NARSEE MONJEE INSTITUTE OF

MANAGEMENT STUDIES

A
PROJECT REPORT ON

BRIEF ANALYSIS OF AI AND OPERATIONS IN SUPPLY CHAIN

MANAGEMENT FOR AIRLINES INDUSTRY

Submitted In The Partial Fulfillment of Requirement For The Award of The Degree of Masters in
Business Administration (Supply Chain Management)

SUBMITTED BY -
Mr. Harsh Devalia
SAP ID- 77221260836

DATE OF SUBMISSION -
29th November 2023
Domain – Supply Chain Management
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

ACKNOWLEDGEMENTS: -

I would like to express my sincere gratitude and appreciation to those who have contributed to the
successful completion of this project on the "Brief Analysis of AI And Operations in Supply Chain
Management for Airlines Industry."
I extend my heartfelt thanks to the faculty and mentors at NMIMS for their guidance and support
throughout the course of my research. Their expertise and valuable insights have played a crucial role
in shaping the direction and scope of this project.
I am also thankful to the society, for providing valuable information and insights during our
discussions and interviews. Their expertise in the field has greatly enriched the content of this report.
My sincere thanks go to my peers and friends for their encouragement and constructive feedback,
which have been instrumental in refining the ideas presented here.
Lastly, I express my gratitude to my family for their unwavering support and understanding
throughout this academic journey. Their encouragement has been a driving force behind my efforts.
Thank you to everyone who has been a part of this academic endeavor. Your support has been
invaluable.

2|Page
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

INDEX

Sr No Particular Page No
List of Tables 5
Objectives 6
Project Scope 7
Executive Summary 8-9
Methodology 10
Questionary 11
Hypothesis 17
1 Chapter- 1 Fuel Planning 18-26
1.1 Introduction 19
Challenges and Solution in Fuel Management Program
1.2 20
Demand Planning
1.2.1 20
Mathematical Model Demand Planning
1.2.2 23
Tendering And Procurement
1.2.3 24
Mathematical Mode Tendering and Procurement
1.2.4 25
2 Chapter- 2 Airline Catering 27-34
2.1 Flight Catering System 28
2.2 Mathematical Model 32

2.3 AI Integration 34

3 Chapter- 3 Crew Scheduling 35-42

3.1 Crew Scheduling Process In Indigo Airlines 36

Mathematical Model
3.2 41
AI Integration
3.3 42

3|Page
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

Sr No Particular Page No
4 Chapter- 4 Route Planning 43-50
4.1 Route Planning Process In Indigo Airlines 44
4.2 Mathematical Model 48

4.3 AI Integration 50

5 Chapter- 5 Crew Scheduling 51-58

5.1 Maintenance Plan 52

Mathematical Model
5.2 56
AI Integration
5.3 58

6 Chapter- 6 Findings 59-63

6.1 Comparison with and without AI 60

6.2 Sector Growth 61

6.3 Challenges in AI Implementation 63

7 Chapter-7 Bibliography 64-65

4|Page
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

LIST OF TABLES

TABLE NO DESCRIPTION PAGE NO

Table 3.1 Hypothetical Table for Crew Paring 30

Table 3.2 Crew Scheduling Software’s 31

Table 4.1 Flight Planning Software’s 41

Comparison of with and without AI

Table 6.1 49
integration in SCM Airlines
Table 6.2 AI Demand 50

5|Page
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

OBJECTIVES
This case study on IndiGo Airlines' supply chain management seeks to comprehensively analyze
critical operational facets, encompassing fuel optimization, inflight catering, baggage handling, crew
scheduling, route planning, and maintenance planning.
The primary objectives are to scrutinize the airline's fuel efficiency strategies, evaluate the
streamlined processes in inflight catering, assess the effectiveness of baggage handling protocols,
analyze crew scheduling practices, optimize route planning for efficiency, and scrutinize maintenance
planning strategies. The study aims to provide valuable insights into IndiGo's supply chain dynamics
and operational excellence in these key areas.
1) Evaluate AI Integration: Assess how AI technologies are integrated into various facets of supply
chain operations in the airline industry, including but not limited to demand forecasting, inventory
management, and logistics.

2) Identify Efficiency Gains: Identify and analyze the specific areas where AI contributes to
operational efficiency within the airline supply chain, such as optimizing route planning,
streamlining cargo handling, and improving overall resource allocation.

3) Assess Cost-effectiveness: Evaluate the cost-effectiveness of AI applications in supply chain

management, considering factors such as reduced operational costs, minimized errors, and
enhanced resource utilization.

6|Page
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

PROJECT SCOPE

This project's scope includes a thorough analysis of how Artificial Intelligence (AI) is incorporated into
airline supply chain management. The main goal is to investigate how artificial intelligence (AI)
technology may improve and change many aspects of airline supply chain operations. Demand
forecasting, inventory control, route optimization, and overall logistics efficiency are important
research topics.
The project's goal is to gain an understanding of the current state of artificial intelligence usage in
airline supply chains by analyzing case studies, industry reports, and existing literature. Through a
comprehensive analysis, the study will pinpoint instances of effective AI integration, obstacles
encountered, and developing patterns in relation to the aviation sector.
Examining real-world applications, such as pilot programs and case studies, will yield practical
insights. The study will evaluate how artificial intelligence (AI) affects the airline industry's total supply
chain performance, cost-effectiveness, and operational efficiency.
In addition, the study will explore technological aspects, covering topics like cybersecurity problems
related to AI adoption, data integration, and compatibility with older systems. The scope includes
making suggestions for resolving issues and putting AI solutions into practice that are compatible with
the particular dynamics of the airline sector.

7|Page
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

EXECUTIVE SUMMARY
This project report delves into the intricacies of IndiGo Airlines' supply chain operations, specifically
focusing on fuel optimization, crew scheduling, route planning, air catering, and maintenance
planning. These critical aspects are scrutinized through the lens of mathematical models, aiming to
enhance operational efficiency and resource utilization. Furthermore, the report illuminates the
pivotal role of Artificial Intelligence (AI) in optimizing these processes, driving improvements in
overall supply chain performance.
Fuel Optimization:
The report starts by dissecting IndiGo's fuel optimization strategies, employing mathematical models
to analyze factors such as aircraft weight, fuel consumption rates, and optimal flight paths. By utilizing
advanced algorithms, the airline seeks to minimize fuel expenses while adhering to safety and
regulatory standards. This section provides a detailed examination of the mathematical models used
and their impact on cost reduction and environmental sustainability.

Crew Scheduling:
The project extends its focus to crew scheduling, an integral element of efficient airline operations.
Mathematical models play a crucial role in optimizing crew assignments, considering factors like duty
time limitations, rest requirements, and skill sets. The report delves into how these models contribute
to achieving an optimal balance between operational needs, crew well-being, and regulatory
compliance.
Route Planning:
In the context of route planning, the report explores mathematical models that account for variables
such as air traffic, weather conditions, and fuel efficiency. IndiGo's emphasis on optimizing flight paths
not only contributes to cost reduction but also enhances punctuality and customer satisfaction. This
section elucidates the specific mathematical frameworks employed to determine the most efficient
routes for the airline's network.
Air Catering:
The project scrutinizes the intricacies of air catering supply chain operations at IndiGo, employing
mathematical models to optimize inventory management, procurement, and distribution. The report
details how AI algorithms analyze historical data, passenger preferences, and flight schedules to
enhance the efficiency of catering services. This optimization not only minimizes waste but also
ensures a seamless and tailored inflight dining experience for passengers.

8|Page
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

Maintenance Planning:
The maintenance planning segment investigates how mathematical models contribute to the
optimization of aircraft maintenance schedules. By predicting component lifespans, analyzing
historical performance data, and considering regulatory requirements, IndiGo aims to reduce
downtime and maintenance costs. The report outlines the specific mathematical approaches used
and their impact on the airline's overall operational reliability.
Role of AI in Optimization:
A pivotal aspect of this report is the exploration of how AI enhances optimization processes across
these supply chain operations. IndiGo leverages machine learning algorithms to analyze vast datasets,
predict trends, and dynamically adapt to changing conditions. The report elucidates how AI-driven
decision support systems contribute to real-time adjustments, improving the agility and
responsiveness of IndiGo's supply chain operations.

