Hotel Booking Demand Analysis - EDA Report

A comprehensive exploratory data analysis of hotel booking demand patterns

🎯 Introduction

The article titled "Hotel Booking Demand Datasets", authored by Antonio Nuno de Almeida, Ana Nunes, and Luis, addresses the importance and utility of datasets related to hotel booking demand. Published on January 2, 2019, this work operates within a multidisciplinary context that combines aspects of data analysis, hotel management, and machine learning techniques.

The research focuses on collecting and analyzing hotel booking data, enabling researchers and industry professionals to better understand consumer behavior patterns. Through the use of these datasets, predictive models can be developed to help anticipate demand, optimize revenue management, and improve customer experience. This approach not only benefits hotels in terms of operational efficiency but also contributes to more informed decision-making based on concrete data.

💡 Key Insight: The work highlights the relevance of hotel demand datasets as key tools for analysis and continuous improvement in the tourism industry.

🎯 Objectives

General Objectives

Conduct comprehensive EDA: Perform exploratory data analysis of the dataset, generating visualizations, preparing data, and extracting initial conclusions using R/RStudio
Explore consumer behavior: Investigate different aspects of consumer behavior in the hotel sector

🎯 Specific Objectives

Develop prediction models: Create models to classify the probability of hotel booking cancellations

📊 Dataset Overview

Dataset Statistics

Metric	Value
Total Records	119,390
Variables	32
Hotel Types	Resort & City Hotels
Time Period	July 1, 2015 - August 31, 2017

The dataset contains 119,390 records and 32 variables sourced from ScienceDirect, a peer-reviewed academic bibliography platform by Elsevier. The article describes two hotel demand datasets: one from a resort hotel ("Resort Hotel") and another from an urban hotel ("City Hotel").

🔍 Case Study Analysis

📚 Data Origin

👥 Authors: Antonio Nuno de Almeida, Ana Nunes, and Luis
📅 Publication Date: January 2, 2019
🔗 Source: ScienceDirect
🏛️ Institution: University Institute of Lisbon, Portugal

✅ Data Credibility

The data originates from queries made to the hotels' Property Management System (PMS), stored in SQL format from previous administrative records, ensuring reliability.

🎯 Applicable Use Cases

Stakeholder	Application
Hotel Chains	Portfolio optimization, customer satisfaction improvement, market expansion
Independent Hotels	Niche opportunities, service differentiation
Academic Researchers	Consumer behavior studies in tourism and hotel management
Tourism Organizations	Understanding tourist flow and policy development

💼 Business Problems Addressed

Improve hotel profitability through dynamic pricing strategies and promotions during low seasons
Enhance customer experience with personalized services for returning customers
Optimize revenue fluctuation with appropriate marketing strategies for low-activity periods

💡 Key Insights

Cancellation Patterns: High or low cancellation rates could indicate business problems. Potential factors include climate changes, high vacation seasons, overly flexible cancellation policies.

Dynamic Pricing Strategy: Hotels should implement pricing strategies considering both seasonality and room type, offering discounts or special packages during high seasons for less visited rooms.

Seasonality Analysis: Seasonality relates to both room types and seasons, allowing for price increases during high-demand periods.

Customer Loyalty: Due to the existence of new and repeat customers, special services could be created for returning guests - premium access to hotel events, coupons, or visit credits.

❓ Key Analysis Questions

How many bookings are made by hotel type?
Is demand increasing over time?
What are the high and low demand seasons?
How many bookings include children and/or babies?
Is parking space availability important?
In which months are most booking cancellations made?

