Skip to content

Mobility and Data Analytics: Data Sources for Accurate Analysis

In the field of transport planning, good baseline data is key to the success of any project. Whether applied to a traffic study, a demand analysis, for the development of a transport model or simply to make a good initial diagnosis of mobility, baseline data is the basis for our project to become a reliable and solid tool with which to make strategic decisions.

At the same time, the collection of baseline data has historically been one of the biggest headaches for transport planners. They represent a considerable cost of the total value of the project and it is often difficult to find reliable and robust data sources that tell us in a comprehensive way what the mobility of our study area is like.

In recent years, innovation and advances in Big Data have fully entered society, and the world of transport and mobility has been one of the great beneficiaries of this. Nowadays, knowing what people’s mobility is like through forms such as tracking mobile phone signals, GPS data from our devices, tracking our purchases with a bank card or the comments we post on our social networks while we travel, has meant a revolution in transport planning. There are many companies that, aware of the usefulness of this data for its application in this type of analysis, have been commercialising this type of data for some years now, flooding the market with multiple options that sometimes make end users not know which is the best source of data to apply in their studies.

To shed some light on all this, we tell you about the main new (and not so new) data sources to understand and analyse mobility. Spoiler: none of them is the best or the worst, they all have advantages and limitations and their application will depend on the needs of your study.

There are no better and worse sources of data, it will depend on several factors whether we should choose one or the other in our studies.

CAMERAS AND OTHER DEVICES
There are several types of cameras on the market, all of which are focused on vehicle surveys. Let’s take a look at the main ones:

  • Artificial vision cameras. This consists of the temporary installation of high-tech cameras, whose videos are subsequently processed with artificial vision software that is capable of automatically counting the directional count of vehicles and their vehicle classification.
  • What data is obtained with them?
    Vehicle gauging in road sections, with vehicle classification, i.e. vehicle counting.
    Short-range vehicle origin-destination (O/D) matrices, i.e. movements at roundabouts.

    • Advantages
      • Practically all vehicles are gauged during the time the camera is installed.
      • Non-intrusive technology.
    • Limitations
      • Recording time limited to the duration of the battery or power supply (between 24 hours and 7 days).
      • Possibility of the device being vandalised while installed.
  • Radars. They allow the collection of vehicle gauges, allowing data to be taken on one-way and two-way roads. These devices usually incorporate a data management programme to automatically calculate results reports.
  • What data is obtained with them?
      • Vehicle gauging in road sections, with vehicle classification.
    • Advantages
      • Practically all vehicles are gauged during the time the radar is installed.
      • Capacity to measure up to 12 lanes and with reliability in inclement weather.
      • Non-intrusive technology.
    • Limitations
      • Possibility of the device being vandalised while installed.
  • ANPR/RAM cameras. These are cameras that -by means of the automatic reading of license plates- make it possible to identify vehicle flows and obtain reliable origin-destination matrices, both at urban, regional and national level (preserving the anonymity of the data).
  • What data do they provide?
      • Long, medium or short distance vehicle O/D matrices.
      • Vehicle capacity in road sections, with vehicle classification.
    • Advantages
      • Obtaining complete vehicle O/D matrices, which is difficult to obtain with other technologies or methodologies.
      • Non-intrusive technology.
    • Limitations
      • Possibility of the device being vandalised while installed.
  • Drones. Data collection is carried out using a drone equipped with a camera equipped with artificial vision.
  • What data is obtained with them?
      • Vehicle gauging in road sections, with vehicle classification.
      • Short-range vehicle O/D matrices, i.e. movements at roundabouts, crossroads, junctions, etc.
    • Advantages
      • Full view of the road, junction or roundabout under study.
      • Allows recording in places that are difficult to access.
      • Practically all vehicles passing on the road under study are recorded during the time the camera is installed.
      • Non-intrusive technology.
    • Limitations
      • Short recording time limited to the battery of the drone (between 20 minutes to 1 hour approximately).
      • In general terms, drone pilot certification and registration with AESA, the Ministry of the Interior and possibly nearby airports are required in order to be able to fly.
Revealed and stated preference surveys are a very useful source when conducting demand studies.

REVEALED AND STATED PREFERENCE SURVEYS

This type of survey is usually carried out in the framework of demand studies, when data are needed to feed the so-called four-stage models.

Revealed Preference Surveys provide insight into the current travel behaviour of individuals, while Declared Preference Surveys try to reflect what individuals would do in certain hypothetical mobility situations (such as new transport lines, new modes, fare changes, etc.).

Historically, they were carried out in person (at stations, airports, roads…) or by telephone, although there is a trend towards online panels, and until a few years ago they were the only way to know the general mobility in all modes of transport at the same time.

  • What data is obtained with them?
    • O/D matrices for all modes of transport, and reasons for travel.
    • Socio-economic profile of the user.
    • Value of time and future decision intention.
  • Advantages
    • Measurement of all modes of transport, travel motives and other user parameters necessary for the development of transport models (income level, motorisation, age range, etc.).
    • Obtaining the time value parameter necessary for demand forecasting, a parameter that is difficult to obtain by other means.
  • Limitations
    • The surveys are carried out on a sample population, and it is necessary to extrapolate the results to the total study universe.
    • Costly method, requires the design of the survey and the intervention of interviewers (in person or by telephone) and the training and organisation process that this entails.
    • Intrusive method, requires the user to answer survey questions.
Mobile phone data are increasingly used in mobility studies.