Conclusion:
In conclusion, this project report provides a comprehensive analysis of IndiGo Airlines' supply chain
operations with a focus on fuel optimization, crew scheduling, route planning, air catering, and
maintenance planning. The integration of mathematical models into these processes underscores the
airline's commitment to efficiency, cost-effectiveness, and customer satisfaction.
The role of AI emerges as a transformative force, driving continuous improvement and adaptability in
the dynamically evolving landscape of airline operations. As IndiGo continues to innovate and refine
its supply chain strategies, this report serves as a valuable resource for understanding the symbiotic
relationship between mathematical modeling, AI, and operational excellence in the aviation industry.

9|Page
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

METHOLODOGY: -
Research methodology refers to the systematic process of planning, executing, and analyzing
research. It involves defining research questions, selecting appropriate data collection methods,
gathering, and interpreting data, and drawing conclusions. A well-designed methodology ensures the
reliability and validity of research findings, contributing to the overall rigor of the study.
METHOD USED – MIXED APPROACH
The project's study methodology, "Brief Analysis of AI and Operations in Supply Chain Management
for the Airlines Industry," takes a mixed-method approach, combining qualitative and quantitative
methodologies to achieve a comprehensive knowledge of the complex dynamics at work.
Qualitative Data
In-depth interviews with key industry stakeholders and case studies of airlines that have adopted AI
in their supply chain operations are used to acquire qualitative data. These approaches enable a more
in-depth examination of the real-world obstacles, rewards, and strategic considerations connected
with integrating AI technologies. The study intends to unearth insights that may not be easily evident
in quantitative data through qualitative analysis, creating a richer narrative about the influence of AI
on supply chain management.
Quantitative Data
On the quantitative side, questionnaires are sent to a random sample of airlines in the industry. These
surveys are intended to collect structured data on supply chain key performance indicators (KPIs) such
as efficiency, cost reduction, and responsiveness. Furthermore, data analytics tools are used to
analyses past operational data, collecting trends and patterns that demonstrate the quantitative
impact of AI adoption.
The mixed-method approach combines the strengths of qualitative and quantitative research to
provide a thorough and well-rounded assessment of the topic. Qualitative data adds depth and
context, allowing for a more comprehensive understanding of the human and strategic aspects of AI
integration. In the meanwhile, quantitative data provides statistical by permitting the detection of
trends, correlations, and measurable results.
Charts Used in Analysis: -
1) Pie Chart
2) Bar Chart

10 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

Questionary List (Given To 15 Members Related to Aviation Sector, Including SCM Engineers, Pilots
and Collected the Data Which Was Mentioned Below in the Report)
1) What key aspects of supply chain operations are critical for Indigo Airlines' efficiency and
success?
a) Flight scheduling
b) Inventory management
c) Passenger booking
d) Marketing strategies
2) How has Indigo Airlines historically managed its inventory in the supply chain?
a) Manual tracking
b) Automated systems
c) Outsourced logistics
d) Combination of manual and automated methods
3) In the context of Indigo Airlines, which factor is most crucial for ensuring on-time delivery of
aircraft parts and components?
a) Weather conditions
b) Supplier reliability
c) Route optimization
d) Employee scheduling
4) How does Indigo Airlines currently handle demand forecasting in its supply chain?
a) Statistical models
b) Intuition-based decisions
c) Supplier recommendations
d) Customer feedback
5) Regarding AI integration in supply chain operations, what benefits might Indigo Airlines expect
to achieve?
a) Cost reduction
b) Faster delivery times
c) Enhanced decision-making
d) All of the above

11 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

6) What challenges do you anticipate Indigo Airlines might face in integrating AI into its supply chain
operations?
a) Resistance from employees
b) Technological constraints
c) Data security concerns
d) All of the above
7) How does Indigo Airlines prioritize sustainability in its supply chain operations?
a) Carbon offset programs
b) Eco-friendly packaging
c) Efficient route planning
d) All of the above
8) What role does technology play in enhancing the visibility of goods and materials across Indigo
Airlines' supply chain?
a) RFID tracking
b) Barcoding systems
c) GPS-enabled devices
d) All of the above
9) In terms of AI adoption, which specific supply chain functions might Indigo Airlines prioritize for
automation?
a) Procurement
b) Warehousing
c) Transportation
d) All of the above
10) How can the integration of AI in Indigo Airlines' supply chain contribute to a more resilient and
responsive operation?
a) Predictive maintenance
b) Real-time analytics
c) Demand sensing
d) All of the above

12 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

The above questionary was sent to 15 persons working in Indigo Airlines, from that we get
response from 7 peoples, from analysis of there submitted response, the interpretation was as
follows.
1) Critical Key Aspects of SCM

2) Historic Management of Inventory

3) Crucial Factor For On Time Delivery Of Aircraft Parts

13 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

4) Current Demand Planning Method

Predictive
Real-time Demand Planning Method
maintenance
analytics
0%0%

5) AI Benefits Indigo Might to Achive

6) Resistance while Implementing AI in SCM

14 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

7) Prioritize sustainability in its supply chain operations: -

8) Visibility of Goods and Material in SCM

9) Function of SCM in Priority to Automate

15 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

10) Integration of AI Can Benefit Operation

16 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

HYPOTHESIS
Hypothesis: The integration of artificial intelligence (AI) technologies in supply chain management
operations will significantly enhance efficiency, reliability, and cost-effectiveness in the airline
industry.
Explanation:
Operational Efficiency: It is hypothesized that the implementation of AI-driven technologies, such
as predictive analytics and machine learning algorithms, will streamline and optimize various
operational processes within the supply chain of the airline industry. This may include inventory
management, demand forecasting, route optimization, and maintenance scheduling, ultimately
leading to improved efficiency.
Enhanced Decision-Making: The hypothesis suggests that AI-powered tools can provide real-time
data analytics and insights, enabling better decision-making for supply chain managers in the
airline industry. Through data-driven decision support systems, the industry can respond more
effectively to dynamic factors such as weather disruptions, fuel price fluctuations, and changing
customer demands.
Reduced Costs and Resource Optimization: By leveraging AI technologies, the hypothesis
anticipates that airlines can identify areas for cost reduction and resource optimization. This may
involve minimizing excess inventory, optimizing fuel consumption through intelligent route
planning, and enhancing overall resource allocation to achieve a more cost-effective and
sustainable supply chain.
Improved Reliability and Risk Management: The hypothesis proposes that AI can contribute to
increased reliability in the supply chain by predicting and mitigating potential disruptions.
Machine learning models can analyze historical data to identify patterns and trends, allowing for
proactive risk management strategies that minimize the impact of unforeseen events on the
airline supply chain.
Customer Satisfaction and Service Levels: With AI-driven improvements in supply chain
management, the hypothesis suggests that airlines can enhance customer satisfaction by ensuring
on-time deliveries, minimizing delays, and offering a more seamless and responsive service. This
could lead to increased customer loyalty and positive brand perception.
Limitation: -
Here we can show the benefits and can draft the models but can’t run the AI deployment
because it needs addition AI team to make ML models.

17 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

CHAPTER- 1
AIRLINES FUEL PLANNING

18 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

1) FUEL OPTIMIZATION PROCESS FOR AIRLINES

1.1) INTRODUCTION: -
The airlines need to procure the jet fuel for their aircrafts. The procurement of jet fuel for the airlines
is one of the most important domains which should be critically handled otherwise there may be loss
of business unit. According to the IATA and ICAO reports, the fuel is the one of the airlines largest
expenses. The fuel expense is about 18 to 30 % of the total airlines operating cost.
The fuel management program of the airlines contains varieties of the in between activities which
are in short mentioned as below.

• The fuel planning process starts with demand forecasting, a task overseen by the SCM team.
They draw upon inventory levels and historical fuel ordering records to generate the precise
demand projection. This projection is indispensable, including insights gleaned from analyzing
past demand trends and evolving consumption patterns, essential for strategic fuel
management.

• The airline seeks new fuel suppliers or wishes to renew existing contracts, a formal tendering
process ensues. Vendors are invited to submit competitive bids, and the evaluation criteria
encompass both commercial factors, such as pricing and contractual terms, and technical
considerations to guarantee the quality and reliability of the fuel supply. The documentation
generated during this tendering phase is meticulously preserved for reference and compliance
purposes.

• One this demand forecast; it is forwarded to the procurement department. Here,

procurement specialists rigorously evaluate potential fuel vendors, considering quotations
and technical data. In some instances, a pre-existing annual purchase contract with a supplier
may lead to the issuance of a blanked Purchase Order (PO). This PO embodies essential data,
such as quantities, delivery schedules, and pricing terms.

• Once procurement complete, attention shifts to the efficient management of fuel supply
logistics. This may entail the management of dedicated storage facilities or, the supplier may
directly supply fuel in the tanker so that the storage inventory management for airlines
process is removed. The record-keeping of fuel receipts, quality checks, and adherence to
delivery schedules is imperative to guarantee seamless operations.

