Surveys

This page lists all the survey types which are currently supported by MobiSurvStd.

All surveys can be processed without spatial / zonal data when running with the --skip-spatial flag.

Enquête mobilité des personnes 2019 (EMP)

Code: emp2019

Link: https://www.statistiques.developpement-durable.gouv.fr/resultats-detailles-de-lenquete-mobilite-des-personnes-de-2019

Tested version: November 2024

Expected format:

emp_2019/
├── *k_deploc_public_V4.csv  # The `*` character can be anything (there is a typo in the original filename)
├── k_individu_public_V3.csv
├── q_2rmot_public_V3.csv
├── q_menage_public_V3.csv
├── q_voitvul_public_V3.csv
├── tcm_ind_kish_public_V3.csv
├── tcm_ind_public_V3.csv
└── tcm_men_public_V3.csv

Notes:

• Many variables from the original survey data are not read because they are not part of any other survey type.
• The survey does not guarantee any representativeness at the regional level.
• Only one person per household is surveyed for their trips and additional characteristics.
• The car type is not specified. Cars can include passenger cars, utilitary cars and recreational cars.
• The survey includes data on 58 “voiturettes”, 59 “quads”, and 1 “tricycle” as a special vehicle category, which is not read by MobiSurvStd.
• Walking legs are not specified in the survey data for trips combining walking with other modes.
• The surveyed day is drawn among the days of the week before the interview on which the person traveled. Therefore, the persons with no trip (nb_trips = 0) are the persons who did not travel during the whole week, not the persons who did not travel during an average day. These persons have no surveyed day (trips_weekday is null).
• A representative weekday (e.g., an average Monday) can be built from the union of:
  • the persons surveyed for that weekday (trips_weekday of the household is "monday"), with their trips, weighted by sample_weight_day;
  • the persons who did not travel on that weekday (traveled_on_monday is false), with no trip, weighted by sample_weight_surveyed.

Enquêtes mobilité certifiés Cerema (EMC²)

Code: emc2

Link: https://www.cerema.fr/fr/activites/mobilites/connaissance-modelisation-evaluation-mobilite/enquetes-mobilite-emc2

Download link (for researchers): https://data.progedo.fr/series/adisp/enquetes-menages-deplacements-emd-enquetes-mobilite-certifiee-cerema-emc

Expected format (recent Progedo exports):

my_emc2_survey/
├── *-Donnees_CSV  # Subdirectory names do not matter.
│   └── fichiers_standards
│       ├── *_std_depl.csv
│       ├── *_std_men.csv
│       ├── *_std_pers.csv
│       └── *_std_traj.csv
└── *-Documentation  # Subdirectory names do not matter.
    └── SIG
        ├── *_ZF(_*)?.(TAB|shp)           # Optional "Zones fines" file
        ├── *_GT(_*)?.(TAB|shp)           # Optional "Générateurs de trafic" file
        ├── *_GT_externes(_*)?.(TAB|shp)  # Optional "Générateurs de trafic" external file
        └── *_DTIR(_*)?.(TAB|shp)         # Optional "Zones de tirage" file

Expected format (older Progedo exports):

my_emc2_survey/
├── Csv
│   └── Fichiers_Standard
│       ├── *_std_depl.csv
│       ├── *_std_men.csv
│       ├── *_std_pers.csv
│       └── *_std_traj.csv
└── Doc
    └── SIG
        ├── *_ZF(_*)?.(TAB|shp)    # Optional "Zones fines" file
        ├── *_GT(_*)?.(TAB|shp)    # Optional "Générateurs de trafic" file
        └── *_DTIR(_*)?.(TAB|shp)  # Optional "Zones de tirage" file

Tested surveys:

• Alençon 2018
• Angers 2022
• Angoulême 2023 New
• Besançon 2018
• Bordeaux 2021
• Bouzonville 2019
• Brest 2018
• Chambéry 2022
• Clermont-Ferrand 2023
• Évreux 2018
• Gap 2018
• Grenoble 2020
• Lannion 2022
• Le Havre 2018
• Lens / Artois 2023 New
• Marseille 2020
• Nevers 2023 New
• Nice 2023
• Pointe-à-Pitre 2021
• Poitiers 2018
• Reims 2021
• Rennes 2018
• Sables d’Olonne 2021
• Saint-Étienne 2021
• Strasbourg 2024
• Toulouse 2023
• Tours 2019
• Valenciennes 2019
• Vannes 2023 New
• Vendée 2020

Notes:

• The survey detailed zones correspond to the “zones fines” (“ZF”), which are usually the size of IRIS zones or smaller.
• For most surveys, there are special locations corresponding to “Générateurs de trafic” (“GT”), which represent e.g., train stations or hospitals.
• Draw zones correspond to the survey’s “zones de tirage”.
• For each trips, the non-walking legs are recorded. MobiSurvStd reconstruct the walking legs using the access / egress walking time variables. This means that the leg travel time is known only for the walking legs (as it is not a recorded variable for the other legs).