📝 Dataset Variables

Variable Dictionary

Variable	Type	Description
`hotel`	Categorical	Hotel Type (Resort - City)
`is_canceled`	Numeric	Indicates if booking was canceled (1 Yes / 0 No)
`lead_time`	Numeric	Days between booking and arrival
`arrival_date_year`	Numeric	Year of arrival date
`arrival_date_month`	Categorical	Month of arrival date
`arrival_date_week_number`	Numeric	Week number of arrival year
`arrival_date_day_of_month`	Numeric	Day of month of arrival
`stays_in_weekend_nights`	Numeric	Number of weekend nights stayed
`stays_in_week_nights`	Numeric	Number of weekday nights stayed
`adults`	Numeric	Number of adults in booking
`children`	Numeric	Number of children in booking
`babies`	Numeric	Number of babies in booking
`meal`	Categorical	Type of meal booked
`country`	Categorical	Customer's country
`market_segment`	Categorical	Market segment of booking
`distribution_channel`	Categorical	Booking distribution channel
`is_repeated_guest`	Numeric	Indicates if returning guest
`previous_cancellations`	Numeric	Number of previous canceled bookings
`reserved_room_type`	Categorical	Reserved room type
`assigned_room_type`	Categorical	Assigned room type
`booking_changes`	Numeric	Number of booking changes
`deposit_type`	Categorical	Deposit type (No Deposit, Non Refund, Refundable)
`agent`	Numeric	ID of booking agent
`company`	Numeric	ID of managing company
`days_in_waiting_list`	Numeric	Days on waiting list
`customer_type`	Categorical	Customer type
`adr`	Numeric	Average Daily Rate
`required_car_parking_spaces`	Numeric	Number of required parking spaces
`total_of_special_requests`	Numeric	Number of special requests
`reservation_status`	Categorical	Reservation status (Check-Out, Canceled, No-Show)
`reservation_status_date`	Date	Reservation status date

📈 Exploratory Data Analysis {#exploratory-data-analysis}

🔧 Data Loading

setwd("C:/Users/Usuario/OneDrive - Universidad Peruana de Ciencias/DATA SCIENCE")

datahotel <- read.csv("hotel_bookings.csv",header=T,sep = ",")

🔍 Data Inspection

Dataset dimensions

dim(datahotel)

[1] 119390     32

Column names

colnames(datahotel)
[1] "hotel"                         
 [2] "is_canceled"                   
 [3] "lead_time"                     
 [4] "arrival_date_year"             
 [5] "arrival_date_month"            
 [6] "arrival_date_week_number"      
 [7] "arrival_date_day_of_month"     
 [8] "stays_in_weekend_nights"       
 [9] "stays_in_week_nights"          
[10] "adults"                        
[11] "children"                      
[12] "babies"                        
[13] "meal"                          
[14] "country"                       
[15] "market_segment"                
[16] "distribution_channel"          
[17] "is_repeated_guest"             
[18] "previous_cancellations"        
[19] "previous_bookings_not_canceled"
[20] "reserved_room_type"            
[21] "assigned_room_type"            
[22] "booking_changes"               
[23] "deposit_type"                  
[24] "agent"                         
[25] "company"                       
[26] "days_in_waiting_list"          
[27] "customer_type"                 
[28] "adr"                           
[29] "required_car_parking_spaces"   
[30] "total_of_special_requests"     
[31] "reservation_status"            
[32] "reservation_status_date"

