Final Project 606 Visualizations
Mehreen Ali Gillani
2025-11-26
Flight Data Analysis: U.S. Domestic Air Travel Patterns (Jan-July 2025)
Project Overview
This analysis examines U.S. domestic air travel patterns during the first seven months of 2025, utilizing data from the T-100 Domestic Market (U.S. Carriers) dataset. The comprehensive visualization explores key metrics including passenger volume, flight operations, freight, mail transportation, and distance patterns across the national air network.
Analysis Scope
The investigation employs extensive univariate and multivariate data visualization techniques to uncover insights about:
Passenger distribution and travel demand fluctuations
Flight frequency patterns across major routes
Origin-destination dynamics and popular travel corridors
Carrier performance comparisons across three categories: Major, National, and Regional carriers
Key Focus Areas
Through correlation analysis and comparative visualization, this report specifically examines relationships between:
Passenger volume and flight frequency
Distance traveled and carrier type preferences
Geographic patterns in origin and destination cities
Market segmentation across different carrier classifications
Methodology
The analysis employs R programming with visualization packages including ggplot2 to create interactive and informative representations of complex air travel data, providing stakeholders with actionable insights into 2025 domestic aviation trends.
Read Flights
url <- "https://raw.githubusercontent.com/mehreengillani/Final_project_Data606/refs/heads/main/flights_data_clean.csv"
flights <- read_csv(url)## Rows: 32476 Columns: 16
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (10): unique_carrier, unique_carrier_name, origin_city_name, dest_city_n...
## dbl (6): passengers, freight, mail, distance, month, distance_group
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.# Verify the new dataframe
dim(flights)## [1] 32476 16colnames(flights)## [1] "passengers" "freight" "mail"
## [4] "distance" "unique_carrier" "unique_carrier_name"
## [7] "origin_city_name" "dest_city_name" "month"
## [10] "distance_group" "class" "distance_cat"
## [13] "carrier_type" "month_name" "route"
## [16] "distance_simple"view(flights)1. UNI-variate analysis:
1.1 NUMERICAL VARIABLES
# Set theme for better visuals
theme_set(theme_minimal())
# Passengers distribution
p1 <- ggplot(flights, aes(x = passengers)) +
geom_histogram(bins = 30, fill = "steelblue", alpha = 0.7) +
labs(title = "Distribution of Passengers",
x = "Number of Passengers", y = "Count") +
scale_x_continuous(labels = scales::comma)
# Distance distribution
p2 <- ggplot(flights, aes(x = distance)) +
geom_histogram(bins = 30, fill = "darkorange", alpha = 0.7) +
labs(title = "Distribution of Distance",
x = "Distance", y = "Count") +
scale_x_continuous(labels = scales::comma)
# 1. Create binary indicator
flights <- flights %>%
mutate(has_cargo = ifelse(mail > 0 | freight > 0, "Yes", "No"))
# Add title to the table output
cat("=== CARGO VS PASSENGER-ONLY FLIGHTS ===\n")## === CARGO VS PASSENGER-ONLY FLIGHTS ===table(flights$has_cargo)##
## No Yes
## 27741 4735# Or with proportions
prop.table(table(flights$has_cargo))##
## No Yes
## 0.8542 0.1458# 2. Plot cargo vs no-cargo flights
p3 <- ggplot(flights, aes(x = has_cargo, fill = has_cargo)) +
geom_bar() +
geom_text(stat = "count", aes(label = ..count..),
vjust = -0.5, size = 4, fontface = "bold") +
labs(title = "Flights with Cargo vs Passenger-Only",
x = "Carries Cargo?", y = "Number of Flights")
# 3. Distribution of non-zero cargo
cargo_flights <- flights %>% filter(mail > 0 | freight > 0)
# mail distribution
p4 <-ggplot(cargo_flights, aes(x = mail)) +
geom_histogram(bins = 30, fill = "brown") +
labs(title = "Mail Distribution \n (Non-zero flights only)")
# Freight distribution
p5 <- ggplot(cargo_flights, aes(x = freight)) +
geom_histogram(bins = 30, fill = "darkred") +
labs(title = "Freight Distribution\n(Non-zero flights only)")
# Numerical plots grid
grid.arrange(p1, p2, p5, p4, ncol = 2)
print(p3)## Warning: The dot-dot notation (`..count..`) was deprecated in ggplot2 3.4.0.
