In ZiggerZZ/Rproject: Probabylity analysis of SNCF train delays

Introduction

To create the map we need the department associated with the departure and arrival trainstation. To automate this search and implement it in a scalable way we used google maps api to get from a string the first result latitude and longitude, then we converted it back to a postal code and country (to filter out non French destinations). Then we needed to add the code per train line. Once we have a table with line names, departement codes and cotes we can run our carto function.

We want to get a dataframe containing departure/arrivals and their department code

We will use the following packages:

library(trainpack)
library(dplyr)

# use data from package
SNCF_regularite <- trainpack::SNCF_regularite

Take TGV out of Gare de Départ Name because our geocoding function is not working well with TGV

SNCF_regularite <- SNCF_regularite %>% 
  rowwise %>% 
  mutate(gare_de_depart = gsub(" TGV","", gare_de_depart),
         gare_d_arrivee = gsub(" TGV","", gare_d_arrivee)) %>% 
  ungroup()

We then want to add a lat/lon column for each gare de départ/arrivée

sncf_lon_lat <- SNCF_regularite %>% 
  group_by(gare_de_depart, gare_d_arrivee) %>% 
  summarise() %>% 
  mutate( lon_depart = do.call(rbind, lapply(gare_de_depart,FUN = geocoding))$lon,
          lat_depart = do.call(rbind, lapply(gare_de_depart,FUN = geocoding))$lat,
          lon_arrivee = do.call(rbind, lapply(gare_d_arrivee,FUN = geocoding))$lon,
          lat_arrivee = do.call(rbind, lapply(gare_d_arrivee,FUN = geocoding))$lat
          )

# Francfort did not work well so we'll change it manually. In the future we would need to implement this straight in the geocoding function.
sncf_lon_lat[sncf_lon_lat$gare_de_depart=="FRANCFORT",] <- sncf_lon_lat %>%
  filter(gare_de_depart == "FRANCFORT") %>% 
  mutate(lon_depart = geocoding("gare de FRANCFORT")$lon, lat_depart = geocoding("gare de FRANCFORT")$lat)

Find the postcode of each Gare de Départ and assign a country to filter out non French countries

sncf_lon_lat_postcode <- sncf_lon_lat %>% 
  rowwise() %>% 
  mutate(postcode_depart = as.character(mapply(rev_geocoding, lat = lat_depart, lon = lon_depart)), 
         postcode_arrivee = as.character(mapply(rev_geocoding, lat= lat_arrivee, lon = lon_arrivee))) %>% 
  mutate(country_depart = mapply(rev_geocoding_country, lat = lat_depart, lon = lon_depart),
         country_arrivee = mapply(rev_geocoding_country, lat = lat_arrivee, lon = lon_arrivee))

Extract two first characters of postcode

sncf_lon_lat_postcode <- sncf_lon_lat_postcode %>% 
  filter(country_depart=="France") %>% 
  mutate(CODE_DEPT_depart = substr(postcode_depart,1,2),
         CODE_DEPT = substr(postcode_arrivee,1,2))

Insert cote in the dataframe thanks to the fonction_cote function.

#cote_tbl <- sncf_lon_lat_postcode %>% mutate(cote = mapply(fonction_cote, gareA = gare_de_depart, gareB = gare_d_arrivee, month = 12)) 

Finally, we save it as a .csv

# convert to df
#cote_tbl <- as.data.frame(cote_tbl, stringsAsFactors = FALSE)

#cote_tbl <- write.csv(cote_tbl, file = "cote_tbl.csv")

ZiggerZZ/Rproject documentation built on May 31, 2019, 6:40 p.m.