Dataset structure

str(datahotel)
'data.frame':	119390 obs. of  32 variables:
 $ hotel                         : chr  "Resort Hotel" "Resort Hotel" "Resort Hotel" "Resort Hotel" ...
 $ is_canceled                   : int  0 0 0 0 0 0 0 0 1 1 ...
 $ lead_time                     : int  342 737 7 13 14 14 0 9 85 75 ...
 $ arrival_date_year             : int  2015 2015 2015 2015 2015 2015 2015 2015 2015 2015 ...
 $ arrival_date_month            : chr  "July" "July" "July" "July" ...
 $ arrival_date_week_number      : int  27 27 27 27 27 27 27 27 27 27 ...
 $ arrival_date_day_of_month     : int  1 1 1 1 1 1 1 1 1 1 ...
 $ stays_in_weekend_nights       : int  0 0 0 0 0 0 0 0 0 0 ...
 $ stays_in_week_nights          : int  0 0 1 1 2 2 2 2 3 3 ...
 $ adults                        : int  2 2 1 1 2 2 2 2 2 2 ...
 $ children                      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ babies                        : int  0 0 0 0 0 0 0 0 0 0 ...
 $ meal                          : chr  "BB" "BB" "BB" "BB" ...
 $ country                       : chr  "PRT" "PRT" "GBR" "GBR" ...
 $ market_segment                : chr  "Direct" "Direct" "Direct" "Corporate" ...
 $ distribution_channel          : chr  "Direct" "Direct" "Direct" "Corporate" ...
 $ is_repeated_guest             : int  0 0 0 0 0 0 0 0 0 0 ...
 $ previous_cancellations        : int  0 0 0 0 0 0 0 0 0 0 ...
 $ previous_bookings_not_canceled: int  0 0 0 0 0 0 0 0 0 0 ...
 $ reserved_room_type            : chr  "C" "C" "A" "A" ...
 $ assigned_room_type            : chr  "C" "C" "C" "A" ...
 $ booking_changes               : int  3 4 0 0 0 0 0 0 0 0 ...
 $ deposit_type                  : chr  "No Deposit" "No Deposit" "No Deposit" "No Deposit" ...
 $ agent                         : chr  "NULL" "NULL" "NULL" "304" ...
 $ company                       : chr  "NULL" "NULL" "NULL" "NULL" ...
 $ days_in_waiting_list          : int  0 0 0 0 0 0 0 0 0 0 ...
 $ customer_type                 : chr  "Transient" "Transient" "Transient" "Transient" ...
 $ adr                           : num  0 0 75 75 98 ...
 $ required_car_parking_spaces   : int  0 0 0 0 0 0 0 0 0 0 ...
 $ total_of_special_requests     : int  0 0 0 0 1 1 0 1 1 0 ...
 $ reservation_status            : chr  "Check-Out" "Check-Out" "Check-Out" "Check-Out" ...
 $ reservation_status_date       : chr  "2015-07-01" "2015-07-01" "2015-07-02" "2015-07-02" ...

Data Processing

Missing Data Identification

colSums(is.na(hotel_data))
                         hotel                    is_canceled 
                             0                              0 
                     lead_time              arrival_date_year 
                             0                              0 
            arrival_date_month       arrival_date_week_number 
                             0                              0 
     arrival_date_day_of_month        stays_in_weekend_nights 
                             0                              0 
          stays_in_week_nights                         adults 
                             0                              0 
                      children                         babies 
                             4                              0 
                          meal                        country 
                             0                              0 
                market_segment           distribution_channel 
                             0                              0 
             is_repeated_guest         previous_cancellations 
                             0                              0 
previous_bookings_not_canceled             reserved_room_type 
                             0                              0 
            assigned_room_type                booking_changes 
                             0                              0 
                  deposit_type                          agent 
                             0                              0 
                       company           days_in_waiting_list 
                             0                              0 
                 customer_type                            adr 
                             0                              0 
   required_car_parking_spaces      total_of_special_requests 
                             0                              0 
            reservation_status        reservation_status_date 
                             0                              0

Missing Data Treatment

# Create helper dataset "dataset_clean"
dataset_clean <- na.omit(hotel_data)
# Remove columns with more than 50% NA values
dataset_clean <- hotel_data[, colMeans(is.na(hotel_data)) < 0.5]
View(dataset_clean)

Outlier Identification

# Create boxplot to visualize outliers
boxplot(dataset_clean$adr, main = "Box Plot - ADR")

# Identify outliers using Interquartile Range (IQR)
Q1 <- quantile(dataset_clean$adr, 0.25, na.rm = TRUE)
Q3 <- quantile(dataset_clean$adr, 0.75, na.rm = TRUE)
IQR <- Q3 - Q1

# Define limits
lower_bound <- Q1 - 1.5 * IQR
upper_bound <- Q3 + 1.5 * IQR
outliers <- dataset_clean$adr[dataset_clean$adr < lower_bound | dataset_clean$adr > upper_bound]

Statistic	Value
IQR	56.7
Lower Bound	-15.8
Upper Bound	211
Q1	69.3
Q3	126
Outliers Count	3,793