## ℹ Please use `after_stat(count)` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
1.2 CATEGORICAL VARIABLES
# Carrier type distribution
p5 <- ggplot(flights, aes(x = carrier_type)) +
geom_bar(fill = "skyblue", alpha = 0.7) +
labs(title = "Distribution by Carrier Type",
x = "Carrier Type", y = "Count") +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
# Month distribution
p6 <- ggplot(flights, aes(x = month_name)) +
geom_bar(fill = "lightgreen", alpha = 0.7) +
geom_text(stat = 'count', aes(label = ..count..), vjust = -0.2) +
labs(title = "Distribution by Month",
x = "Month", y = "Count") +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
# Origin city distribution (top 15)
top_origins <- flights %>%
count(origin_city_name) %>%
arrange(desc(n)) %>%
head(15)
p7 <- ggplot(top_origins, aes(x = reorder(origin_city_name, n), y = n)) +
geom_bar(stat = "identity", fill = "coral", alpha = 0.7) +
labs(title = "Top 15 Origin Cities",
x = "Origin Cities", y = "Count") +
coord_flip()
# Destination city distribution (top 15)
top_dests <- flights %>%
count(dest_city_name) %>%
arrange(desc(n)) %>%
head(15)
p8 <- ggplot(top_dests, aes(x = reorder(dest_city_name, n), y = n)) +
geom_bar(stat = "identity", fill = "goldenrod", alpha = 0.7) +
labs(title = "Top 15 Destination Cities",
x = "Destination Cities", y = "Count") +
coord_flip()
# Distance category distribution
p99 <- ggplot(flights, aes(x = distance_cat)) +
geom_bar(fill = "lightpink", alpha = 0.7) +
labs(title = "Distribution by Distance Category",
x = "Distance Category", y = "Count") +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
# Create new distance categories (5 categories)
flights_distance <- flights %>%
mutate(distance_cat_5 = cut(distance,
breaks = c(0, 300, 600, 1200, 2400, Inf),
labels = c("Short(0-300mi)",
"Medium-Short(301-600mi)",
"Medium(601-1200mi)",
"Long(1201-2400mi)",
"Very Long(2400+mi)"),
include.lowest = TRUE))
# Plot with new categories
p9 <- ggplot(flights_distance, aes(x = distance_cat_5)) +
geom_bar(fill = "#2E86AB") +
geom_text(stat = 'count', aes(label = ..count..), vjust = -0.2) +
labs(title = "Flights by Distance", x = "Distance", y = "Count") +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
# Class distribution
p10 <- ggplot(flights, aes(x = class)) +
geom_bar(fill = "lightsteelblue", alpha = 0.7) +
labs(title = "Distribution by Class",
x = "Class", y = "Count")
# Get top 15 routes
top_routes <- flights %>%
count(route) %>%
arrange(desc(n)) %>%
head(15)
# Plot top 15 routes
p11 <- ggplot(top_routes, aes(x = reorder(route, n), y = n)) +
geom_bar(stat = "identity", fill = "lightblue", alpha = 0.8) +
geom_text(aes(label = scales::comma(n)),
hjust = -0.2, size = 3.5, color = "darkblue") +
labs(title = "Top 15 Busiest Routes",
subtitle = "Most frequently traveled flight routes",
x = NULL,
y = "Number of Flights") +
scale_y_continuous(labels = scales::comma, expand = expansion(mult = c(0, 0.1))) +
coord_flip()
# 3. ARRANGE PLOTS IN GRID
# Arrange plots with proper spacing and sizing
grid.arrange(p5, p6, ncol = 2)
#padding = unit(3, "cm")
grid.arrange(p9,p10, ncol = 2)
