Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Passenger Satisfaction Prediction: Asaaju Babatunde

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 10

1

PASSENGER SATISFACTION
PREDICTION

A S A A J U B A B AT U N D E
2

PRESENTATION OBJECTIVE

The main aim of this presentation is to explore marketing survey DATA


on customer satisfaction for business sustainable growth and
profitability.
3

STATEMENT OF THE PROBLEM


The airline industry like every other business is faced with opportunities (high traffic due to
increase in demand) as well as challenges from competition. And all players in the industry are
required to continuously innovating in quality of services and technology enhancement to stay
in business.

However, the marketing survey feedback of about 90,917 passengers on service


SATISFACTION is about average response rate.

60

50

40

Rate
30

20

10

0
Satisfaction Dissatisfaction or Neutral
4

POSSIBLE CONTRIBUTING FEATURES

Online Support Customer Type Class Gender Age

Departure/Arrival
Travel Type Seat Comfort Flight Distance Legroom Service
Convenient

Departure Inflight
Cleanliness Inflight Wifi Food and Drink
Delay-time Entertainment

Arrival
Check-In Services Gate Location Baggage Handling On-board Service
Delay-time

Ease of Online
Online Boarding
Booking
5

ANALYTICS FRAMEWORK

4
3
2
1 STEP 03
STEP 04
DEPLOYMENT.
STEP 02 Model Building
STEP 01 Data & Evaluation
Understanding Preparation
the data
6

INSIGHTS

Does the data collected represent the problem to be


What additional work required to make the data more useful?
solved? YES
• Correlation exists between Arrival & Departure Delay-time
• 90,917rows and 24 columns
• Outliers and Missing Value Treatments
• 4 variables with outliers
• Splitting the data into training and testing data
• 42,075 Missing Values
Understanding the data Data Pre-processing

In what way can the data be visualized to get the


answer required? “Can the model be put into practice? YES
• All the variables are statistically significant • 95% Sensitivity
• Supervised Learning Algorithms • 95% Accuracy
• Best Model: Random Forest
Deployment
Model Building Evaluation
7

10 TOP IMPORTANT FEATURES


M e a n D e c r e a s e A c c u r a c y

Departure
Baggage /Arrival Time
Handling Convenient
Ease of Online Food and Drink
Online Support Booking
Customer Type Travel Type Check-in Services

Seat Comfort In-Flight Entertainment


8
RECOMMENDATIONS

 The survey research method (quantitative) used requires improvement for


flexibility and simplicity to avoid recording large missing values in future.

 It is imperative for the airline management to understand which features or


parameters play important role in swaying passengers’ feedback positively.

 The airline management needs to innovate in both procedural and technological


services since almost all the features are statistically significant.
9

CREATIVE SOLUTIONS

Mobile APP Online Booking & Check-In 24/7 Online Support Customer Services

Acquire Modern Aircraft with Decent


Ergonomics Inflight Video Stream Entertainment

Anti-tamper Baggage Tagging


Nutritious food and drink

Improved Time Management


Staff Training on Online Support
10

THANK YOU

You might also like