Outlier Treatment

# Winsorization to limit outliers
install.packages("DescTools")
library(DescTools)

# Calculate 5th and 95th percentiles
lower_bound <- quantile(dataset_clean$adr, 0.05, na.rm = TRUE)
upper_bound <- quantile(dataset_clean$adr, 0.95, na.rm = TRUE)

# Manual winsorization by replacing values outside limits
dataset_clean$adr[dataset_clean$adr < lower_bound] <- lower_bound
dataset_clean$adr[dataset_clean$adr > upper_bound] <- upper_bound

# Re-run BoxPlot
boxplot(dataset_clean$adr, main = "Box Plot - ADR (After Winsorization)")

Data Visualization

Question 1: Bookings by Hotel Type

hotel_count <- hotel_data %>%
  group_by(hotel) %>%
  summarise(count = n())

ggplot(hotel_count, aes(x = hotel, y = count, fill = hotel)) +
  geom_bar(stat = "identity") +
  labs(title = "Number of Bookings by Hotel Type", x = "Hotel Type", y = "Number of Bookings") +
  theme_minimal()

Question 2: Demand Over Time

# Convert month name to number if necessary
hotel_data$arrival_date_month <- match(hotel_data$arrival_date_month, month.name)

# Create 'year_month' column combining year and month
hotel_data$year_month <- make_date(hotel_data$arrival_date_year, hotel_data$arrival_date_month, 1)

# Verify year_month column creation
head(hotel_data$year_month)

# Group by year_month and count bookings
demand_over_time <- hotel_data %>%
  group_by(year_month) %>%
  summarise(count = n())

# Visualization: line chart of demand over time
ggplot(demand_over_time, aes(x = year_month, y = count)) +
  geom_line(color = "blue") +
  labs(title = "Booking Demand Over Time", x = "Date", y = "Number of Bookings") +
  theme_minimal()

Question 3: Seasonal Patterns

seasonality <- hotel_data %>%
  group_by(arrival_date_month) %>%
  summarise(count = n()) %>%
  arrange(match(arrival_date_month, month.name))

ggplot(seasonality, aes(x = factor(arrival_date_month, levels = month.name), y = count, fill = count)) +
  geom_bar(stat = "identity") +
  labs(title = "Booking Seasonality", x = "Month", y = "Number of Bookings") +
  theme_minimal()

Question 4: Lowest Demand Period

lowest_demand_month <- seasonality[which.min(seasonality$count), ]
lowest_demand_month

Question 5: Bookings with Children/Babies

children_bookings <- hotel_data %>%
  filter(children > 0 | babies > 0)

nrow(children_bookings)

ggplot(children_bookings, aes(x = factor(children > 0, levels = c(TRUE, FALSE)), fill = babies > 0)) +
  geom_bar() +
  labs(title = "Bookings with Children and/or Babies", x = "Children Present", y = "Number of Bookings") +
  theme_minimal()

Question 6: Parking Space Importance

parking_spaces <- hotel_data %>%
  group_by(required_car_parking_spaces) %>%
  summarise(count = n())

ggplot(parking_spaces, aes(x = required_car_parking_spaces, y = count, fill = required_car_parking_spaces)) +
  geom_bar(stat = "identity") +
  labs(title = "Importance of Parking Spaces", x = "Number of Parking Spaces", y = "Number of Bookings") +
  theme_minimal()

Question 7: Monthly Cancellations

hotel_data$arrival_date_month <- match(hotel_data$arrival_date_month, month.name)

hotel_data$year_month <- make_date(hotel_data$arrival_date_year, hotel_data$arrival_date_month, 1)

cancellations <- hotel_data %>%
  filter(is_canceled == 1) %>%
  group_by(year_month) %>%
  summarise(cancel_count = n())

ggplot(cancellations, aes(x = year_month, y = cancel_count)) +
  geom_bar(stat = "identity", fill = "red") +
  labs(title = "Monthly Cancellations", x = "Month", y = "Number of Cancellations") +
  theme_minimal()

📊 Conclusions {#conclusions}

Hotel Type Analysis

Based on booking data by hotel type, we can conclude that city hotels tend to be more popular than resorts. This could be due to various factors such as central location, business travel convenience, and urban amenities.