# State distribution plots
grid.arrange(p7, p8, ncol = 2)
#Route
grid.arrange(p11, ncol = 2)
# summary for key metrics only
summary_table <- flights %>%
select(carrier_type, month_name, class) %>%
summary()
print(summary_table)## carrier_type month_name class
## Length:32476 Length:32476 Length:32476
## Class :character Class :character Class :character
## Mode :character Mode :character Mode :charactersummary(flights_distance$distance_cat_5)## Short(0-300mi) Medium-Short(301-600mi) Medium(601-1200mi)
## 6891 7606 10984
## Long(1201-2400mi) Very Long(2400+mi)
## 6329 666# Summary statistics for numerical variables
summary_stats <- flights %>%
select(passengers, distance, freight, mail) %>% #,class, carrier_type, month_name, distance_group) %>%
summary()
print(summary_stats)## passengers distance freight mail
## Min. : 20.0 Min. : 50.0 Min. : 0.0 Min. : 0.0
## 1st Qu.: 59.0 1st Qu.: 350.0 1st Qu.: 0.0 1st Qu.: 0.0
## Median : 142.0 Median : 672.0 Median : 0.0 Median : 0.0
## Mean : 271.3 Mean : 820.1 Mean : 458.4 Mean : 576.9
## 3rd Qu.: 412.0 3rd Qu.:1114.0 3rd Qu.: 0.0 3rd Qu.: 0.0
## Max. :1000.0 Max. :5071.0 Max. :199648.0 Max. :173669.02. Multi-variate Analysis
# Scatter plot with alpha for overplotting
p1 <- ggplot(flights, aes(x = distance, y = passengers)) +
geom_point(alpha = 0.3, size = 1, color = "darkblue") +
labs(title = "Passengers vs Distance",
subtitle = "Scatter plot showing relationship between flight distance and passenger count",
x = "Distance (miles)",
y = "Number of Passengers") +
scale_y_continuous(labels = scales::comma) +
scale_x_continuous(labels = scales::comma) +
theme(plot.title = element_text(face = "bold", size = 14))
print(p1)
# Hexbin plot for dense data
p2 <- ggplot(flights, aes(x = distance, y = passengers)) +
geom_hex(bins = 50) +
scale_fill_viridis_c(name = "Number of\nFlights") +
labs(title = "Passengers vs Distance",
subtitle = "Hexbin plot showing density of flights",
x = "Distance (miles)",
y = "Number of Passengers") +
scale_y_continuous(labels = scales::comma) +
scale_x_continuous(labels = scales::comma) +
theme(plot.title = element_text(face = "bold", size = 14))
print(p2)
# Set theme
theme_set(theme_minimal())
## PLOT 1: Passengers vs Distance colored by Carrier Type
p1 <- ggplot(flights, aes(x = distance, y = passengers, color = carrier_type)) +
geom_point(alpha = 0.6, size = 1.5) +
scale_color_viridis_d(name = "Carrier Type") + # Changed to _d for discrete
labs(title = "Passengers vs Distance by Carrier Type",
subtitle = "Relationship between flight distance and passenger count across different carrier types",
x = "Distance (miles)",
y = "Number of Passengers") +
scale_y_continuous(labels = scales::comma) +
scale_x_continuous(labels = scales::comma) +
theme(plot.title = element_text(face = "bold", size = 14),
plot.subtitle = element_text(size = 10, color = "gray50"),
legend.position = "bottom") +
guides(color = guide_legend(nrow = 2, byrow = TRUE))
print(p1)
## PLOT 2: Freight vs Distance by Carrier Type
p2 <- ggplot(flights, aes(x = distance, y = freight, color = carrier_type)) +
