library(sf)
library(dplyr)
library(readr)
library(stringr)
library(ggplot2)
library(tigris)
library(scales)

1. Purpose

This project inventories active business licenses and active PLCB liquor licenses within the East Passyunk study area to support a corridor-scale understanding of commercial activity and neighborhood character.

The deliverable is a dashboard-style summary: a map of active licenses and a compact breakdown of license composition.

2. Data and Study Area

2.1 Load business licenses

bl <- st_read("data/business_licenses.geojson", quiet = TRUE) %>%
  st_transform(2272) #Philly/SEPA analysis ESPG

2.2 Load study area boundary

ep_boundary <- st_read("data/study_areas_2025_1.shp", quiet = TRUE) %>%
  st_transform(2272)

2.3 Load PLCB geocoded liquor licenses (gathered by planner and prof. Michael Fichman)

plcb_raw <- readr::read_csv("data/PHL_PLCB_geocoded.csv", show_col_types = FALSE) %>%
  mutate(
    lon = as.numeric(lon),
    lat = as.numeric(lat)
  )

plcb_sf <- plcb_raw %>%
  mutate(
    lon = as.numeric(lon),
    lat = as.numeric(lat)
  ) %>%
  filter(!is.na(lon), !is.na(lat)) %>%
  st_as_sf(coords = c("lon", "lat"), crs = 4326, remove = FALSE) %>%
  st_transform(2272)

3. Prepare for Analysis by Joining

3.1 Clip both datasets to East Passyunk

bl_ep <- st_intersection(bl,ep_boundary)

plcb_ep <- st_intersection(plcb_sf, ep_boundary)

3.2 Filter to active licenses and standardize columns

bl_ep_active <- bl_ep %>%
  filter(licensestatus == "Active") %>%   # filter to active status
  mutate(
    category = case_when(
      licensetype %in% c(
        "Food Preparing and Serving",
        "Food Preparing and Serving (30+ SEATS)",
        "Sidewalk Cafe",
        "Food Establishment, Retail Permanent Location",
        "Food Caterer",
        "Food Estab, Retail Non-Permanent Location (Annual)",
        "Food Manufacturer / Wholesaler"
      ) ~ "Dining",
      licensetype %in% c(
        "Amusement",
        "Limited Lodging Operator",
        "Special Assembly Occupancy"
      ) ~ "Entertainment",
      TRUE ~ "Other"
    ),
    source = "Business License"
  ) %>%
  arrange(addressobjectid, category) %>%
  group_by(addressobjectid) %>%
  slice(1) %>%
  ungroup() %>%
  st_zm(drop = TRUE, what = "ZM")

Filter for PLCB active licenses

plcb_ep_active <- plcb_ep %>%
  filter(Status == "Active") %>%
  mutate(
    category = "PLCB License",
    source = "PLCB License", 
    licensetype = License.Type) %>%
  st_zm(drop = TRUE, what = "ZM")

3.3 Combine business licenses and PLCB licenses into one dataset

licenses_ep <- bind_rows(
  bl_ep_active %>%
    transmute(
      source,
      category,
      licensetype = licensetype,
      name = business_name,
      address = address,
      geometry
    ),
  plcb_ep_active %>%
    transmute(
      source,
      category,
      licensetype = License.Type,
      name = Licensee,
      address = Premises.Address,
      geometry
    )
)

# quick sanity check
nrow(licenses_ep)
## [1] 255

4. Dashboard Summary

4.1 Key Counts

## 4.1 Key Counts by Category

summary_counts <- licenses_ep %>%
  st_drop_geometry() %>%
  count(category, sort = TRUE) %>%
  mutate(
    percent = round(100 * n / sum(n), 2)
  )

summary_counts

4.2 License Composition Chart

## 4.2 License Composition Chart

license_type_summary <- licenses_ep %>%
  st_drop_geometry() %>%
  count(category, licensetype, sort = TRUE) %>%
  group_by(category) %>%
  mutate(
    pct_within_category = n / sum(n)
  ) %>%
  ungroup()

ggplot(
  license_type_summary,
  aes(x = reorder(licensetype, n), y = n, fill = category)
) +
  geom_col(alpha = 0.9) +
  coord_flip() +
  geom_text(aes(label = n), hjust = -0.15, size = 4) +
  scale_y_continuous(expand = expansion(mult = c(0, 0.15))) +
  labs(
    title = "Food, Beverage & Entertainment License Distribution — East Passyunk",
    subtitle = "Active licenses only",
    x = NULL,
    y = "Count",
    fill = "Category"
  ) +
  theme_minimal(base_size = 13) +
  theme(
    plot.title = element_text(face = "bold"),
    legend.position = "bottom"
  )

5.1 Load Clip Roads

## 5.1 Local Roads for Context

options(tigris_use_cache = TRUE)

roads <- roads(state = "PA", county = "Philadelphia") %>%
  st_transform(2272)

ep_buffer <- st_buffer(ep_boundary, dist = 800)  # ~800 ft buffer
roads_clipped <- st_intersection(roads, ep_buffer)

5.1 Dashboard Map

## 5.2 Map of Active Licenses in East Passyunk

category_palette <- c(
  "Dining" = "#7b3294",
  "Entertainment" = "#686569",
  "Other" = "#ADB068",
  "PLCB License" = "#D17621"
)

ggplot() +
  geom_sf(data = roads_clipped, color = "gray55", linewidth = 0.25) +
  geom_sf(
    data = ep_boundary,
    fill = NA,
    color = "#F50A0A",
    linewidth = 0.9,
    linetype = "dashed"
  ) +
  geom_sf(
    data = licenses_ep,
    aes(color = category),
    size = 2.2,
    alpha = 0.9
  ) +
  scale_color_manual(values = category_palette) +
  labs(
    title = "Active Business and Liquor Licenses in East Passyunk",
    color = "License Category"
  ) +
  theme_void() +
  theme(
    plot.background = element_rect(fill = "black", color = NA),
    panel.background = element_rect(fill = "black", color = NA),
    plot.title = element_text(
      hjust = 0.5,
      color = "white",
      size = 16,
      face = "bold"
    ),
    legend.background = element_rect(fill = "black", color = NA),
    legend.key = element_rect(fill = "black", color = NA),
    legend.text = element_text(color = "white"),
    legend.title = element_text(color = "white", face = "bold")
  )