diff --git a/summary_script.R b/summary_script.R
new file mode 100644
index 0000000..e2c6eed
--- /dev/null
+++ b/summary_script.R
@@ -0,0 +1,102 @@
+# Separate each row by separator |
+# health_base_fr <- health_base |> as.data.frame() |> filter(str_detect(everything(-id), '|'))
+
+
+library(tidyverse)
+# Set seed
+set.seed(1024)
+
+health_base <- read_csv("C:/Users/Ali/RMIT University/JIBE working group - simulationResults/ForUrbanTransition/reference/health/04_death_and_disease/pp_healthDiseaseTracker_2039_fixBug_processed.csv")
+
+health_cyc <- read_csv("C:/Users/Ali/RMIT University/JIBE working group - simulationResults/ForUrbanTransition/reference/health/04_death_and_disease/pp_healthDiseaseTracker_2039_fixBug_processed.csv")
+
+# Set sample size
+sample_size <- 10000
+
+# Keep until 2039
+health_base <- health_base |> dplyr::select(1:21) |> slice_sample(n = sample_size)
+
+# Keep until 2039
+health_cyc <- health_cyc |> dplyr::select(1:21) |> slice_sample(n = sample_size)
+
+get_expanded_rows <- function(health_base_fr){
+  health_base_fr |> 
+  pivot_longer(cols = -c(id)) |> 
+  mutate(unpacked = str_split(value, "\\|")) |> 
+  unnest() |> 
+  mutate(value = str_trim(unpacked))  |> 
+  dplyr::select(-unpacked)
+}
+
+
+# Expand each row separated by | character
+health_base_summary <- get_expanded_rows(health_base)
+
+
+# Expand each row separated by | character
+health_cyc_summary <- get_expanded_rows(health_cyc)
+
+# States
+states_base_sum <- health_base_summary |> 
+  group_by(name, value)|> 
+  summarise(nv = dplyr::n(), 
+            freq = round(100 * nv / sample_size, 1), scenario = "reference")
+
+
+ggplot(states_base_sum) +
+  aes(x = name, y = nv, fill = value) +
+  geom_col() +
+  scale_fill_hue(direction = 1) +
+  theme_minimal() +
+  facet_wrap(vars(value))
+
+
+states_base_freq <- states_base_sum |>  
+  filter(freq > 0)
+
+ggplot(states_base_freq) +
+  aes(x = name, y = freq, fill = value) +
+  geom_col() +
+  scale_fill_hue(direction = 1) +
+  theme_minimal() +
+  facet_wrap(vars(value))
+
+
+
+# States
+states_cyc_sum <- health_cyc_summary |> 
+  group_by(name, value)|> 
+  summarise(nv = dplyr::n(), 
+            freq = round(100 * nv / sample_size, 1), scenario = "cycling intervention")
+
+
+ggplot(states_cyc_sum) +
+  aes(x = name, y = nv, fill = value) +
+  geom_col() +
+  scale_fill_hue(direction = 1) +
+  theme_minimal() +
+  facet_wrap(vars(value))
+
+
+states_cyc_freq <- states_cyc_sum |>  
+  filter(freq > 0)
+
+ggplot(states_cyc_freq) +
+  aes(x = name, y = freq, fill = value) +
+  geom_col() +
+  scale_fill_hue(direction = 1) +
+  theme_minimal() +
+  facet_wrap(vars(value))
+
+combine_summary <- bind_rows(states_base_sum, states_cyc_sum) |> mutate(scenario = factor(scenario, levels = c("reference", "cycling intervention")))
+
+combine_summary %>%
+  filter(freq >= 2L & freq <= 85L) %>%
+  ggplot() +
+  aes(x = name, y = freq, fill = scenario) +
+  geom_col(position = "dodge2") +
+  scale_fill_hue(direction = 1) +
+  theme_minimal() +
+  theme(axis.text.x = element_text(angle = 90L)) +
+  facet_wrap(vars(value))
+