geom_point(alpha = 0.7, size = 1.5) +
labs(title = "Freight vs Distance by Carrier Type",
subtitle = "Relationship between flight distance and freight volume across different carrier types",
x = "Distance (miles)",
y = "Freight Volume") +
scale_y_continuous(labels = scales::comma) +
scale_x_continuous(labels = scales::comma) +
theme(plot.title = element_text(face = "bold", size = 14),
plot.subtitle = element_text(size = 10, color = "gray50"),
legend.position = "bottom") +
guides(color = guide_legend(nrow = 2, byrow = TRUE))
print(p2)
# Distance Distribution by Month
p1 <- ggplot(flights, aes(x = factor(month), y = distance, fill = factor(month))) +
geom_boxplot(alpha = 0.7) +
labs(title = "Distance Distribution by Month",
subtitle = "Boxplot showing distance variations across months",
x = "Month", y = "Distance (miles)") +
scale_fill_viridis_d() +
guides(fill = "none") +
theme(axis.text.x = element_text(angle = 0, hjust = 0.5))
print(p1)
# Line plot with confidence intervals showing monthly trends
p3 <- flights %>%
group_by(month) %>%
summarise(
avg_distance = mean(distance, na.rm = TRUE),
se_distance = sd(distance, na.rm = TRUE) / sqrt(n())
) %>%
ggplot(aes(x = month, y = avg_distance)) +
geom_line(color = "steelblue", size = 1) +
geom_point(color = "steelblue", size = 2) +
geom_ribbon(aes(ymin = avg_distance - 1.96*se_distance,
ymax = avg_distance + 1.96*se_distance),
alpha = 0.2, fill = "steelblue") +
labs(title = "Average Distance by Month",
subtitle = "Line plot with confidence intervals showing monthly trends",
x = "Month", y = "Average Distance (miles)") +
scale_x_continuous(breaks = 1:12)+ ylim(500, NA)## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.print(p3)
# First aggregate the data
monthly_agg <- flights %>%
count(month, carrier_type) %>%
mutate(month_name = factor(month.abb[month], levels = month.abb))
p2 <- ggplot(monthly_agg,
aes(x = month_name, y = n,
color = carrier_type, group = carrier_type)) +
geom_line(size = 1, alpha = 0.8) +
geom_point(size = 2) +
scale_color_brewer(palette = "Set1", name = "Carrier Type") +
labs(title = "Monthly Flight Trends by Carrier Type",
subtitle = "Line chart showing seasonal patterns",
x = "Month",
y = "Number of Flights") +
scale_y_continuous(labels = scales::comma) +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
print(p2)
summary(monthly_agg)## month carrier_type n month_name
## Min. :1 Length:21 Min. : 645 Jan :3
## 1st Qu.:2 Class :character 1st Qu.: 784 Feb :3
## Median :4 Mode :character Median : 996 Mar :3
## Mean :4 Mean :1546 Apr :3
## 3rd Qu.:6 3rd Qu.:2566 May :3
## Max. :7 Max. :3199 Jun :3
## (Other):3# Ensure correct ordering
p3 <- flights %>%
mutate(passenger_group = case_when(
passengers <= 150 ~ "1. Small (≤150)",
passengers <= 300 ~ "2. Medium (151-300)",
passengers <= 500 ~ "3. Large (301-500)",
TRUE ~ "4. Very Large (>500)"
)) %>%
group_by(month, passenger_group) %>%
summarise(
avg_distance = mean(distance, na.rm = TRUE),
.groups = "drop"
) %>%
ggplot(aes(x = month, y = avg_distance, color = passenger_group, group = passenger_group)) +
geom_line(size = 1) +
geom_point(size = 2) +