• After supply chain operations, effective financial management takes precedence. This
encompasses the comprehensive review and processing of supplier invoices, ensuring that
the contractual terms are fulfilled and accurate representation of the fuel received.
Furthermore, the finance department does the timely payments to suppliers, maintaining a
crucial financial flow within the fuel procurement process.

19 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

1.2) CHALLENGES AND SOLUTION IN FUEL MANAGEMENT PROGRAME: -

1.2.1) DEMAND PLANNING: -
Demand forecasting is a strategic
process that predicts future
consumer needs for a product or
service.
It employs historical data, market
analysis, and statistical models to
anticipate demand patterns, enabling
businesses to optimize inventory,
production, and distribution.
Accurate forecasting enhances
operational efficiency, reduces costs,
and ensures adequate stock levels,
ultimately improving customer
satisfaction.
This proactive approach empowers companies to adapt swiftly to market fluctuations and maintain a
competitive edge in meeting consumer demands. The airlines must do demand forecasting for the
estimation of their fuel.

CHALLENGES: -

• The planning is essential for estimating the demand of the fuel for each month. The fuel
forecasting is done based on the historic data available; the number of tickets booked the
consumptions; the calculation and analysis of the data is very challenging as the fuel data
comes from thousands of servers and excel spreadsheet and compiling each data manually is
very challenging. Also, the data analysis should be so accurate to predict the forecast. The data
models and AI will help in this scenario.

SOLUTION: -

• The Sample fuel consumption and fuel demand forecasting plan is attached in the below
section the number of routes is noted and the daily to and from flights are specified, the Indigo
airlines operated the Airbus A320 and Airbus A321 aircrafts, the fuel consumption per km per
seat is obtained from OEMs manual, from that data the total fuel required per day is calculated
and based on that the monthly demand of fuel is estimated.

20 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

• Although the forecasted plan in attached sheet is limited to few routes and is not including
with other parameters. The mathematical model is created with adding following parameters,
like weather condition in upcoming month, total numbers of routes in which the aircraft is
operated, the type of aircraft, historic fuel purchase, the ATC information of traffic at airports
on which the airlines are servicing, the sales of tickets forecasted in upcoming month. This
mathematical model is attached in below file.

• As the mathematical model is created the AI integration starts. It involves data processing,
model selecting, feature engineering (Create new features or transformations that might
improve model performance), Model validation (Train the selected machine learning model
on historical data. Use a portion of the data for training and reserve another portion for
validation, real time data mapping (implement mechanisms to continuously feed real-time
data into the AI model. This includes live weather updates, air traffic data, and aircraft
performance metrics.), Feedback loop (Establish a feedback loop to incorporate insights and
lessons learned from real-world operations into the AI model's training data)

• With the implementing of AI technology, the demand forecasting for the airlines is accurate
and the Just in Time can be accomplished for the inventory levels.

21 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

Fig- 1(A)
(AIRLINES SAMPLE FUEL DEMAND FORECASTING PER MONTH)
Hyperlink for the File: -

22 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

1.2.2) MATHEMATICAL MODEL FOR FUEL DEMAND FORECASTING: -

Fuel demand (D) depends on multiple factors including the number of flights, aircraft types, distance
traveled, weather conditions, economic indicators, and time trends.

MATHEMATICAL MODEL:
D = α * N + β1 * Type + β2 * Dp + β3 * Weather + β4 * Economic + β5 * Time + ε
Where:
✓ D is the fuel demand (in gallons or any appropriate unit).
✓ N is the number of flights.
✓ Type represents the aircraft type (a categorical variable).
✓ Dp is the average distance per flight (in miles).
✓ Weather includes various weather-related parameters (e.g., temperature, wind speed,
precipitation).
✓ Economic includes economic indicators (e.g., GDP, fuel prices).
✓ Time represents time-related variables (e.g., seasonality, trends).
✓ α, β1, β2, β3, β4, β5 are the model coefficients to be estimated through regression analysis.
✓ ε represents the error term (unexplained variation).

The variables α, β1, β2, β3, β4, β5 and ε are calculated by regression analysis of the historic data, the
coefficients are analyzed and substituted in the equation to find out the monthly demand of the fuel.
The python Jupiter notebook is used to find the values of following coefficients, for calculating the
coefficients the large number of historic data is to be needed. The sample code is written in below
section.

23 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

1.2.3) TENDERING AND PROCUREMENT: -

The tendering is the process in
which the airlines release the
tender for procuring the fuel as
the quantity obtained from
demand forecasting for the next
month. The vendor submits their
bids, according to the financial
and technical data, the airlines
select the best vendor and award
them the contract of suppling
the required quantity of fuel.
Once the vendor is selected the
airline award them the purchase
order in which the quantity and
price is clearly mentioned, the blanked PO is also offered in which only the price for suitable supplying
period is mentioned and the supplying of fuel quantity is done as and when needed by airlines on
that blanked PO price.

CHALLENGES: -

• Generally, the airlines use the manual process to analysis the bids from the vendors using the
excel formulas. This is very time consuming as there are large number of vendors across the
globe who are submitting their bids bases on different time zones and different currency’s.
The human error chances are also there in selection of wrong bids, also the transparency in
bidding process is also not obtained.
• The releasing of purchase order by manual process is also very time consuming, as there are
numbers of terms and conditions which need to be write in PO and the proper quantity w.r.t
price finalized in bid is to be described. This all needs a proper technical manpower.

SOLUTION: -

• The detailed mathematical model is prepared for the tendering analysis process, this
mathematical model is then trained with proper algorithm and live data to ensure the right
supplier is selected. Let’s plot the mathematical model as under.

24 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

1.2.4) MATHEMATICAL MODEL FOR TENDERING PROCESS: -

Let's define a utility function (U) that combines these factors to represent the overall value of a
quotation:
U=α * Price + β * Quality + γ * reliability
Where:
✓ Quotation Price (P) - The cost of fuel per unit (e.g., gallon or liter).
✓ Fuel Quality (Q) - A numerical rating representing the quality of the fuel. Higher values
indicate better quality.
✓ Supplier Reliability (R) - A numerical rating representing the reliability of the supplier. Higher
values indicate more reliable suppliers.
✓ α, β, and γ are weights that represent the importance of each factor. These weights should
add up to 0 to 1.
✓ To select the best quotation, calculate the utility score for each quotation and choose the one
with the highest score.
Selection Process:
Normalize the factors (Price, Quality, Reliability) to a common scale between 0 and 1. You can use
methods like min-max scaling for this purpose.

(A) Price Normalization:

• Determine the minimum and maximum values for the price among all quotations. Let's say
the minimum price is Pmin and the maximum price is Pmax.
• For each quotation's price Pi, calculate the normalized price using the min-max scaling
formula:
𝑃𝑖−𝑃𝑚𝑖𝑛
• 𝑁𝑃𝑖 =
𝑃𝑚𝑎𝑥−𝑃𝑚𝑖𝑛

(B) Quality Normalization:

• Quality values should scale them. Assuming quality values are already in the range [0, 1], no
further normalization is needed.

(C) Reliability Normalization:

• Similar to price, determine the minimum and maximum values for reliability among all
quotations. Let's say the minimum reliability is Rmin and the maximum reliability is Rmax.
• For each quotation's reliability (Ri), calculate the normalized reliability (NRi) using the min-
max scaling formula:
𝑅𝑖−𝑅𝑚𝑖𝑛
• 𝑁𝑅𝑖 =
𝑅𝑚𝑎𝑥−𝑅𝑚𝑖𝑛

So will have normalized values for price, quality, and reliability for each quotation, ensuring that
they are all on a common scale between 0 and 1. These normalized values can then be used in the
utility function to make fair comparisons and select the best quotation.

25 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

The weights α, β, and γ represent the importance or priority to assign to each factor (price, quality,
and reliability) when deciding. These weights are determined based on specific requirements,
objectives, and preferences.

Here is a general approach to calculating these weights:

(C) Identify Priorities:
Determine the relative importance of each factor in decision-making process. Consider
what matters most to organization or situation. For example, lets prioritize low cost
(price) over everything else, or might value quality and reliability equally.
(D) Assign Importance Scores:
Assign a numerical importance score to each factor based on priorities. These scores can
be on a scale of 0 to 1, where 0 indicates no importance, and 1 indicates maximum
importance.
Ensure that the sum of the importance scores equals 1 to maintain a proper weight
distribution.