Demand Trends

When analyzing booking demand over time, we observe a consistent increase in bookings for both hotel types, especially during vacation months and special events.

Seasonal Patterns

High seasons coincide with vacation months (June, July, and August) and major festive events, where higher booking volumes are observed. This pattern provides valuable insights for revenue management strategies.

💡 Recommendations {#recommendations}

Dynamic Pricing Strategy

Implement dynamic rates adjusted to seasonality, demand, and room type. During low seasons, offering specific promotions and discounts could incentivize bookings, while premium pricing during high seasons can maximize revenue.

Cancellation Policy Review

Review cancellation policies, especially in months where higher cancellation rates are detected. Consider implementing more restrictive policies during peak periods while maintaining flexibility during low-demand periods.

Customer Loyalty Programs

Create loyalty programs to reward returning customers, providing exclusive benefits such as discounts, room upgrades, event access, or personalized services. This can significantly improve customer retention and lifetime value.

Continuous Service Improvement

Implement a continuous improvement strategy in customer service based on analysis of special requests. This will help personalize each guest's experience, improving satisfaction and retention rates.

Data-Driven Decision Making

Establish regular monitoring and analysis cycles of booking patterns to quickly adapt to market changes and optimize operational strategies.

Bibliography

Antonio, N., De Almeida, A., & Nunes, L. (2018). Hotel booking demand datasets. Data In Brief, 22, 41-49. https://doi.org/10.1016/j.dib.2018.11.126
Gabara, E. (2024, July 18). What your hotel booking performance can tell you about demand patterns. Cloudbeds. https://www.cloudbeds.com/es/articulos/patrones-demanda-hotelera/
International, S. (2024, June 23). Hotel market research. SIS International Research. https://www.sisinternational.com/es/pericia/industrias/investigacion-de-mercado-hotelero/

Analysis Complete

This report was generated as part of the CC216 coursework for exploratory data analysis in the hospitality industry.

🔍 For more insights and detailed analysis, please refer to the complete dataset and R code implementations. Autor: Nicole Sihuincha

Name		Name	Last commit message	Last commit date
Latest commit History 11 Commits
Captura de pantalla 2024-09-27 010713.png		Captura de pantalla 2024-09-27 010713.png
Captura de pantalla 2024-09-27 011036.png		Captura de pantalla 2024-09-27 011036.png
README.md		README.md
best-hotel-booking-form-wordpress-plugins.jpg		best-hotel-booking-form-wordpress-plugins.jpg
hotel_bookings.csv		hotel_bookings.csv

Folders and files

Latest commit

History

Repository files navigation

Hotel Booking Demand Analysis - EDA Report

Table of Contents

🎯 Introduction

🎯 Objectives

General Objectives

🎯 Specific Objectives

📊 Dataset Overview

Dataset Statistics

🔍 Case Study Analysis

📚 Data Origin

✅ Data Credibility

🎯 Applicable Use Cases

💼 Business Problems Addressed

💡 Key Insights

❓ Key Analysis Questions

📝 Dataset Variables

Variable Dictionary

📈 Exploratory Data Analysis {#exploratory-data-analysis}

🔧 Data Loading

🔍 Data Inspection

Dataset dimensions

Column names

Dataset structure

Data Processing

Missing Data Identification

Missing Data Treatment

Outlier Identification

Outlier Treatment

Data Visualization

Question 1: Bookings by Hotel Type

Question 2: Demand Over Time

Question 3: Seasonal Patterns

Question 4: Lowest Demand Period

Question 5: Bookings with Children/Babies

Question 6: Parking Space Importance

Question 7: Monthly Cancellations

📊 Conclusions {#conclusions}

Hotel Type Analysis

Demand Trends

Seasonal Patterns

💡 Recommendations {#recommendations}

Dynamic Pricing Strategy

Cancellation Policy Review

Customer Loyalty Programs

Continuous Service Improvement

Data-Driven Decision Making

Bibliography

Analysis Complete

About

Topics

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Packages