labs(title = "Average Flight Distance by Passenger Group",
subtitle = "Grouped by passenger capacity",
x = "Month", y = "Average Distance (miles)") +
scale_x_continuous(breaks = 1:12, labels = month.abb) +
scale_color_brewer(palette = "Set1", name = "Passenger Group",
labels = c("Small (≤150)", "Medium (151-300)",
"Large (301-500)", "Very Large (>500)")) +
theme_minimal()
print(p3)
p1_labeled <- flights %>%
mutate(passenger_group = factor(case_when(
passengers <= 150 ~ "Small (≤150)",
passengers <= 300 ~ "Medium (151-300)",
passengers <= 500 ~ "Large (301-500)",
TRUE ~ "Very Large (>500)"
), levels = c("Small (≤150)", "Medium (151-300)", "Large (301-500)", "Very Large (>500)"))) %>%
count(carrier_type, passenger_group, month) %>%
mutate(month_name = factor(month.abb[month], levels = month.abb)) %>%
ggplot(aes(x = month_name, y = carrier_type, fill = n)) +
geom_tile(color = "white", size = 0.5) +
geom_text(aes(label = ifelse(n > 50, scales::comma(n), "")),
size = 2.8, color = "white", fontface = "bold") +
facet_wrap(~ passenger_group, ncol = 2) +
scale_fill_gradient(
name = "Flights",
low = "skyblue",
high = "darkblue",
trans = "sqrt",
labels = scales::comma
) +
labs(title = "Monthly Flight Distribution by Carrier and Passenger Group",
x = "Month", y = "Carrier Type") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
print(p1_labeled)
p2 <- flights %>%
group_by(carrier_type, month) %>%
summarise(total_passengers = sum(passengers, na.rm = TRUE), .groups = "drop") %>%
mutate(month_name = factor(month.abb[month], levels = month.abb)) %>%
ggplot(aes(x = month_name, y = total_passengers, fill = carrier_type)) +
geom_bar(stat = "identity", position = "dodge", width = 0.7) +
scale_fill_brewer(palette = "Set2", name = "Carrier Type") +
scale_y_continuous(labels = scales::comma, expand = expansion(mult = c(0, 0.1))) +
labs(title = "Monthly Passenger Volume by Carrier Type",
subtitle = "Total passengers transported each month",
x = "Month", y = "Total Passengers") +
theme_minimal() +
theme(
axis.text.x = element_text(angle = 45, hjust = 1),
plot.title = element_text(face = "bold", size = 16),
plot.subtitle = element_text(size = 12, color = "gray50")
)
print(p2)
p4 <- ggplot(flights, aes(x = carrier_type, y = passengers, fill = carrier_type)) +
geom_boxplot(alpha = 0.7) +
labs(title = "Passenger Distribution by Carrier Type",
x = "Carrier Type", y = "Passengers") +
scale_y_continuous(labels = scales::comma) +
theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
scale_fill_brewer(palette = "Set2")
print(p4)
# Prepare the data
top_routes_10 <- flights %>%
group_by(origin_city_name, dest_city_name) %>%
summarise(
total_passengers = sum(passengers, na.rm = TRUE),
total_flights = n(),
#avg_passengers = mean(passengers, na.rm = TRUE),
.groups = "drop"
) %>%
arrange(desc(total_passengers)) %>%
head(10) %>%
mutate(route_label = paste(origin_city_name, "→", dest_city_name))
p11_clean <- ggplot(top_routes_10,
aes(x = reorder(route_label, total_passengers),
y = total_passengers)) +
geom_col(fill = "lightblue", width = 0.6, alpha = 0.8) +
