Unlocking the power of data visualization with R

Alternative fueling stations data-set is from the Bureau of transportation statistics updated at January 3, 2022. The U.S. Department of Energy collects this data in partnership with Clean Cities coalitions and their stakeholders to help fleets and consumers find alternative fueling stations.¹

library(tidyverse)
# library(tidytuesdayR)
# tuesdata <- tidytuesdayR::tt_load(2022, week = 9)
# stations <- tuesdata$stations
stations <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2022/2022-03-01/stations.csv')

Stations dataset is made of 70 variables and about 60000 observations.

We are interested in:

Open date
States and cities
Geo codes
Fuel type code:
- Biodiesel (BD),
- Compressed Natural Gas (CNG),
- Electric (ELEC),
- Ethanol (ethanol-gasoline blends, E85),
- Hydrogen (HY),
- Liquefied Natural Gas (LNG),
- Propane (Liquefied Petroleum Gas, LPG)
Status code:
- Available (E)
- Planned (P)
- Temporarily Unavailable (T)
Access code:
- Public
- Private

Each point on the map is counted as one station in the station count

stations <- stations%>%
  janitor::clean_names()

stations%>%head

stations%>%select(contains("ev"))

stations%>%names%>%sort

stations%>%count(bd_blends,bd_blends_french)
stations%>%count(e85_blender_pump,e85_blender_pump,e85_other_ethanol_blends)
stations%>%count(ev_network,ev_on_site_renewable_source)
stations%>%count(hydrogen_is_retail,hydrogen_pressures,hydrogen_standards,hydrogen_status_link)
stations%>%count(lng_on_site_renewable_source)
stations%>%count(cng_on_site_renewable_source)

stations%>%count(facility_type)
stations%>%count(fuel_type_code)
stations%>%count(lpg_primary)

df <- stations%>%
  #filter(between(longitude,-126,-50),latitude>23)%>%
  mutate(fuel_type=case_when(fuel_type_code=="BD"~"Biodiesel (B20 and above)",
                             fuel_type_code=="CNG"~"Compressed Natural Gas (CNG)",
                             fuel_type_code=="E85"~"Ethanol (E85)",
                             fuel_type_code=="ELEC"~"Electric",
                             fuel_type_code=="HY"~"Hydrogen",
                             fuel_type_code=="LNG"~"Liquefied Natural Gas (LNG)",
                             fuel_type_code=="LPG"~"Propane (LPG)",
                             TRUE~fuel_type_code))%>%
  select(open_date,city,state,zip,latitude,longitude,
         access_code,
         facility_type,
         fuel_type_code,
         fuel_type,
         cng_on_site_renewable_source,
         lng_on_site_renewable_source,
         ev_on_site_renewable_source)

world <- map_data("world")%>%
  filter(!region=="Antarctica")
usa <-map_data("state")

library(showtext)
library(sysfonts)
library(extrafont)

showtext::showtext_auto()
showtext::showtext_opts(dpi=320)

font_add_google(name="Noto Sans",family="notosans")

plot <- ggplot() +
  geom_polygon(data=world,aes(x=long,y=lat,group=group),
               fill="grey85",color="#654321",size=0.1) +
  geom_polygon(data = usa,
               aes(x = long, y = lat, group = group),
               fill="grey85",color="#654321",size=0.1,linetype="dashed") +
  geom_point(data=df,aes(x=longitude,y=latitude,group=state,color=fuel_type_code),
             show.legend = T,size=0.1,#shape=".",
             key_glyph = rectangle_key_glyph(fill=color)) + 
  
  geom_point(data=subset(df,fuel_type_code=="CNG"),
             aes(x=longitude,y=latitude,group=state,color=fuel_type_code),
             show.legend = T,shape=21,stroke=0.1,
             key_glyph = rectangle_key_glyph(fill=color)) + 
  guides(colour=guide_legend(title.position="top", nrow = 1,
                                     title.hjust =0.5))+
  ggthemes::scale_color_tableau(name = "Fuel type code",
    breaks=c("ELEC", "E85", "LPG","CNG","BD","LNG","HY"))+
  coord_map("ortho", orientation = c(41.071548, -102.211644, 0)) +
  labs(title="",
       subtile="",
       x="",y="") +
  theme_void() +
  theme(text = element_text(family="notosans"),
        legend.position =  c(0.54,0.61),
        legend.box.spacing = unit(0,"pt"),
        legend.direction = "horizontal",
        legend.key.size = unit(10,"pt"),
        legend.justification = "center",
        plot.margin = margin(0,0,0,0,unit="pt"),
        panel.background = element_rect(fill="#d8ebff",color="#d8ebff"),
        plot.background = element_rect(fill="#d8ebff",color="#d8ebff"))