MOBILE PHONE DATA

In recent years, the use of data from mobile devices for mobility analysis has become widespread.

In Spain, many of the major telephone companies, as well as other companies with agreements with these companies, commercialise mobile network signalling data (in an anonymised form) to study mobility. These data represent a significant share of the market and it is the companies themselves -aware of their share- that usually market these data already expanded to the total population.

  • What data do you get with them?
    • O/D matrices for (almost) all modes of medium and long distance transport.
    • Other data relating to the study areas: presence, overnight stays, etc.
  • Advantages
    • Possibility of accessing large population samples in short time frames.
    • Eliminates the possible bias derived from carrying out surveys, where people tend to simplify their answers.
  • Limitations
    • It only allows distinguishing the “road” mode, without distinguishing between bus and private vehicle, which is a limitation for the development of multimodal models.
    • It does not allow to obtain the classic trip motives, only usually home-work trips based on their frequency and stay in typical hours.
    • It does not allow the derivation of the time value parameter.
    • Although they provide certain socio-economic parameters (such as gender, age range, etc.), they are data that should be used with caution in demand studies since they are associated with the ownership of the contract.

In addition to these data, many other experiences from telephony have proliferated for mobility analysis. In recent years, more and more companies are marketing demand data collected through the use of associated mobile application packages (SDKs) for this purpose. In turn, social networks have provided insights into users’ mobility patterns for certain uses. For example, by analysing general mobility patterns from the analysis of geolocated posts published on networks such as Twitter. Or the so-called sentiment analysis, which allows us to know the perception that users have of a means of transport based on the analysis through algorithms of the nature of the publications where users refer to these means of transport. These and many other examples illustrate the vast amount of information on mobility patterns that can be extracted from our interaction with mobile phones.

GPS data marketed by various companies enriches and adds new dimensions to mobility studies.

GPS DATA

As with telephony data, obtaining mobility patterns by tracking the GPS data that we generate with different devices has become one of the most solid and widely used sources in recent years.

Companies such as TomTom, Inrix, Wejo and many others market this kind of data through agreements with vehicle fleets, navigation devices, sensors and other data sources that enable highly accurate geolocation of vehicle movements.

  • What data do you get with them?
    • Partial short- and long-range vehicle O/D matrices.
    • Partial vehicle gauges, as well as other trip parameters such as travel times, speeds, etc.
  • Advantages
    • Highly accurate, robust and reliable data for determining vehicle mobility patterns.
    • Quick to obtain, do not require the installation of any device, non-intrusive.
  • Limitations 
    • Partial data, corresponding to the provider’s market share, and usually traded without expanding to the full universe.
    • Limited to vehicular traffic.

    BY WAY OF CONCLUSION…

    Although additional data sources can be found in the market, these are the most common ones as they provide the most complete information for the needs of mobility analysis and traffic studies.

    As we mentioned at the beginning of this article, although some of the sources mentioned have clear advantages over others or important differences in terms of cost -not only economic but also in terms of management-, the truth is that we cannot state that any of them is clearly better than the others.

    They all have advantages and limitations, and in many cases we will need to use a combination of several in our studies in order to have as complete a picture as possible of our field of analysis.

    Therefore, the first step is to know its characteristics in order to be able to assess in each case, based on the budget of our project, the timeframe available, the data needed and the overall characteristics of the study, which is the most appropriate data source for the mobility analysis.

Vectio Traffic & Transport Planning

 

Related Posts

Mobility by Vectio 1×02 with Nati Armentia

Surviving the traffic jam: how China faces with traffic challenges

Candela Martín
Transport Engineer

The impact of mass tourism and new technologies on mobility

Irene Méndez
Transport Engineer

Mobility by Vectio 1×01 with Jorge Rodríguez

Beyond speed bumps

Irene Méndez
Transport Engineer

Mobility Studies for logistics, industrial and office developments at Madrid

Jorge Rodríguez
Technical Director

First steps in transport modelling

Ana Álvarez
Transport Modeller

Public Transport and Alternative Fuels: Fuelling Change

Johanna Díaz
Urban Planner

So, what if congestion is the solution?

Itziar Buruchaga
Transport Engineer

Planning Considerations for Logistic Delivery Center Parking

Yang Wang
Principal Engineer

Does a Four-day Workweek mean a Carbon-Zero Future?

The five-sol coaches and free public transport

Jorge Rodríguez
Technical Director

The history of the railway through cinema

Irene Méndez
Transport Engineer

Thinking Mobility from the outset: Transport-Oriented Developments

Mobility and Data Analytics: Data Sources for Accurate Analysis

Transport simulations, great allies in mobility management

Azariel Menéndez
Draftsman

Covid-19 and mobility: Has traffic gone back to normal after the pandemic?

Ana Álvarez
Transport Modeller

The 5 stages of Big Data applied to Transport Engineering

From roads that divide to roads that unite: scenarios for a non-segregated infrastructure

Driver behaviour: the key to analyze traffic functionality

Yang Wang
Principal Engineer
Categories
×

For further information, contact our commercial team.