geom_text(aes(label = scales::comma(total_passengers)),
hjust = -0.2, size = 4.5, fontface = "bold", color = "#2E86AB") +
geom_text(aes(label = paste("(", total_flights, "flights)")),
hjust = 1.1, size = 3.5, color = "darkblue") +
labs(title = "Top 10 Busiest Air Routes",
subtitle = "Passenger volume and flight frequency",
x = NULL, y = NULL) +
scale_y_continuous(labels = NULL, expand = expansion(mult = c(0, 0.25))) +
coord_flip() +
theme_minimal() +
theme(
plot.title = element_text(face = "bold", size = 20, hjust = 0.5),
plot.subtitle = element_text(size = 14, color = "darkblue", hjust = 0.5),
axis.text.y = element_text(size = 13, face = "bold", margin = margin(r = 15)),
panel.grid = element_blank(),
axis.text.x = element_blank(),
axis.ticks = element_blank()
)
print(p11_clean)
# Get top 10 carriers by passenger volume
top_carriers <- flights %>%
group_by(unique_carrier_name) %>%
summarise(
total_passengers = sum(passengers, na.rm = TRUE),
total_flights = n(),
avg_passengers = mean(passengers, na.rm = TRUE),
.groups = "drop"
) %>%
arrange(desc(total_passengers)) %>%
head(10)
# Create the plot
p_carriers <- ggplot(top_carriers,
aes(x = reorder(unique_carrier_name, total_passengers),
y = total_passengers)) +
geom_col(fill = "#4C72B0", width = 0.7, alpha = 0.8) +
geom_text(aes(label = scales::comma(total_passengers)),
hjust = -0.1, size = 4, fontface = "bold", color = "#4C72B0") +
geom_text(aes(label = paste("(", total_flights, "flights)")),
hjust = 1.1, size = 3.5, color = "gray40") +
labs(title = "Top 10 Airlines by Passenger Volume",
subtitle = "Total passengers transported with flight count",
x = NULL,
y = NULL) +
scale_y_continuous(labels = NULL, expand = expansion(mult = c(0, 0.2))) +
coord_flip() +
theme_minimal() +
theme(
plot.title = element_text(face = "bold", size = 18, hjust = 0.5),
plot.subtitle = element_text(size = 14, color = "gray50", hjust = 0.5),
axis.text.y = element_text(size = 12, face = "bold", margin = margin(r = 15)),
panel.grid = element_blank(),
axis.text.x = element_blank()
)
print(p_carriers)
# Select only numeric variables for correlation analysis
numeric_vars <- flights %>%
select(passengers, freight, mail, distance, month, distance_group)
# Correlation matrix
cor_matrix <- cor(numeric_vars, use = "complete.obs")
print(cor_matrix)## passengers freight mail distance month
## passengers 1.00000000 -0.02126790 -0.02126495 -0.08164203 -0.02140240
## freight -0.02126790 1.00000000 0.25377673 -0.05877046 0.02786811
## mail -0.02126495 0.25377673 1.00000000 -0.10700328 0.01025854
## distance -0.08164203 -0.05877046 -0.10700328 1.00000000 0.02446867
## month -0.02140240 0.02786811 0.01025854 0.02446867 1.00000000
## distance_group -0.08671762 -0.04834417 -0.08833010 0.97416571 0.02670123
## distance_group
## passengers -0.08671762
## freight -0.04834417
## mail -0.08833010
## distance 0.97416571
## month 0.02670123
## distance_group 1.00000000# Correlation plot
corrplot(cor_matrix, method = "color", type = "upper",
tl.cex = 0.8, tl.col = "black",
title = "Correlation Matrix of Flight Variables",
mar = c(0,0,1,0))