plot

library(gt)
library(gtExtras)
table <- df%>%
  count(fuel_type,fuel_type_code)%>%
  mutate(pct=round(n/sum(n)*100,2))%>%
  arrange(-pct)%>%
  select(-n)%>%
  gt::gt(
    caption="Bureau of Transportation Statistics")%>%
  gt_plt_bar(column = pct, keep_column = T)%>%
  tab_header(
    title = "Fuel Alternatives") %>%
  cols_label(fuel_type_code="",fuel_type = "Fuel Type",pct="% ")%>%
  gt_theme_guardian()

table

#class(table)
# table %>%
#   gtsave(filename = "table.png")

fuel_font <- df%>%
  select(contains("fuel"),longitude,latitude,city) %>%
  group_by(city)%>%
  summarise(fuel_type_code,avg_lng=mean(range(longitude)),avg_lat=mean(range(latitude)),.groups="drop")%>%
  ungroup()%>%
  distinct()

 

barplot <- df%>%
  count(fuel_type_code,fuel_type,cng_on_site_renewable_source,lng_on_site_renewable_source,ev_on_site_renewable_source) %>%
  pivot_longer(cols=c("cng_on_site_renewable_source","lng_on_site_renewable_source","ev_on_site_renewable_source"),
               names_to="source_name",values_to="source")%>%
  filter(!fuel_type=="NA",!source=="NONE")%>%
  select(-source_name)%>%
  drop_na() %>%
  
  ggplot(aes(x=fuel_type,y=source))+
  geom_col(position="dodge",aes(fill=source),show.legend = F)+
  geom_text(aes(label=source),position=position_dodge(width = 1),
            angle=0,hjust=1,vjust=0.5,size=5)+
  #coord_polar(theta = "x")+
  coord_flip()+
  #facet_wrap(vars(fuel_type))+
  ggthemes::scale_fill_tableau()+
  theme_void()+
  theme(text = element_text(family="notosans",size=12),
        strip.placement = "inside",
        axis.text.y = element_blank(),
        plot.background = element_blank(),
        panel.background = element_blank())

barplot

ggsave("barplot.png")

library(cowplot)
ggdraw()+
  draw_plot(plot,scale=1.6)+
  draw_image("barplot.png",scale=0.5,x=-0.3,y=0.1)+
  draw_image("table.png",scale=0.3,x=-0.32,y=-0.28)+
  draw_image("NTAD_Logo_32.png",scale=0.3,x=0.32,y=0.28)+
  draw_image("eia.png",scale = 0.2,x=0.35,y=-0.3)+
  draw_line(x = c(0.05, 0.65),y = c(0.88, 0.88),color = "gold", size = 4)+
  draw_label("Electric",x=0.018,y=0.75,size=12,fontfamily = "notosans",fontface = "bold",angle=90)+
  draw_label("Compr.nat.Gas",x=0.018,y=0.49,size=12,fontfamily = "notosans",fontface = "bold",angle=90)+
  draw_label("Fuel type by sources | US DOT | EIA",x=0.12,y=0.37,size=8,
             fontfamily = "notosans",angle=0)+
  draw_label("US Alternative Fuel Stations",x=0.35,y=0.9,size=30,
             fontfamily = "notosans",fontface = "bold")+
  draw_label("Although gasoline remains by far the dominant \ntransportation fuel, a variety of alternative fuels \nare currently in use,primarily by government \nand private fleets.\nThese fuels include electricity, propane,\nhigher ethanol-gasoline blends (E85),\nhydrogen, and natural gas.\nIn aggregate, there are currently about 10,000\n alternative fuel stations in the United States,\ncompared to approximately 160,000\n gasoline stations in the country.(EIA)",size=8,fontfamily = "notosans",
             x=0.86,y=0.47)+
  draw_label("DataSource: #TidyTuesday 2022/w9 | Alternative Fuel Stations | US DOT | EIA\nInfographics: Federica Gazzelloni",x=0.35,y=0.05,size=10,fontfamily = "notosans")

ggsave("w9_stations.png",width = 10)

Footnotes

datasource ↩︎