Enquête Globale Transport 2018-2020

Code: egt2020

Link: https://data.progedo.fr/studies/doi/10.13144/lil-1581

Expected format (version 3):

egt_2020/
├── 01_menage_egt1820.csv
├── 02_individu_egt1820.csv
├── 03_deplacement_egt1820.csv
├── 04_trajet_egt1820.csv
├── 05_voiture_egt1820.csv
├── 06_drm_egt1820.csv
└── 07_velo_egt1820.csv

Expected format (older versions):

egt_2020/
└── Csv
    ├── a_menage_egt1820.csv
    ├── b_individu_egt1820.csv
    ├── c_deplacement_egt1820.csv
    ├── d_trajet_egt1820.csv
    ├── e_voiture_egt1820.csv
    ├── f_drm_egt1820.csv
    └── g_velo_egt1820.csv

Notes:

• Due to some households not being surveyed with the expected method (phone vs face to face), you should not rely to much on the sample weights of this survey.
• There is no draw zones because the survey was interrupted before the end due to Covid-19.
• There is no detailed zone but the coordinates of the origins / destinations are known (rounded to 100 meters).
• The car types are not specified but utilitary cars and recreational cars ar excluded.
• Motorcycles’ parking data is not read because the survey’s documentation is incorrect.
• Variable main_mode_group can differ from the original MODP_H7 variable because MobiSurvStd uses a different definition of mode groups and a different methodology to identify the main mode.
• The start_insee and end_insee values are derived from the leg’s coordinates, using INSEE data from 2025.
• The work_insee and study_insee values are derived from the given coordinates, using INSEE data from 2025. There can be some errors because of the 100m rounding.

Enquête Mobilité par GPS 2023 (EMG)

Code: emg2023

Link: https://www.institutparisregion.fr/mobilite-et-transports/deplacements/enquete-regionale-sur-la-mobilite-des-franciliens/

Expected format:

emg/
├── EMG_BD_Deplacements_*.xlsx
├── EMG_BD_Individus_*.xlsx
└── EMG_Distance.xlsx     # Optional, gives trip euclidean distances.

Notes:

• Only one person per household is surveyed. Characteristics of all other members are unknown.
• The number of cars and motorcycles per household is known but their characteristics are unknown.
• The survey is conducted over a week (seven days) for each person. The trip-level variable trip_date can be used to identify the day of the trips. The person might not have traveled each day of the surveyed week.
• Variables traveled_on_{weekday} are defined from the recorded days of the surveyed week: they are true if at least one trip was recorded and false if the day was explicitly recorded without trip. They are null if the GPS trace was missing, if the person was outside Île-de-France, if the weekday was not surveyed, or if the weekday was surveyed several times with inconsistent results. To build a representative weekday, select the persons with a non-null traveled_on_{weekday} value, weighted by sample_weight_surveyed (rescaled to account for the null values), and the trips with the corresponding trip_weekday.
• Trips done before 4 a.m. are usually registered to the previous day, with a departure / arrival time value over 24h.
• Origins and destinations are defined at the INSEE level.

Implementation in MobiSurvStd by Sebastian Hörl

Enquêtes Déplacements Grands Territoires (EDGT)

Code: edgt

Link: https://www.cerema.fr/fr/activites/mobilites/connaissance-modelisation-evaluation-mobilite/enquetes-mobilite-emc2

Download link (for researchers): https://data.progedo.fr/studies?q=edgt

Expected format:

my_edgt_survey/
├── Csv
│   ├── Fichiers_Standard_Face_a_face
│   │   ├── *_std_faf_depl.csv
│   │   ├── *_std_faf_men.csv
│   │   ├── *_std_faf_pers.csv
│   │   └── *_std_faf_traj.csv
│   └── Fichiers_Standard_Telephone
│       ├── *_std_tel_depl.csv
│       ├── *_std_tel_men.csv
│       ├── *_std_tel_pers.csv
│       └── *_std_tel_traj.csv
└── Doc
    └── SIG
        ├── *(_ZF|zones_fines).(TAB|shp|MIF)        # Optional "Zones fines" file (the actual regex is more complex)
        ├── *(_GT|générateur*).(TAB|shp|MIF)        # Optional "Générateurs de trafic" file (the actual regex is more complex)
        └── *(_DTIR|_secteurstirage).(TAB|shp|MIF)  # Optional "Zones de tirage" file

Tested surveys:

• Amiens 2010
• Angers 2012
• Annecy 2017
• Annemasse 2016
• Bayonne 2010
• Clermont-Ferrand 2012
• Dijon 2016
• Dunkerque 2015
• Lyon 2015
• Marseille 2009
• Metz 2017
• Montpellier 2014
• Nancy 2013
• Nantes 2015 (Also available in an anonymized open-data format)
• Nice 2009
• Saint-Denis-de-la-Réunion 2016
• Saint-Quentin-en-Yvelines 2010
• Valence 2014

Notes:

• The survey detailed zones correspond to the “zones fines” (“ZF”), which are usually the size of IRIS zones or smaller.
• For most surveys, there are special locations corresponding to “Générateurs de trafic” (“GT”), which represent e.g., train stations or hospitals.
• Draw zones correspond to the survey’s “zones de tirage”.
• For each trips, the non-walking legs are recorded. MobiSurvStd reconstruct the walking legs using the access / egress walking time variables. This means that the leg travel time is known only for the walking legs (as it is not a recorded variable for the other legs).

