Presented by
June 11th 2021
About Us!
Abracadata was founded in 2021 as a data science consultant that aims to provide the best business recommendations for your company. Our members consist of five, namely:
We thank all parties who have helped a lot in the process of completing this project, especially Mr. Hafizh who has given many inputs and suggestions during the work on this project.
Table Of Contents
Provide an understanding of the existing problems and data gathering.
Problem &
Data
The problem solving process from data exploration to model evaluation tests.
The Process
Business recommendations for existing problems.
Target
Abracadata
Problem Statement
Invistico Airline asked Abracadata as business consultant to provide them with business recommendations to increase their customer satisfaction level without reducing revenue.
Abracadata
We focus on finding what services need to be improved so that invistico customer satisfaction increases.
We are looking for the best way to improve the quality of service so that the number of satisfied customers increases.
With the increase of satisfied customers, it will increase the level of customer satisfaction without reducing revenue.
Issues
Abracadata
Considering the current condition where the effects of the Covid-19 pandemic have made business travel decline, according to the data, there was a 96% decline, therefore we decided to focus on personal travel because the data is more dominant.
*Data taken from : Investopedia.com
The Process
Abracadata
We focused on Eco & Eco Plus customer
because Business class have minimum data compared to Eco class.
Business
Eco
Eco Plus
1431.0
1242.0
17610.0
15395.0
2415.0
2094.0
Percentage satisfaction in feature class.
Business
Eco
Eco Plus
53.5
46.5
53.4
46.6
53.6
46.4
Dataset from Invistico Airlines
The percentage of Eco Plus class trips is the lowest when compared to other classes, as well as the satisfaction level which is still below the Business and Eco class, therefore we combine Eco and Eco Plus into one value in class features called Eco.
Abracadata
MODELLING
Get service that will increase customer satisfaction.
Eco
53.6
46.4
Satisfaction rate of Eco Class
ECO Class Optimalization
HyperParameter Tuning
To determine the right model, we have done pre-modelling based on five models namely Logistic Regression, Decision Tree, XGBoost, AdaBoost and Random Forest. From this model, we choose to use logistic regression because it is simple and easy to use for business.
Model
Random Forest
Logistic Regression
XGBoost
AdaBoost
Decision Tree
Accuracy
0.99
0.86
0.93
0.9
0.91
Precision
0.93
0.83
0.94
0.89
0.91
Model
Random Forest
Logistic Regression
XGBoost
AdaBoost
Decision Tree
Accuracy
0.93
0.85
0.93
0.9
0.89
Precision
0.94
0.84
0.94
0.89
0.89
Accuracy
is a metric used to correctly assess a prediction of satisfied or dissatisfied customers
What is Accuracy and Precision?
Precision
is a metric used to correctly assess a prediction of satisfied customers based on the total predictions of satisfied customers
.
.
Abracadata
Feature Importance
After the modeling process, we determine the important features that we will use as a reference for business recommendations. We also display various kinds of feature importance with different types of models.
“We provide business recommendations for Inflight Entertainment, Online Service and Seat Comfort as variables that need to be improved by Invistico to increase customer satisfaction without reducing revenue”
Abracadata
The Target
Abracadata
Inflight Entertainment
Seat Comfort
Online Service
The Target
Cozy at all cost
Impact on Satisfaction
| Condition | | | Satisfaction | |
01 | Current Eco Class Satisfied | | 46,4% | ||
02 | Impact on Eco Class Satisfied | | 71,8% | ||
03 | Impact on Invistico Customer Satisfaction | | 80,8% | ||
Impact on Revenue & Profit
Customer Lifetime Value | |
Average Ticket Price | Rp. 1.750.000,- |
Average Customer Flying Year | 3X |
Customer Retention | 6,09% |
Gross Profit Margin | 16,11% |
seat occupancy | 74% |
Customer Lifetime Value | 39.089,39 |
Number of customers | 8.585.906 |
Economy Class Income | Rp.45.076.006.500.000 |
2,30% Profit | Rp. 1.036.748.149.500 |
Investation | |
Inflight Entertainment | Rp.1.913.130.000.000,- |
Seat Comfort | Rp.6.658.577.451,- |
Onlinve Service | Rp.1.740.000.000,- |
TOTAL | Rp.1.921.528.577.451,- |
Rp. 2.073.496.299.000
Profit in 2 Years
Return On Investment (ROI)
In 2 years
Conclusion
Based on the explanation that we have done, we can conclude that by applying the business recommendations that we have conveyed, it is proven that Invistico airline can increase their level of satisfaction without reducing revenue. Based on ROI, in the next 2 years the investment issued by Invistico can be paid off