Calculate the utility score for each quotation using the utility function:

Ui=α * Normalized Price(i)+ β * Normalized Quality(i) + γ * Normalized Reliability(i)

Where, I represent the quotation being evaluated.

Select the quotation with the highest utility score as the best choice

26 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

CHAPTER 2
FLIGHT CATERING SYSTEM

27 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

2) FLIGHT CATERING SYSTEM: -

The inflight catering system supply chain is quite interesting to know. The airlines follow series of
procedures to serve meal to the passengers. To provide this service, airlines depend more on the
aspects of hygiene, safety, security, production quality, just in time production etc.

Today's marketing is mainly based on marketing research applied to the experiences of customers in
various flight sectors by a certain airline. Passengers are most worried about safety, on-time
performance, scheduling/ticketing concerns, the aircraft's physical surroundings such as seat and leg
comfort, and gate check-in and boarding, according to passenger surveys (research at the University
of Surrey).

2.1) SUPPLY CHAIN DESIGNING OF FLIGHT CATERING SYSTEM: -

MARKETING ANALYSIS - STAGE-1)

Today's marketing is mainly reliant on marketing research applied to the experiences of customers
in various flight sectors by a certain airline. Passengers are most worried about safety, on-time
performance, scheduling/ticketing concerns, the aircraft's physical surroundings such as seat and leg
comfort, and gate check-in and boarding, according to passenger surveys (research at the University
of Surrey).
This suggests that, while food is important, it is unlikely to influence a passenger's choice of airline.

In-flight meals, in my opinion, do not influence pre-purchase decisions, but they do become a hugely
significant post-purchase component.

28 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

CATERING SPECIFICATIONS - IN-FLIGHT CATERING SERVICE

In-flight catering services, according to a legal source, refer to the delivery of prepared and packaged
food and beverages to any public airport for consumption aboard an aircraft while in flight.
The in-flight catering service strategy is based on the operational and marketing considerations listed
below:
1) On Aircraft class such as: -
• Economic Class.
• Business Class.
• First Class.

2) On Flight Type: -
• Full Service
• Low Cost

3) The flight length: -

• Long Haul Flight
• Short Haul Flight
Whether it is a short-haul flight or a long-haul flight, inflight catering service is an integral part of a
full-service airline flight plan. Airline passengers will get at least a cup of tea and something to eat
during their trip.
In this stage, the airline companies plan the food menus for each flight depending on the marketing
plan, the cost limitations, the passenger's preferences, the aircraft cabin class, and the aircraft kitchen
load limitations, etc.
IN-FLIGHT CATERING WORK FLOW – STAGE 2
The Suppliers
Food and non-food items are purchased from a variety of suppliers by airline catering providers. They
maintain an active interaction with the local supplier network, frequently calling tenders and
negotiating cheaper prices for high volume purchases for their aircraft catering business.
Most of the food and non-food goods purchased by caterers are homogeneous, and equivalents are
readily available in the market. Airline food providers typically choose suppliers who offer high-quality
products at a moderate cost.

29 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

The contract between the airline catering providers and the supplier contains a clear legal clause. The
contract places a high value on punctuality, and airline catering suppliers may levy penalties for late
deliveries while also reserving the right to select an alternate supplier in the event of delivery failures.
There is also a food safety, quality, and security protocol that the supplier must follow, such as having
a quality management system and a security program.

IN-FLIGHT CATERING WORKFLOW – STAGE 3

The Airline Catering Providers
The assembly processes. This procedure accepts one or more items as input and produces one or
more items as output. The main meal and numerous trays set components (jam, milk gigger, water
couplet, salad, and dessert) are consumed in the meal tray set assembling process. It then generates
a whole tray set unit, such as a hot breakfast tray set, a hot supper tray set, or a cabin/tech crew meal.
Menu planning is a vital activity in the planning phase of an airline's catering system. Airlines and
their food providers collaborate to suit the needs of onboard passengers. Another important action
in this process is galley stowage planning.
The airline catering providers must have an engaging work relationship with many external suppliers
to find the right products; both customers nominated products and sourced products to meet the
airline companies’ expectations.

IN-FLIGHT CATERING WORK FLOW – STAGE 4

Airline companies are responsible for the design of onboard service. This is affected by the time of
flight, length of flight, point of embarkation and disembarkation, nationality or ethnicity of
passengers, seat class (economy, business or first), budget allowed by the airline companies, price of
food, seasonality of food, cost of labor to make a food item, the time required to serve the food,
number of flight attendants available to serve food, the time needed to consume food, the ability of
meal to be consumed in a small place on a plane, the time and effort needed to clear an item, the
needs and desires of the passengers, odors that may penetrate the cabin, etc.
Airlines must have a safety management system and a security program that has been certified by
the state's civil aviation authority. It is also the airline's responsibility to have an audit program in
place to ensure that the catering service provider adheres to the highest safety and security standards
when preparing, storing, transporting, and loading catering goods into airplanes.

30 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

The Cabin Crew

To ensure food safety, effective interaction between the catering service provider and cabin crew is
required. The cabin crew should be aware of the type of inflight meal service and check it against the
catering requirements. The cabin crew also needed to check the correct stowage position, with the
catering container sealed and protected from any hazards such as dust, temperature, and bugs. Or,
any security threats, as well.
At the designated time during the flight, the cabin crew then carry out the service of meals, snacks
and other items
The Passenger
At present, passengers' request for airline meals is not only to fill their stomachs but to enjoy featured
and high-quality cuisines such as those offered by high–end restaurants on land. Various major airline
companies have dedicated themselves to improving catering quality to attract more passengers. So,
airline companies need to maintain the quality of food, it also increases profit as well as reputation
and customer loyalty.

31 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

2.2) MATHEMATICAL MODEL: -

Creating a mathematical model for in-flight catering food purchase and transportation from the
supplier to the cabin involves several variables and constraints. Below is a simplified mathematical
model that can be used as a starting point.
Let us define some variables:
1. Food Items (i): These are the different types of food items or products that need to be
transported, such as meals, beverages, snacks, etc.
2. Suppliers (j): These are the various suppliers or vendors from whom the food items can be
purchased.
3. Flights (k): These are the different flights that require in-flight catering services.
4. Quantity (Q_ijk): This variable represents the quantity of each food item (i) purchased from
supplier (j) for flight (k).
5. Transportation Cost (C_jk): The cost of transporting food items from supplier (j) to flight (k).
6. Demand (D_ik): The demand for each food item (i) on flight (k).
7. Inventory (I_ijk): The inventory level of each food item (i) available from supplier (j) for flight
(k).
8. Storage Capacity (S_j): The storage capacity at each supplier (j).

Objective function: The objective is to minimize the total cost, which includes the purchase cost
and transportation cost:
=∑I, ∑j, ∑k(Qijk*Purchase cost(ij+C(jk))
Subject to the following constraints:
Demand Constraint: Ensure that the total quantity of each food item on each flight meets the
demand:
=∑j Q(ijk)=D(ik)
Inventory Constraint: Ensure that the inventory level at each supplier does not exceed its capacity:
=∑k Q(ijk) < I(ij)
Capacity Constraint: Ensure that the total quantity of each food item purchased from each supplier
does not exceed the supplier's capacity:
=∑k Q(ijk) < S(j)

32 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

Non-negativity Constraint: Ensure that the quantities purchased are non-negative:

Q(ijk) > 0
Integer Constraint: If the quantities must be integers (e.g., whole units of food items), add an integer
constraint to Qijk.
This model provides a basic framework for managing in-flight catering food purchase and
transportation. In practice, you would need to define the actual values for parameters like purchase
costs, transportation costs, demand, inventory levels, and storage capacities.

33 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

2.3) INTEGRATION OF AI IN FLIGHT CATERING SYSTEM:-

The integration of AI in airlines catering services involves leveraging advanced technologies to
enhance efficiency and improve overall customer satisfaction. AI can optimize various aspects of
catering, such as menu planning, inventory management, and supply chain logistics.
Machine learning algorithms can analyse passenger preferences, dietary restrictions, and historical
data to personalize meal options. Additionally, AI-driven predictive analytics can optimize inventory
levels, reducing waste and costs.
Real-time data processing allows for adaptive adjustments to changing demand, ensuring that airlines
provide a streamlined and personalized catering service, ultimately improving the passenger
experience.