# Scatterplot matrix for key variables
pairs(numeric_vars[,1:4], pch = 19, cex = 0.3,
main = "Scatterplot Matrix: Passengers, Freight, Mail, Distance")
# Focus on passenger relationships
library(ggplot2)
ggplot(flights, aes(x = distance, y = passengers)) +
geom_point(alpha = 0.3, color = "steelblue") +
geom_smooth(method = "lm", color = "red") +
labs(title = "Passengers vs Distance",
subtitle = paste("Correlation:", round(cor(flights$distance, flights$passengers, use = "complete.obs"), 3)))## `geom_smooth()` using formula = 'y ~ x'
3. Correlation
# 1. Convert categorical variables to numeric for
flights_cor <- flights %>%
mutate(
carrier_type_num = as.numeric(as.factor(carrier_type)),
distance_cat_num = as.numeric(as.factor(distance_cat)),
month_num = as.numeric(as.factor(month_name)),
origin_num = as.numeric(as.factor(origin_city_name)),
dest_num = as.numeric(as.factor(dest_city_name)),
class_num = as.numeric(as.factor(class))
) %>%
select(passengers, freight, mail, distance,
carrier_type_num, distance_cat_num, month_num, origin_num, dest_num, class_num)
# 2. Correlation matrix with categoricals
cor_matrix_full <- cor(flights_cor)
print(cor_matrix_full)## passengers freight mail distance
## passengers 1.00000000 -0.021267897 -0.021264951 -0.081642030
## freight -0.02126790 1.000000000 0.253776729 -0.058770464
## mail -0.02126495 0.253776729 1.000000000 -0.107003279
## distance -0.08164203 -0.058770464 -0.107003279 1.000000000
## carrier_type_num -0.01081651 0.087823039 0.179941829 -0.415237133
## distance_cat_num 0.07333467 0.065004507 0.083056913 -0.790855056
## month_num -0.02222689 0.004091336 -0.003150455 0.009557458
## origin_num 0.02626858 -0.031514731 -0.061274364 0.106696292
## dest_num 0.03114342 -0.003693259 0.013122768 0.092808880
## class_num -0.35672610 0.001549364 -0.062322734 -0.009161023
## carrier_type_num distance_cat_num month_num origin_num
## passengers -0.01081651 0.073334667 -0.022226885 0.02626858
## freight 0.08782304 0.065004507 0.004091336 -0.03151473
## mail 0.17994183 0.083056913 -0.003150455 -0.06127436
## distance -0.41523713 -0.790855056 0.009557458 0.10669629
## carrier_type_num 1.00000000 0.352710957 -0.012688904 -0.06183825
## distance_cat_num 0.35271096 1.000000000 -0.007820559 -0.09348377
## month_num -0.01268890 -0.007820559 1.000000000 0.00180490
## origin_num -0.06183825 -0.093483769 0.001804900 1.00000000
## dest_num -0.06231387 -0.080953230 0.002099892 0.01440191
## class_num 0.05849119 -0.006184624 -0.003218176 -0.05693190
## dest_num class_num
## passengers 0.031143422 -0.356726096
## freight -0.003693259 0.001549364
## mail 0.013122768 -0.062322734
## distance 0.092808880 -0.009161023
## carrier_type_num -0.062313870 0.058491191
## distance_cat_num -0.080953230 -0.006184624
## month_num 0.002099892 -0.003218176
## origin_num 0.014401908 -0.056931902
## dest_num 1.000000000 -0.057350874
## class_num -0.057350874 1.000000000# 3. Visualize full correlation matrix
corrplot(cor_matrix_full, method = "color", type = "upper",
tl.cex = 0.7, tl.col = "black",
title = "Correlation Matrix Including Categorical Variables",
mar = c(0,0,2,0))
