoods. This data is aggregated and anonymized – information is collected in a way that doesn’t identify individuals, ensuring privacy.
The intent of mobility data is to help us, analysts and strategists, derive meaningful insights to achieve a certain goal or to innovate. This transportation data helps lead us to answers to such complex questions as:
- Who is utilizing the transportation network and how?
- What types of transportation are people selecting in different types of scenarios?
- When do people choose different types of transportation methods?
- Where are people going; what are the patterns?
- How does a person interact with the transportation network around them?
- How does a person get from point A to B to C and so on? Were multiple transportation types used?
- Why did a person choose to go where they went and/or in the manner that they traveled?
Let’s break that definition of mobility data down a little bit more. What do we mean by telematics and contextual enrichment?
Telematics includes driving behavior, location, mode of transportation, and device data. We’ve been collecting telematics data for decades.
Contextual enrichment refers to the where, when, and how of transportation data. It can include adding a layer of temporal facts, weather, speed, moving direction and geospatial constructs. For example, GPS data means nothing without a map to give it context. This is a fairly new construct.
Here’s one simplified example of what we mean by contextual. The telematics data from a mobile device may indicate that someone appears to be driving, but if they are also at an airport and on a runway, the takeaway may be that they are in a taxiing airplane.
Contextual information is not only additional layers of data; it may also include observations, theories, or models, which are scenarios or use cases in the form of algorithms that crunch telematics data into useful statistics, either on the individual or the aggregate level. This is where the analysis comes in.
Mobility data is only valuable when we set it up to be valuable.
We are taking telematics data and enriching it with contextual data to create mobility data signatures that help us analysts offer companies insight to their areas of focus, whether that be insurance, city planning, or advertising attribution, to help them reach their goals.
First, we – that is, the people who are experienced in the world of mobility and the products and services around it – must determine which questions we’re trying to answer and to what end. In other words, what are we trying to achieve?
From there, we can gather the telematics data and determine how it must be enriched or enhanced to become valuable and provide the information that leads to insights and answers.
Essentially, we are taking telematics data and enriching it with contextual data to create mobility data signatures that help us analysts offer companies insight to their areas of focus, whether that be insurance, city planning, or advertising attribution, to help them reach their goals.
Telematics provides the “what.” Mobility data, when set up appropriately, offers the “why.” Analysts then can translate that mobility data into meaningful and actionable insights.
Mobility data best practices and gold standards
Once we all agree that understanding the context of the data and how it links to the real world and we are all on the same page about the definition, we need to consider standardization to help ensure quality and consistency, especially as we share data.
What might be the constructs of mobility data standardization? Here are a few of the questions we ask ourselves at Arity:
- How much transportation data do we need to collect to be statistically relevant? Each use case may be different. For example, the percentage of drivers on a particular stretch of freeway may be higher or lower depending on the volume of traffic in that area. Imagine a busy highway in Los Angeles compared to Route 66 in Arizona, for example.
- How is the data being derived and what is the level of accuracy of that collection point? For example, cell phone data, from tower to tower, is different than GPS data.
- What are the inherent errors of the data source, and how do we enrich it to help eliminate these errors? For example, mobile phone GPS: does the data show that the vehicle stayed on the road or did it appear to go through buildings?
- What is the definition of a “trip”? We all talk about the number of “trips,” an individual or group might take. Is a trip confined by any limitations, such as time, distance, or transportation type?
- How frequently are we collecting the information? Should frequency be normalized, or should it be determined case by case?
As an industry, we are all still learning, testing, and evolving mobility data best practices.
We are on the cutting edge of shaping what mobility data is and how it should best be developed and leveraged.
Mobility data benefits
We are only scratching the surface of what’s possible with mobility data. What we do know is that mobility data helps us:
- Understand how people move so that we can develop better, more useful products. For example, with a reservation on OpenTable, mobility data can determine whether someone is running late and notify the restaurant, change a reservation, or provide alternatives automatically, if someone chooses that option.
- Manage the business, especially when transportation is an important component of that business.
- Understand, predict, and manage risk. Telematics is a better element for prediction than demographics alone.
- Develop better value propositions based on personal points of interest.
Life will be easier, work will be more productive, and we’ll all work together more seamlessly when we agree on one definition of mobility data; that is, contextually enriched telematics data.
In the next chapter of our mobility data guide, we’ll cover the shift from leveraging only location and device data to mobility data, and how this is truly a historically memorable moment in time.