34 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

CHAPTER – 3
CREW SCHEDULING

35 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

3 CREW SCHEDULING
Crew scheduling in airlines is a critical and complex procedure that involves the efficient organization
of flight and cabin crews to satisfy operational needs while complying to regulatory norms. The
fundamental goal is to have qualified and rested people available for each flight, encouraging safety,
timeliness, and optimal resource utilization.
Crew scheduling encompasses tasks such as assigning pilots, co-pilots, and flight attendants to specific
flights, considering factors like legal duty limits, rest requirements, and individual qualifications. This
process involves intricate coordination, considering flight durations, layovers, and time zones to
prevent fatigue and maintain crew well-being.
Advanced scheduling software is frequently used to optimize crew assignments by taking into account
variables such as personnel availability, training needs, and contractual obligations. Unexpected
disruptions, such as weather or technological challenges, need real-time scheduling alterations.
Effective crew scheduling improves operational dependability, regulatory compliance, and overall
customer happiness. To provide safe and efficient air travel services, airlines must strike a balance
between meeting operational needs, ensuring crew wellbeing, and adhering to severe aviation rules.
3.1) CREW SCHEDULING PROCESS IN INDIGO AIRLINES

1) Scheduling Software: -
IndiGo Airlines manages and optimizes crew
assignments using innovative staff scheduling
software. This innovative technology considers key
elements such as staff availability, qualifications, and
legal responsibility restrictions.
As a result, the entire scheduling process is
streamlined, ensuring that flight and cabin crews are
assigned with precision and efficiency. This advanced
software not only improves operational efficiency but also helps to ensure compliance with
aviation laws. IndiGo may create a well-balanced and optimized crew schedule by strategically
integrating these aspects, adding to the airline's overall reliability and safety requirements.
The List of software’s available is provided later in the report in the tabular form with there
services the software can illustrate.

36 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

2) Regulatory Compliance: -
IndiGo Airlines prioritizes regulatory compliance
when developing crew schedules. The airline
rigorously adheres to aviation laws, ensuring
that crew assignments comply to prescribed
maximum flight hours, duty periods, and rest
requirements.
This unrelenting dedication is essential for
putting the safety and well-being of the airline's
flight and cabin workers first. IndiGo not only
adheres to industry guidelines but also reinforces a culture of safety by meticulously aligning
schedules with regulatory standards, emphasizing the critical role adherence to regulatory norms
plays in safeguarding the health, alertness, and overall welfare of its crew members during
operational duties.

3) Flight Planning Integration: -

IndiGo Airlines seamlessly integrates crew
schedules with flight planning, fostering a
synergistic relationship between operational
aspects. This tight integration ensures a
harmonious alignment of flight schedules
with crew assignments, a pivotal strategy
aimed at optimizing resource utilization and
minimizing disruptions.
By intricately linking these two critical
elements, IndiGo achieves a finely tuned operational balance. This approach enhances efficiency
by synchronizing crew availability with specific flight requirements, reducing turnaround times,
and minimizing delays.
The airline's commitment to this integrated approach not only bolsters operational reliability but
also underscores its dedication to providing a streamlined and punctual travel experience for
passengers while optimizing the productivity and effectiveness of its flight and cabin crews.

37 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

4) Dynamic Adjustments: -
In the dynamic aviation landscape, unforeseen events, such as sudden weather disruptions or
technical issues, can significantly impact flight schedules. For instance, if a storm unexpectedly
affects a flight's departure, crew schedules at IndiGo swiftly engage dynamic adjustment
mechanisms.
The scheduling software may automatically reassign available and qualified crew members to
affected flights, ensuring continuity and minimal disruption. This agile response to real-time
challenges exemplifies how IndiGo's crew scheduling methods proactively adapt to changing
circumstances, guaranteeing that replacements are efficiently deployed when needed.
This flexibility is vital for maintaining operational resilience and delivering a reliable and
responsive service to passengers amidst the unpredictable nature of the aviation environment.
5) Crew Paring: -
IndiGo Airlines strategically implements crew pairing to optimize operational efficiency. For
example, if a pilot is based in a specific location, the scheduling system intelligently pairs them
with flights that align with their location and preferences. This minimizes travel time for the crew,
reduces operational costs, and enhances overall efficiency. The sample table is provided later in
the report.
Consider a scenario where a flight attendant fluent in multiple languages is paired with
international routes requiring language skills. Such thoughtful crew pairings not only contribute
to a smoother operational flow but also align with individual crew strengths, ensuring that the
right personnel are assigned to flights, ultimately benefiting both the airline and the passengers.

38 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

Crew Flight Departure Arrival Duty

Departure Arrival Rest Period
Pair Number Time Time Hours

1 ABC123 City A City B 08:00 AM 10:00 AM 12 hours 8

2 XYZ456 City B City C 12:30 PM 03:00 PM 10 hours 6

3 DEF789 City C City A 04:45 PM 07:30 PM 8 hours 5

4 GHI012 City B City D 10:15 AM 01:45 PM 11 hours 7

5 JKL345 City D City A 03:30 PM 06:15 PM 9 hours 6

Hypothetical Table for Crew Paring (Table-3.1)

6) Communication and Collaboration: -

IndiGo's crew scheduling, seamless
communication, and cross-departmental
collaboration are shown in real-time problem-
solving.
For example, if an unexpected aircraft
maintenance issue emerges, the scheduling staff
instantly communicates with operations and
crew management. They work together to find a
solution, ensuring that a substitute crew is
quickly assigned and that affected flights experience little inconvenience.
This proactive, collaborative approach strengthens operational resilience. IndiGo guarantees that
difficulties are resolved with rapid and coordinated reactions by building a coherent atmosphere
in which information flows effortlessly between departments, ensuring the integrity of crew
schedules and overall operational efficiency. This collaborative mindset is critical in the fast-paced
and time-sensitive aviation sector.

39 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

CREW SCHEDULING SOFTWARES

Software Name Key Specifications

Sabre AirCentre Crew Automated crew scheduling, legality checks, real-

time updates

Jeppesen Crew Optimization, pairing, fatigue risk management

AIMS Crew Management Crew scheduling, tracking, fatigue management

Maestro Crew Scheduling, bidding, tracking, compliance

Laminaar ARMS Automated rostering, crew tracking, compliance

PDC Crew Crew rostering, pairing optimization, compliance

Lufthansa Systems NetLine Crew Automated crew scheduling, roster planning

Table-3.2

40 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

3.2) MATHEMATICAL MODEL FOR CREW SCHEDULING

Objective Function:
The goal of the mathematical model is to minimize the total cost or penalty associated with assigning
crew members to flights. The objective function Z is the sum of the product of the cost (Cij) and binary
assessment variable (Xij) for all crew members and flights.
Decision Variables:
(Xij) Binary variable indicating whether crew member, (i) is assigned to flight (j), t takes the value 1 if
the assignment is made and 0 otherwise.
Constraints:

• Crew Member Assignment Constraint:

∑j Xij < 1 : Ensures that each crew member is assigned to at most one flight.

• Flight Coverage Constraint:

∑j Xij <= 1 : Ensures that each flight is covered by exactly one crew member.

• Binary Nature of Variables:

∑j Xij € {0,1} The assignment variables are binary, representing either the crew member is
assigned to the flight (1) or not (0).
Additional Considerations:

• Duty hour limits for crew members.

• Minimum rest requirements between flights.
• Qualifications and certifications of crew members for specific aircraft types.
• Crew preferences, such as preferred routes, home base considerations, etc.
• Legal and contractual obligations.

Optimization solvers like CPLEX, Gurobi, or other appropriate tools to solve the integer
programming problem and obtain the optimal crew schedule that minimizes the total cost or
penalty while satisfying all constraints.

The above model shows the basic model for reducing the cost of crew scheduling and it is drafter
for single day scheduling only.

41 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

3.3) AI ROLE IN OPTIMIZING CREW SCHEDULING

Artificial intelligence (AI) is transforming airline crew scheduling with novel applications that
improve operational precision. Machine learning predictive analysis uses past data to forecast
crew demand and anticipate future problems, allowing airlines to change schedules proactively.
Optimization algorithms, such as genetic algorithms, negotiate the complexities of crew
scheduling challenges, identifying solutions that save costs while adhering to regulatory
requirements.
AI's real-time adaptive scheduling reacts dynamically to unforeseen occurrences, weather
disruptions, or changes in flight schedules, ensuring optimal personnel utilization.
AI-powered automation speeds roster development while taking legal compliance, contractual
requirements, and individual preferences into account.
Natural Language Processing allows scheduling, operations, and crew management teams to
communicate and collaborate more effectively. Furthermore, AI predictive maintenance
capabilities foresee aircraft demands, resolving faults in advance to reduce last-minute scheduling
disruptions.
Cognitive decision support systems use AI to help schedulers make educated decisions by
analyzing large datasets for optimal crew schedules. Because AI is always learning, these systems
may adapt and improve over time, improving overall operational reliability and increasing the
passenger experience. In essence, artificial intelligence serves as a catalyst for agility, reactivity,
and efficiency in the complicated world of airline crew scheduling.