# Which carrier types and distance categories have the most passengers?
flights %>%
group_by(carrier_type) %>%
summarise(mean_passengers = mean(passengers)) %>%
arrange(desc(mean_passengers))## # A tibble: 3 × 2
## carrier_type mean_passengers
## <chr> <dbl>
## 1 national carriers 364.
## 2 major carriers 256.
## 3 regional carriers 201.flights %>%
group_by(distance_cat) %>%
summarise(mean_passengers = mean(passengers)) %>%
arrange(desc(mean_passengers))## # A tibble: 11 × 2
## distance_cat mean_passengers
## <chr> <dbl>
## 1 4000-4499 miles 558.
## 2 500-999 miles 302.
## 3 less than 500 miles 276.
## 4 1000-1499 miles 267.
## 5 4500-4999 miles 226.
## 6 2000-2499 miles 196.
## 7 1500-1999 miles 188.
## 8 3000-3499 miles 186.
## 9 2500-2999 miles 178.
## 10 3500-3999 miles 94.9
## 11 5000-5499 miles 47# Plot 1: Passengers by Carrier Type
p1 <- ggplot(flights, aes(x = carrier_type, y = passengers, fill = carrier_type)) +
geom_boxplot(alpha = 0.8) +
stat_summary(fun = mean, geom = "point", shape = 18, size = 3, color = "red") +
labs(title = "Passenger Distribution by Carrier Type",
subtitle = "ANOVA: F = 615.1, p < 0.001 ***",
x = "Carrier Type", y = "Number of Passengers") +
scale_fill_brewer(palette = "Set2") +
theme_minimal() +
theme(legend.position = "none")
# Plot 2: Passengers by Distance Category
p2 <- ggplot(flights, aes(x = distance_cat, y = passengers, fill = distance_cat)) +
geom_boxplot(alpha = 0.8) +
stat_summary(fun = mean, geom = "point", shape = 18, size = 3, color = "red") +
labs(title = "Passenger Distribution by Distance Category",
subtitle = "ANOVA: F = 93.51, p < 0.001 ***",
x = "Distance Category", y = "Number of Passengers") +
scale_fill_brewer(palette = "Set3") +
theme_minimal() +
theme(legend.position = "none",
axis.text.x = element_text(angle = 45, hjust = 1))
# Display plots
library(patchwork)
(p1 | p2) 
monthly_trends <- flights %>%
group_by(month) %>%
summarise(
mean_passengers = mean(passengers),
ci_lower = mean_passengers - 1.96 * (sd(passengers) / sqrt(n())),
ci_upper = mean_passengers + 1.96 * (sd(passengers) / sqrt(n())),
.groups = "drop"
) %>%
mutate(month_name = factor(month.abb[month], levels = month.abb))
ggplot(monthly_trends, aes(x = month_name, y = mean_passengers, group = 1)) +
geom_ribbon(aes(ymin = ci_lower, ymax = ci_upper),
fill = "lightblue", alpha = 0.3) +
geom_line(color = "steelblue", size = 1) +
geom_point(color = "darkblue", size = 2) +
ylim(20, NA) +
labs(title = "Monthly Passenger Trends with 95% CI",
subtitle = "Shaded area shows uncertainty in monthly estimates",
x = "Month", y = "Average Passengers")
# Create aggregated data with both flight counts and passenger info
flight_data <- flights %>%
group_by(carrier_type, month) %>% # grouping variables
summarise(
flight_count = n(), # Number of flights
total_passengers = sum(passengers), # Total passengers
avg_passengers = mean(passengers), # Average per flight
.groups = "drop"
)
p3 <- ggplot(flight_data,
aes(x = total_passengers, y = flight_count,
color = carrier_type)) +
# Size only on points, not global
geom_point(aes(size = avg_passengers), alpha = 0.7) +
# No size in geom_smooth
geom_smooth(method = "lm", se = FALSE, aes(group = carrier_type)) +
labs(title = "Flights vs Total Passengers",
subtitle = "Bubble size = Average passengers per flight",
x = "Total Passengers", y = "Number of Flights") +
scale_size_continuous(name = "Avg Passengers") +
scale_x_continuous(labels = scales::comma) +
scale_y_continuous(labels = scales::comma) +
theme_minimal()
print(p3)## `geom_smooth()` using formula = 'y ~ x'
Key Findings:
Passenger loads are right‑skewed—most flights operate below capacity.
Cargo operations are highly specialized; many flights carry no freight/mail. Medium‑haul routes dominate (>75% of flights), reflecting hub‑and‑spoke networks.
Route demand is directional: Washington ↔︎ Pittsburgh has asymmetric passenger/flight counts.
Carrier specialization by distance: regional (<300 mi), national (mid‑haul), major (long‑haul).
Seasonal distance shift: short flights peak in February; longer routes increase in July.
Major carriers lead in flights and passengers across all months, peaking in January.
Regional/national carriers excel in small‑aircraft ops, not jumbo jets.
Large‑capacity flights (500+ passengers) maintain stable distances all year.
Highly fragmented market: 10,601 unique routes served by 77 carriers.