Given the walking time analysis, we can estimate the location of a future school in the ACT that would have the largest impact on reducing walking times for students. One suggested location is estimated for primary and high school, but a number of decent options exist. This tool could be used by the department for future planning purposes.
The “Optimal New School Location” was found by randomly sampling 100 locations in the ACT and then calculating the average weighted walk time to a primary and high school’s if this new school was included. The location resulted in the largest decrease was selected as the “optimal new school”. Other more efficient algorithms for finding this location may exist however, given the limited available time in the competition this crude method was employed.
New Primary School Location
One of the best new primary school locations was estimated to be in North West Taylor. Worth noting that the pin is indicative of the area, not the exact location to place the school. Realistically, primary schools appear to be fairly evenly distributed. As a result, only a small improvement can be found.
Current Average (2024)
17.3 mins
Population-weighted walk time to nearest primary school
With New School
17 mins
Improvement: -0.32 minutes (1.8% reduction)
New High School Location
The best new high school location was estimated to be at the confluence of Wright, Coombs, Duffy and Holder. This makes a lot of sense, as this newly developed area in Canberra has attracted many young families. This is likely a more interesting finding that the primary school location, as high schools are fewer and more unevenly distributed.
Current Average (2024)
29.3 mins
Population-weighted walk time to nearest high school
With New School
27.3 mins
Improvement: -2.01 minutes (6.9% reduction)
Source Code
---title: "Optimal New School Locations in 2024"---Given the walking time analysis, we can estimate the location of a future school in the ACT that would have the largest impact on reducing walking times for students. One suggested location is estimated for primary and high school, but a number of decent options exist. This tool could be used by the department for future planning purposes.The "Optimal New School Location" was found by randomly sampling 100 locations in the ACT and then calculating the average weighted walk time to a primary and high school's if this new school was included. The location resulted in the largest decrease was selected as the "optimal new school". Other more efficient algorithms for finding this location may exist however, given the limited available time in the competition this crude method was employed.```{r setup}#| echo: false#| message: false#| warning: falselibrary(sf)library(leaflet)library(dplyr)library(htmltools)# Load optimal location dataoptimal_primary <-st_read(paste0(here::here(),"/data/new_schools/optimal_primary_location_2024.gpkg"), quiet = T)optimal_high <-st_read(paste0(here::here(),"/data/new_schools/optimal_high_location_2024.gpkg"), quiet = T)# Load 2024 walk time data for comparisonwalk_2024 <-st_read(here::here("data", "walk_analysis", "school_walk_times_2024.gpkg"), quiet =TRUE)primary_2024 <- walk_2024 |>filter(school_type =="primary")high_2024 <- walk_2024 |>filter(school_type =="high")# Create KPI callout functioncreate_kpi_callout <-function(title, value, subtitle, color ="#2166AC") {div(style =paste0("background-color: ", color, "; color: white; padding: 30px; border-radius: 12px; text-align: center; margin: 10px;"),h3(title, style ="margin: 0; font-size: 1.4em;"),h1(paste0(value, " mins"), style ="margin: 10px 0; font-size: 3em; font-weight: bold;"),p(subtitle, style ="margin: 0; font-size: 1.1em; opacity: 0.9;") )}# Create a focused map showing the proposed school location with SA2 backgroundcreate_school_location_map <-function(optimal_data, school_type, rank =1) {# Get the optimal location new_school <- optimal_data[rank,]# Get the corresponding walk time data walk_data <-if(school_type =="Primary") primary_2024 else high_2024# Calculate color palette - using plasma reversed so shorter times are yellow/bright pal <-colorNumeric("plasma", walk_data$mean_walk_time_min, na.color ="#cccccc", reverse =TRUE)leaflet() |>addProviderTiles(providers$CartoDB.Positron) |># Add SA2 polygons showing walk times (50% opacity)addPolygons(data = walk_data,fillColor =~pal(mean_walk_time_min),fillOpacity =~ifelse(population ==0, 0, 0.5), # 50% opacity, transparent if no populationcolor ="#444444",weight =1,label =~paste0( sa2_name, "<br>","Walk Time: ", round(mean_walk_time_min, 1), " mins<br>","Population (5-14): ", population ) |>lapply(htmltools::HTML),labelOptions =labelOptions(sticky =TRUE),group ="SA2 Walk Times" ) |># Add the proposed new school marker on topaddMarkers(data = new_school,lng =~longitude,lat =~latitude,popup =~paste0("<strong>Proposed ", school_type, " School (Rank ", rank, ")</strong><br>","Location: ", round(longitude, 4), ", ", round(latitude, 4), "<br>","Time Reduction: ", round(time_reduction_min, 2), " minutes<br>","Current Average: ", round(current_avg_walk_time_min, 1), " mins<br>","New Average: ", round(new_avg_walk_time_min, 1), " mins" ),icon =makeIcon(iconUrl ="https://raw.githubusercontent.com/pointhi/leaflet-color-markers/master/img/marker-icon-2x-red.png",shadowUrl ="https://cdnjs.cloudflare.com/ajax/libs/leaflet/0.7.7/images/marker-shadow.png",iconWidth =30, iconHeight =49,iconAnchorX =15, iconAnchorY =49,shadowWidth =51, shadowHeight =51 ),group ="New School" ) |># Add legend for walk timesaddLegend(pal = pal,values = walk_data$mean_walk_time_min,title =paste("Walk Time (mins)<br>", school_type, "Schools"),position ="bottomright" ) |># Add layer controladdLayersControl(overlayGroups =c("SA2 Walk Times", "New School"),options =layersControlOptions(collapsed =FALSE) ) |>setView(lng = new_school$longitude, lat = new_school$latitude, zoom =11# Zoomed out slightly to show more context )}```# New Primary School LocationOne of the best new primary school locations was estimated to be in North West Taylor. Worth noting that the pin is indicative of the area, not the exact location to place the school. Realistically, primary schools appear to be fairly evenly distributed. As a result, only a small improvement can be found.::: {.columns}::: {.column width="50%"}```{r}#| echo: false# Get values for rank 2 locationcurrent_avg <-round(optimal_primary$current_avg_walk_time_min[2], 1)projected_improvement <-round(optimal_primary$time_reduction_min[2], 2)new_avg <-round(optimal_primary$new_avg_walk_time_min[2], 1)# Current walk time KPIcreate_kpi_callout(title ="Current Average (2024)", value = current_avg,subtitle ="Population-weighted walk time to nearest primary school",color ="#dc2626"# Red for current state)```:::::: {.column width="50%"}```{r}#| echo: false# Improved walk time KPIcreate_kpi_callout(title ="With New School", value = new_avg,subtitle =paste0("Improvement: -", projected_improvement, " minutes (", round((projected_improvement/current_avg)*100, 1), "% reduction)"),color ="#16a34a"# Green for improvement)```:::::: ```{r}#| echo: falsecreate_school_location_map(optimal_primary, "Primary", rank =2)```# New High School LocationThe best new high school location was estimated to be at the confluence of Wright, Coombs, Duffy and Holder. This makes a lot of sense, as this newly developed area in Canberra has attracted many young families. This is likely a more interesting finding that the primary school location, as high schools are fewer and more unevenly distributed.::: {.columns}::: {.column width="50%"}```{r}#| echo: false# Get values for rank 1 locationcurrent_avg <-round(optimal_high$current_avg_walk_time_min[1], 1)projected_improvement <-round(optimal_high$time_reduction_min[1], 2)new_avg <-round(optimal_high$new_avg_walk_time_min[1], 1)# Current walk time KPIcreate_kpi_callout(title ="Current Average (2024)", value = current_avg,subtitle ="Population-weighted walk time to nearest high school",color ="#dc2626"# Red for current state)```:::::: {.column width="50%"}```{r}#| echo: false# Improved walk time KPIcreate_kpi_callout(title ="With New School", value = new_avg,subtitle =paste0("Improvement: -", projected_improvement, " minutes (", round((projected_improvement/current_avg)*100, 1), "% reduction)"),color ="#16a34a"# Green for improvement)```:::::: ```{r}#| echo: falsecreate_school_location_map(optimal_high, "High")```