42 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

CHAPTER – 4
ROUTE PLANNING

43 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

4 ROUTE PLANNING
Airline route planning involves the strategic determination of flight paths and schedules to
optimize operational efficiency and meet market demands. Airlines analyze factors such as
passenger demand, competition, fuel costs, and regulatory considerations to design a network of
routes. This process aims to maximize profitability, minimize operational costs, and enhance
connectivity. Advanced algorithms and data analysis are employed to identify optimal routes,
considering variables like aircraft range, airport capacity, and seasonal trends. Successful route
planning is crucial for airlines to offer a comprehensive and competitive network, ensuring
sustainable growth and meeting the evolving needs of passengers and the industry.

4.1) ROUTE PLANNING PROCESS IN INDIGO AIRLINES

1) Domestic Route Plan:
IndiGo, a major Indian airline, strategically
plans its domestic routes to serve the diverse
needs of passengers within India. Utilizing its
fleet of fuel-efficient narrow-body aircraft,
such as Airbus A320s, IndiGo connects major
cities like Delhi, Mumbai, and Bangalore, as
well as regional hubs.
The route planning involves analyzing
passenger demand, evaluating competition, and ensuring compliance with domestic aviation
regulations.
For instance, recognizing the high demand for business travel between Delhi and Mumbai, IndiGo
may schedule frequent flights on this domestic route, offering convenient options for passengers.
The optimization of frequencies, aircraft utilization, and adherence to regulatory standards
contributes to IndiGo's efficiency in serving the domestic market.

2) International Route Plan: -

IndiGo primarily focused on domestic routes
within India. However, if IndiGo were to
expand into international routes, the
planning process would involve a strategic
analysis. IndiGo would assess international
markets, considering factors like passenger
demand, competition, regulatory
requirements, and potential partnerships.

44 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

The airline would select suitable aircraft with the range for international flights, such as the Airbus
A320neo or A321neo. By identifying key destinations, analyzing market dynamics, and ensuring
regulatory compliance, IndiGo could develop an efficient and competitive plan for international
route expansion, aligning with its business objectives.

3) Short Haul Routes: -

IndiGo strategically operates short-haul routes, connecting cities within India and neighboring
countries efficiently. Leveraging its fleet of fuel-efficient Airbus A320 family aircraft, the airline
caters to the high-demand, short-distance travel prevalent in the region. For instance, routes like
Delhi to Jaipur or Mumbai to Dubai exemplify IndiGo's focus on short-haul connectivity. The airline
optimizes aircraft utilization and schedules to meet the dynamic needs of short-haul travelers,
offering frequent and convenient flights while maintaining cost-effectiveness. This approach
aligns with IndiGo's commitment to providing affordable and accessible air travel for passengers
on short-distance routes.
4) Medium-Haul Routes:
IndiGo strategically operates medium-haul routes, linking cities within the Indian subcontinent
and neighboring regions. Utilizing its versatile fleet, including Airbus A320 and A321 aircraft, the
airline serves routes like Delhi to Bangkok or Mumbai to Singapore.
These medium-haul connections cater to regional travel demands, offering a balance between
the efficiency of short-haul flights and the extended reach of long-haul services. IndiGo optimizes
its schedules and aircraft deployment to ensure cost-effectiveness while providing passengers
with convenient and direct connections across moderate distances, enhancing overall
connectivity and accessibility within the broader geographic region.
Example: -
Quantitative data for IndiGo's medium-haul routes. Please note that these figures are entirely
fictional and for illustrative purposes only:
Delhi to Bangkok
Mumbai to Singapore
Flight Frequencies:
Delhi to Bangkok: 14 flights per week (2 flights per day)
Mumbai to Singapore: 10 flights per week (approximately 1.5 flights per day)
Passenger Load Factors: Average Ticket Price:
Delhi to Bangkok: 85% Delhi to Bangkok: $250
Mumbai to Singapore: 78% Mumbai to Singapore: $300

45 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

Monthly Passenger Numbers (assuming 4 weeks in a month):

Delhi to Bangkok: 14 flights/week * 85% load factor * 4 weeks = 4760 passengers/month
Mumbai to Singapore: 10 flights/week * 78% load factor * 4 weeks = 3120 passengers/month
Monthly Revenue (approximate):
Delhi to Bangkok: $250 * 4760 = $1,190,000
Mumbai to Singapore: $300 * 3120 = $936,000

5) Hub to Hub: -
Hub-to-hub routes are critical for airlines, serving as key connectors between major aviation hubs.
IndiGo, if adopting this strategy, might link hubs like Delhi and Mumbai, providing seamless
connections for passengers transferring between various domestic and international flights. By
strategically scheduling and optimizing these routes, IndiGo ensures efficient connectivity,
enabling passengers to easily switch flights without long layovers. This enhances the airline's
overall network efficiency, attracts more connecting passengers, and solidifies its hub airports as
central transit points, contributing to the success of its hub-and-spoke operational model.

6) Point to Point
Point-to-point routes involve direct flights
between two specific cities, eliminating the
need for passengers to transfer at an
intermediate hub. For instance, IndiGo might
operate a point-to-point route connecting Delhi
and Bangalore directly. This strategy simplifies
travel for passengers, reducing layovers and
travel time. IndiGo optimizes its schedules and
fleet deployment for such routes, catering to
the demand for direct connections and offering a convenient and time-efficient travel experience.
This approach aligns with IndiGo's commitment to providing direct and accessible air travel
options for passengers traveling between specific city pairs.

46 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

7) Connecting Routes: -
Connecting routes involve flights with one or more intermediate stops, enabling passengers to
connect to other flights to reach their final destination. IndiGo strategically plans connecting
routes, enhancing its network's reach.
For example, a connecting route from Kolkata to Bangkok via Delhi allows passengers to
seamlessly transfer between flights, expanding travel options. IndiGo optimizes schedules and
layover times to provide efficient connections, meeting the needs of travelers with various
destinations. This approach facilitates broader connectivity, enabling passengers to reach
destinations beyond direct routes and contributing to IndiGo's comprehensive and customer-
centric network strategy.

8) Seasonal Route: -
Seasonal routes are strategically planned by airlines, like IndiGo, to cater to variations in travel
demand during specific seasons. For instance, the airline might introduce routes to popular winter
or summer destinations during peak travel times.
This adaptive approach allows IndiGo to align its flight schedules with the ebb and flow of seasonal
demand, maximizing operational efficiency and meeting the preferences of holidaymakers. By
providing dedicated services to sought-after destinations during specific times of the year, IndiGo
ensures flexibility in its route network, capturing seasonal demand and offering diverse travel
options to its passengers.

9) Cargo Routes: -
Routes are specialized aircraft lines designed to convey products and goods between cities or
countries. Airlines, particularly specialized cargo carriers, arrange these routes deliberately in
order to meet the demand for efficient and timely cargo transportation. These routes priorities
cargo movement, frequently utilizing cargo-specific aircraft.
Cargo routes play an important role in global trade by assuring the timely and dependable transit
of goods across a wide range of businesses, including manufacturing, e-commerce, and logistics.
Cargo route optimization improves the overall efficiency of the supply chain and international
trade networks.

47 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

4.2) MATHEMATICAL MODEL FOR ROUTE PLANNING

The objective of this model is to maximize the total profit generated by the airline's routes,
considering the profit contribution from each route. The decision variables Xij determine the
selection of routes, subject to various constraints that ensure the feasibility and practicality of the
solution.
The constraints cover factors such as aircraft capacity, passenger demand, aircraft availability, and
hub operations. The binary nature of the decision variables and the linear structure of the
objective function make this a linear programming model, suitable for optimization algorithms.
The solution to this model provides an optimal combination of routes and aircraft assignments
that maximizes the overall profit for the airline.
Maximize Z = ∑i∑j profit(ij) * X(ij)

• Profit (Profit ij) This term represents the profit per passenger for the route from airport i to
airport j. It is multiplied by the decision variable x(ij), which is a binary variable indicating
whether a flight operates from airport i to airport j (1 if true, 0 if false). The objective is to
maximize the total profit across all routes.

• Decision Variables (Xij)- these binary variables determine the selection of routes. If Xij=1, it
indicates that a flight operates from airport i to airport j, contributing to the overall profit. The
objective is to find the combination of routes that maximizes the total profit.