Enquêtes Déplacements Villes Moyennes (EDVM)

Code: edvm

Link: https://www.cerema.fr/fr/activites/mobilites/connaissance-modelisation-evaluation-mobilite/enquetes-mobilite-emc2

Download link (for researchers): https://data.progedo.fr/studies?q=edvm

Expected format:

my_edvm_survey/
├── Csv
│   └── Fichiers_Standard
│       ├── *_std_depl.csv
│       ├── *_std_men.csv
│       ├── *_std_pers.csv
│       └── *_std_traj.csv
└── Doc
    └── SIG
        ├── *(_ZF|zones_fines).(TAB|shp|MIF)  # Optional "Zones fines" file (the actual regex is more complex)
        ├── *(_GT|générateur*).(TAB|shp|MIF)  # Optional "Générateurs de trafic" file (the actual regex is more complex)
        └── *_DTIR.(TAB|shp|MIF)              # Optional "Zones de tirage" file

Tested surveys:

• Ajaccio 2017
• Albi 2011
• Angoulême 2012
• Arras 2014
• Beauvais 2010
• Béziers 2014
• Bourg-en-Bresse 2017
• Carcassonne 2015
• Châlon-sur-Saône 2014
• Cherbourg 2016
• Créil 2017
• Dinan 2010
• La Rochelle 2011
• La-Roche-sur-Yon 2013
• Laval 2011
• Le Creusot 2012
• Les Sables d’Olonne 2011
• Longwy 2014
• Niort 2016
• Puisaye-Forterre et Aillantais 2012
• Quimper 2013
• Roanne 2012
• Saint-Brieuc 2012
• Saintes 2016
• Saint-Louis 2011
• Thionville 2012
• Var 2012

Notes:

• The survey detailed zones correspond to the “zones fines” (“ZF”), which are usually the size of IRIS zones or smaller.
• For most surveys, there are special locations corresponding to “Générateurs de trafic” (“GT”), which represent e.g., train stations or hospitals.
• Draw zones correspond to the survey’s “zones de tirage”.
• For each trips, the non-walking legs are recorded. MobiSurvStd reconstruct the walking legs using the access / egress walking time variables. This means that the leg travel time is known only for the walking legs (as it is not a recorded variable for the other legs).

Enquêtes Ménages Déplacements (EMD)

Code: emd

Link: https://www.cerema.fr/fr/activites/mobilites/connaissance-modelisation-evaluation-mobilite/enquetes-mobilite-emc2

Download link (for researchers): https://data.progedo.fr/studies?q=emd

Expected format:

my_emd_survey/
├── Csv
│   └── Fichiers_Standard
│       ├── *_std_depl.csv
│       ├── *_std_men.csv
│       ├── *_std_pers.csv
│       └── *_std_traj.csv
└── Doc
    └── SIG
        ├── *(_ZF|zones_fines).(TAB|shp|MIF)  # Optional "Zones fines" file (the actual regex is more complex)
        ├── *(_GT|générateur*).(TAB|shp|MIF)  # Optional "Générateurs de trafic" file (the actual regex is more complex)
        └── *(_DTIR|secteur_*).(TAB|shp|MIF)  # Optional "Zones de tirage" file

Tested surveys:

• Douai 2012
• Fort-de-France 2014
• Grenoble 2010
• Lille 2016
• Nîmes 2015
• Rouen 2017
• Saint-Étienne 2010
• Strasbourg 2009
• Toulouse 2013
• Valenciennes 2011

Notes:

• The survey detailed zones correspond to the “zones fines” (“ZF”), which are usually the size of IRIS zones or smaller.
• For most surveys, there are special locations corresponding to “Générateurs de trafic” (“GT”), which represent e.g., train stations or hospitals.
• Draw zones correspond to the survey’s “zones de tirage”.
• For each trips, the non-walking legs are recorded. MobiSurvStd reconstruct the walking legs using the access / egress walking time variables. This means that the leg travel time is known only for the walking legs (as it is not a recorded variable for the other legs).

Enquête Globale Transport 2010

Code: egt2010

Link: https://data.progedo.fr/studies/doi/10.13144/lil-0883

Expected format:

egt_2010/
├── Csv  # Or Format_csv
│   ├── deplacements_semaine.csv  # Filenames are case-insensitive (the first letter can be capitalized).
│   ├── menages_semaine.csv
│   ├── personnes_semaine.csv
│   └── trajets_semaine.csv
├── Doc
│   └── Carreaux_shape_mifmid
│       └── carr100m.shp       # Option 1: Carreaux are in a shapefile within a directory.
└── Carreaux_shape_mifmid.zip  # Option 2: Carreaux are in a zipfile within the parent directory.

Notes:

• The detailed zones correspond to the 100m by 100m cells.
• For now, only the weekday trips are standardized.