• Double Summation (∑i∑j) The objective function considers the profit contribution for each
possible route combination across all airports.

1) Capacity Constraint:
∑j Passengers(ij) *X(ij) < Aircraft Capacity(k) *y(k)
This constraint ensures that the total number of passengers on flights from airport i does not
exceed the capacity of aircraft type k. It accounts for the fact that each aircraft type has a limited
passenger capacity.
2) Demand Constraint:
∑I Passengers(ij) *X(ij) < Demand(ij)
This constraint ensures that the total demand for the route from airport i to airport j is met. It
considers the demand for each route.

48 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

3) Aircraft Availability Constraint:

∑k yk < max aircraft(k)

This constraint limits the total number of aircraft of each type. It ensures that the airline does not
exceed the maximum number of available aircraft for any given type.

4) Hub Constraint:

∑j Xij - ∑I Xji = {1, -1, 0 }

1 if I is a hub (outflow)
-1 if I is the hub (inflow)
0 otherwise

This constraint ensures that the flow at hub airports is balanced. It ensures that for non-hub
airports, the difference between inflow and outflow is zero, and for hub airports, it enforces a
balance depending on whether the airport is an inflow or outflow hub.

5) Binary Variable Constraint:

X(ij) € {0.1]
This constraint ensures that x(ij), is a binary variable, indicating whether a particular route is
selected (1) or not (0).

6) Non-Negative Constraint:

Y(k) > 0
his constraint ensures that the number of aircraft assigned to each type is non-negative.

49 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

4.3) AI IN ROUTE PLANNING

AI transforms aviation route planning using optimization algorithms, machine learning, and real-
time data analysis. Airlines can use predictive analytics to forecast demand, optimize hub-and-
spoke networks, and anticipate problems.
Machine learning models forecast gasoline use, allowing for more fuel-efficient route planning. AI
improves air traffic control by recommending effective aircraft paths and reducing congestion.
Crew scheduling is optimized to achieve a balance of operational efficiency and regulatory
compliance. Predictive maintenance ensures proactive monitoring of aircraft health, reducing
unwanted downtime.
AI-driven route planning saves money by considering fuel, staff, and maintenance costs. Finally,
artificial intelligence enables airlines to dynamically change plans based on real-time situations,
improving overall efficiency, lowering costs, and improving customer experience while adhering
to safety and regulatory standards.

SOFTWARES AVAILABLE FOR FLIGHT PLANNING

Navigation
Software Name Features Integration Database User Interface
Updates

Route Planning,
Crew Planning, Regular User-Friendly,
ForeFlight EFB Integration
Weather Briefings, Updates Interactive
Charts

Electronic Flight
Jeppesen Avionics Regular Intuitive,
Bag (EFB), NavData
FliteDeck Pro Systems Updates Graphical
Updates

Online Flight
Interactive, Web-
SkyVector Planning, Weather Web-Based Real-time
Driven
Visualization

Flight / Crew
Garmin Regular Modern, User-
Garmin Pilot Planning, Weather,
Avionics Updates Friendly
Navigation

PFPX Route
FSX/P3D, X- User- Detailed,
(Professional Optimization, Fuel
Plane Initiated Professional
Flight Planner X) Planning
Table- 4.1

50 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

CHAPTER 5
MAINTENANCE
PLANNING

51 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

5 MAINTENANCE PLANNING: -
To maintain the fleet's longevity, dependability, and safety, airlines must proactively schedule and
oversee maintenance tasks. This is known as maintenance planning.
Based on operating usage, manufacturer recommendations, and regulatory requirements, airlines
create detailed plans that specify routine inspections, maintenance, and component
replacements. This planning considers calendar time, cycles, and flight hours. AI is frequently used
to support advanced predictive maintenance, which evaluates real-time data from aircraft sensors
to foresee possible problems and schedule proactive maintenance, reducing unplanned
downtime.
Optimizing aircraft availability, improving safety, and maximizing operational effectiveness are the
objectives. To comply with aviation rules and ensure that aircraft fulfil strict safety standards while
minimizing delays to flight schedules, effective maintenance planning is essential.
5.1) MAINTENANCE PLAN: -
1) DGCA Regulations: -
IndiGo, like other airlines, prioritizes regulatory compliance
by strictly adhering to directives established by aviation
authorities such as the Directorate General of Civil Aviation
(DGCA) in India. This commitment ensures that
maintenance planning aligns with prescribed safety
standards and operational guidelines.
The airline's initiation of maintenance activities is grounded
in a comprehensive understanding and adherence to these
regulations, emphasizing the paramount importance of
safety and regulatory conformity in all aspects of its fleet
management and maintenance procedures.

2) Manufacturer Guidelines:
IndiGo meticulously adheres to maintenance
guidelines established by aircraft manufacturers,
particularly Airbus, for its A320 fleet. This
involves a structured approach to scheduled
inspections, checks, and component
replacements, aligning with Airbus'
recommendations.
The maintenance schedule is intricately crafted,
considering critical factors such as flight hours, cycles, and calendar time. By adhering to
manufacturer guidelines, IndiGo ensures that its aircraft undergo routine assessments and

52 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

necessary component replacements at prescribed intervals, maintaining the fleet's airworthiness

and operational reliability.
This strategic alignment with manufacturer recommendations underscores IndiGo's commitment
to upholding the performance and safety standards set by the aircraft manufacturer, contributing
to the overall longevity and dependability of its A320 fleet.

3) Routine Inspection: -

Routine inspections form a cornerstone of

IndiGo's maintenance strategy,
encompassing daily, weekly, and monthly
checks.
These assessments aim to proactively
identify and rectify minor issues, preventing
their escalation.
The inspections cover a spectrum of
evaluations, including visual assessments,
system tests, and operational checks. By regularly scrutinizing aircraft conditions, IndiGo ensures
early detection of potential issues, maintaining operational integrity, and enhancing safety.
This proactive approach aligns with industry best practices, contributing to the airline's
commitment to passenger safety, regulatory compliance, and the overall reliability of its fleet.

4) Predictive Maintenance: -
Predictive Maintenance is a cornerstone of
IndiGo's approach, leveraging advanced
data analytics and sensor information from
aircraft systems. This proactive strategy
enables the airline to predict potential
component failures before they occur,
facilitating timely maintenance
interventions and minimizing unscheduled
downtime.

By real-time data, IndiGo optimizes the scheduling of downtime, strategically planning for
maintenance activities during scheduled intervals.

53 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

This approach not only enhances aircraft reliability and operational efficiency but also contributes
to substantial cost savings by preventing unplanned disruptions and ensuring that maintenance
activities align with the aircraft's operational needs and regulatory requirements.

5) Component Replacement: -
IndiGo systematically manages the replacement
or overhaul of critical aircraft components,
including engines, landing gear, and avionics. This
process adheres to predetermined schedules or
specific usage thresholds. The airline follows
meticulous guidelines, considering factors such as
flight hours, cycles, and manufacturer
recommendations.
By replacing or overhauling key components at
strategic intervals or when they reach specified operational limits, IndiGo ensures the sustained
reliability, safety, and airworthiness of its fleet.
This approach aligns with industry best practices, contributing to the longevity and optimal
performance of aircraft systems critical to operational integrity.
(Demand Forecasting in Components)
IndiGo employs demand forecasting to anticipate the need for component replacements in its
aircraft. Leveraging historical data, usage patterns, and manufacturer recommendations, the
airline predicts when key components, such as engines, landing gear, and avionics, will reach
specified operational limits.
This proactive approach ensures that replacement parts are sourced, scheduled, and available
when needed, minimizing downtime, and optimizing maintenance efficiency. By accurately
forecasting demand for component replacements, IndiGo enhances operational reliability,
reduces disruptions, and maintains its commitment to passenger safety and regulatory
compliance.

54 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

6) Documents: -
IndiGo prioritizes meticulous documentation and record-keeping of all maintenance activities,
encompassing inspections, repairs, and component replacements. These comprehensive records
serve as a crucial repository for the airline, facilitating regulatory compliance by providing a
transparent account of adherence to safety standards.

Equally important is the ability to track the aircraft's entire maintenance history, enabling
informed decision-making for future maintenance planning. The detailed documentation ensures
that the airline meets stringent aviation regulations, enhances safety protocols, and fosters a
systematic approach to maintenance operations, contributing to the overall airworthiness and
reliability of IndiGo's fleet.

7) Continuous monitoring: -
Continuous monitoring is integral to
IndiGo's operational strategy,
employing real-time systems that
oversee the health of aircraft
components during flights.
These monitoring systems proactively
detect anomalies or irregularities,
triggering immediate alerts.
This swift response initiates further
inspection or maintenance actions to
address potential issues before they escalate. By leveraging advanced technology for continuous
monitoring, IndiGo ensures the ongoing health and performance of its fleet, enhancing safety,
preventing unforeseen failures, and contributing to the overall reliability and operational
efficiency of its aircraft.
This proactive approach aligns with industry best practices, emphasizing the airline's commitment
to passenger safety and operational excellence.

55 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

5.2) MATHEMATICAL MODEL

The model involves considering various factors such as the number of aircraft in the fleet,
maintenance schedules, operational requirements, and resource constraints. Here is a simplified
mathematical model that you can use as a starting point. Keep in mind that real-world airline
maintenance planning is a complex task, and this model is a basic representation:
Let us define some variables:
N: Total number of aircraft in the fleet.
ti: Time (in hours, days, or other time units) since the last maintenance for aircraft i.
Mi: Maximum allowable time between maintenance for aircraft i.
Ci: Cost of maintenance for aircraft i if it exceeds the maximum allowable time.
R: Total available maintenance resources (e.g., man-hours, facilities, etc.).

The objective is to minimize the total cost of maintenance while ensuring that all aircraft are
adequately maintained within their allowable time limits.

Let's denote the decision variables:

xi: Binary variable indicating whether maintenance is performed on aircraft i(xi=1 if maintenance
is performed=0 otherwise).
Now, the mathematical model can be formulated as an optimization problem:
Objective Function:
Minimize Z = ∑ Ci⋅xi
Subject to:
1) Maintenance Resource Constraint:
∑ xi ≤ R

2) Maintenance Schedule Constraint:

Ti + MaintenanceDuration⋅xi ≤ Mi, ∀i

3) Binary Decision Variable:

Xi ∈ {0,1},

56 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

The objective function aims to minimize the total cost of maintenance.

The first constraint ensures that the total number of aircraft undergoing maintenance does not
exceed the available maintenance resources.
The second constraint enforces the maintenance schedule for each aircraft, considering the time
since the last maintenance and the maintenance duration.
The third constraint defines the binary nature of the decision variables.

57 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

5.3) INTEGRATION OF AI IN MAINTENANCE PLANNING

Artificial Intelligence (AI) can play a significant role in enhancing maintenance planning for airlines
by improving efficiency, reducing costs, and increasing overall safety. Here are several ways in
which AI can be applied to maintenance planning in the airline industry:
Predictive Maintenance:
Fault Prediction: AI algorithms can analyze historical data, sensor readings, and other relevant
information to predict potential faults or failures in aircraft components. By anticipating issues
before they occur, airlines can schedule maintenance proactively, minimizing downtime and
reducing the chances of in-flight failures.
Condition Monitoring:
Real-time Monitoring: AI systems can continuously monitor the condition of aircraft components
in real-time. This involves analyzing data from various sensors to assess the health of critical
systems. Any deviations from normal operating parameters can trigger alerts for maintenance
crews.
Optimized Maintenance Scheduling:
Data-Driven Decision Making: AI can process large datasets to identify patterns and trends in
aircraft performance. This information can be used to optimize maintenance schedules, ensuring
that maintenance tasks are performed when needed and minimizing unnecessary checks.
Regulatory Compliance:
Automated Reporting: AI systems can automate the generation of reports required for regulatory
compliance. This ensures that maintenance activities adhere to aviation safety standards and
guidelines.
Cognitive Decision Support:
Human-Machine Collaboration: AI can provide maintenance planners with decision support tools,
helping them analyze complex data and make informed decisions. This collaboration between AI
and human expertise can lead to better-informed and more effective maintenance planning.
Implementing AI in maintenance planning requires robust data infrastructure, integration with
existing systems, and continuous monitoring and improvement. As technology advances, AI is
likely to play an increasingly crucial role in optimizing maintenance operations within the airline
industry.

58 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

CHAPTER – 6
MAJOUR FINDINGS

59 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

6 MAJOUR FINDINGS FROM THE PROJECT

After the qualitative and quantitate data, we researched for the supply chain planning of the
airlines we predicted following points.

6.1) COMPARISION WITH AND WITHOUT AI INTEGRATION: -

Parameters Without AI Integration With AI Integration

Evaluate AI Relies on historical data and Utilizes AI algorithms for real-time

Integration manual decision-making. analysis and adaptive learning.

Identify Efficiency Limited efficiency gains due to Proactive decision-making, real-

Gains manual processes. time optimization, and automation.

Improved cost-effectiveness
Assess Cost- Higher costs, suboptimal
through optimization and reduced
effectiveness decisions, and manual errors.
errors.

Decision support based on Advanced decision support with

Enhanced Decision
historical trends and human real-time data and machine
Support
experience. learning.

AI analyzes data to identify and

Relies on human foresight, may
Risk Management mitigate risks, suggesting
not effectively identify risks.
contingency plans.

Improved planning, on-time

Customer Inconsistencies may lead to
deliveries, and personalized
Satisfaction delays and dissatisfaction.
services enhance satisfaction.

Table 6.1

60 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

6.2) AIRLINES SECTOR GROWTH

Growing Demand of AI in Airlines Sector
The market for artificial intelligence in aviation was valued at US$ 653.74 million globally in 2021,
and it is projected to grow at a compound annual growth rate (CAGR) of 35.38% from 2022 to
2030, when it is predicted to reach US$ 9,985.86 million.

Year Market in Dollars (Million)

2021 653.74
2022 885
2023 1198
2024 1622
2025 2195
2026 2672
2027 4024
2028 5448
2029 7376
2030 9985.86
Table – 6.2

61 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

Airlines Revenue Growth

As we can see in the below graph forecasted by the ICCT department, the growth is significant
after the year 2020, the implementation and availability of AI resources for supply planning makes
the revenue optimized.

62 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

6.3) CHALLENGES IN AI IMPLEMENTATION

A) Data Quality and Integration:

AI systems require high-quality, accurate, and timely data to make informed decisions. The
airline industry deals with vast amounts of diverse data from various sources, and integrating
this data can be a complex task.
B) Lack of Skilled Workforce:
Implementing AI requires skilled professionals who understand both the technology and the
intricacies of the airline industry. The shortage of AI talent can hinder successful
implementation.
C) Regulatory Compliance:
The aviation industry is highly regulated, and implementing AI systems may raise concerns
related to compliance with safety and security regulations.
D) Change Management:
Introducing AI may lead to resistance from employees who fear job displacement or changes
in their roles. Change management is crucial for a smooth transition.
E) Cost Considerations:
Implementing AI involves significant upfront costs for technology, training, and infrastructure
upgrades. ROI may take time to materialize.
F) Cybersecurity Risks:
With increased connectivity and reliance on digital systems, the aviation industry becomes
susceptible to cybersecurity threats. AI systems may be vulnerable to attacks.
G) Dynamic Operational Environment:
The airline industry is subject to rapid changes, including fluctuations in fuel prices,
geopolitical events, and unexpected disruptions. AI systems must be adaptable to such
dynamic conditions.

63 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

CHAPTER 7
BIBLOGRAPHY

64 | P a g e
Brief Analysis of AI And Operations in Supply Chain Management
for Airlines Industry

BIBLIOGRAPHY

Book:
Author: Christopher, M.
Title: Logistics & Supply Chain Management
Year: 2016
Publisher: Pearson Education
ISBN: 9781292083797

Journal Article:
Author: Van Mieghem, J. A.
Title: The Impact of Big Data on Operations and Supply Chain Management.
Journal: Production and Operations Management
Year: 2015
Volume: 24(4)
Pages: 541-552
DOI: 10.1111/poms.12243

Industry Report:
Author: International Air Transport Association (IATA)
Title: Air Cargo's Transformation with Twin Peaks: Resetting the Global Air Cargo Industry
Year: 2021
URL:https://www.iata.org/en/iata-repository/publications/economics/twin-peaks-cargo-
transformation/

Online Article:
Author: Smith, J.
Title: How AI Is Transforming the Airline Industry
Source: Harvard Business Review
Year: 2018
URL: https://hbr.org/2018/09/how-ai-is-transforming-the-airline-industry

Indigo Airlines
URL:https://www.goindigo.in/?cid=Display|Affiliate|LF|LA6|1243500||5a6aa87c4754fa575
9696ff55df19dd5

Airbus
URL- https://www.airbus.com/en

65 | P a g e