Journal of Engineering, Management

and Information Technology

Vol. 03, No. 04 (2025) 285-294, DOI: DOI:10.61552/JEMIT.2025.04.006 - http://jemit.aspur.rs

ENHANCING PROFITABILITY AND PRODUCT MIX

OPTIMIZATION THROUGH LINEAR PROGRAMMING
PROBLEM: A CASE STUDY ON THE INDIAN GARMENT
INDUSTRY
Siddharth Ghuge Received 02.04.2024.
Tanishaa Gupta Revised 27.05.2024.
Harsh Sharma Accepted 25.06.2024.
Purvi Shetty
Debasmita Mukherjee 1

Keywords: ABSTRACT
The present research work investigates the application of Linear Programming
garment industry, linear programming Problem (LPP) to generate profits while efficiently managing resources such
problem, resource utilization, optimal as raw materials, machinery, labor, transportation, and processing time. The
solution. research focuses on Sunshel Textiles India Pvt Ltd., a garment industry situated
in Gujarat, India. This company was chosen because of its diversified product
range, which requires important management decisions in determining the
quantity of products to be produced and sold. The study sampled seven
products: suits, blazers, trousers, shirts, half t-shirts, shorts, full-sleeve t-shirts.
The data used for the analysis is taken from company records. A LPP model
has been formulated to improve the company’s operations and the optimal
results are being calculated using solver add-in in Microsoft Excel 2019.
Analysis of the collected data indicates that Sunshel Textiles Industry has the
potential to gain extra 5.5% profit by incorporating the optimized model in their
managerial decisions.
© 2025 Journal of Engineering, Management and Information Technology

1. INTRODUCTION various sectors of the economy. This pressure led to the

introduction of management theories aimed at dissecting
Garment industries are constantly suffering from a the business landscape and addressing the external
shortage of manufacturing inputs, which leads to lower challenges presented by the industry's operating
capacity utilization and thus lower production. However, environment, as well as the practical problems arising
economic growth depends on management choices that from operations within the industry. To address these
are made at the enterprise level to increase output, either challenges, a quantitative approach to modeling and
by reducing costs or by maximizing output, thus analyzing decision-making problems has been
increasing productivity within the real sector. As a result, developed. One such quantitative approach is the use of
business managers are constantly looking for optimal LPP models, using mathematical techniques to ascertain
decisions that align with their goals, which primarily the most advantageous course of action. It was developed
focus on maximizing profits. by Soviet mathematician A.N.Kolmogorov before WWII
The expansion of industry puts pressure on management (Woubante, 2017).
to determine the most effective levels of planning, The objective entails optimizing the linear objective
organizing, leading and controlling production across the function taking into account the constraints of linear
equality and inequality. This process involves

1
Corresponding author: Debasmita Mukherjee
Email: [email protected] 285
 Enhancing Profitability and Product Mix Optimization Through Linear Programming Problem: A Case Study on the
Indian Garment Industry
formulating the problem as a set of linear equations and leads to the exploration of different approaches, such as
inequalities, followed by determining optimal values for energy-efficient buffer allocation and binary level
the decision variables that simultaneously satisfy all optimization (Magnanini et al. 2022). Woubante
imposed constraints. Linear optimization is widely used (Woubante, 2017) used data from a textile manufacturing
in many fields, including operations research, economics, unit in Ethiopia to estimate the linear programming
engineering, and finance, to name a few. It can be applied model parameters. Using LINGO 16.0 software, he
to a variety of problems such as resource allocation, showed that improving customer fulfilment can increase
production planning, transportation, and network company profits by 59.84%. Likewise, Tesfaye et al.
optimization (Aliu et al. 2023; Syifa at al. 2023; (Tesfaye et al. 2016) examined data from the Ethiopian
Shamsuzzaman et al.2023). garment industry and used a linear programming model
India's domestic textile and apparel market has seen a to highlight the potential for resource utilization to
significant decline, from US$106 billion in 2019–2020 to increase by 46.41% compared to current levels. Their
an estimated US$75 billion in 2020–21. However, it is model revealed a potential increase in the company's
projected to recover at a compound annual growth rate profits by 145.5%.
(CAGR) of 10% in 2019-20 and reach USD 190 billion In this paper a garment industry, Sunshel Textiles, based
by 2025-26. From this market, clothes make up 73% of in Gujarat, India is being considered. The data has been
the total. The apparel sector is an important part of the collected from the company records. A LPP model
Indian economy, with a significant proportion (93%) of addressing the effective production planning in order to
the workforce employed in unorganized small and get maximum profit has been proposed and analysed in
medium enterprises (SMEs). this article.
Apparel companies are responding to this challenge by
cutting costs and optimizing their supply chains, making
apparel more affordable than other consumer goods. This 2. MATERIALS AND METHODS
strategy has led to an increase in consumer purchases,
with the number of garments purchased per consumer The primary data used in this research paper is from
increasing by 60% between 2000 and 2014. This trend Sunshel Textiles, a large sized enterprise. The data was
has been particularly evident in emerging economies collected through industrial visits to Sunshel Textiles
such as India, driven by a growing middle class and rising which is located in Umbergaon, Gujarat.
living standards (Majumdar & Sinha 2018). The visit involved direct observation of the
Product mix, the range of products offered by a company, manufacturing process, interviews with the operations
is critical to maintaining customer preferences and manager and various staff. This data was then analyzed
revenue. However, offering a wide assortment brings using LPP techniques in Microsoft Excel to optimize the
problems in terms of cost and storage complexity, as well production process and minimize costs while meeting
as customer confusion. Fashion trends change rapidly, production targets.
which complicates inventory management, and retailers
need to efficiently select the right product mix to reduce 2.1 Introduction to our dataset
costs (Koschmann & Sheth 2018). The Table 1 shows details of various products produced
In 1940, Georg Dantzig presented the simple method for by a company and the attributes including fabric used,
solving linear programming problems. This echnique thread required, labor cost, overhead, average demand,
was later applied to food-related problems by Stigler in inventory, profit margin, production capacity, monthly
1945 (Render et al. 2023, Lin et al. 2020). Another study capacity. The products include suits, blazers, trousers,
focused on improving poultry nutrition, aiming to reduce shirts, half t-shirts, shorts and full-sleeve t-shirts.
costs while meeting nutritional requirements (Mallick et
al. 2020). The complexity of multi-objective problems
Table 1. Apparel Production Metrics
Average
Fabric Thread Labor Overh-- Average Profit Production Monthly current
Product Used Used Cost ead demand Margin Cap Cap Production
Suit 5 1400 1900 1300 12000 30% 600 14400 13500
Blazer 1.85 550 900 720 12000 30% 600 14400 13000
Trousers 1.3 270 140 120 70000 30% 3200 76800 72000
Shirts 1.55 380 105 90 35000 50% 1800 43200 38000
Half t-shirt 1.15 140 75 65 53000 55% 2600 62400 58000
Shorts 0.8 110 60 55 14000 60% 700 16800 15500
Full-sleeve t-
shirt 1.35 200 85 70 27000 45% 1300 31200 28000

286
 Journal of Engineering, Management and Information Technology
Vol. 03, No. 04 (2025) 285-294, doi: 10.61552/JEMIT.2025.04.006

The necessary information required for our analysis was SP/unit: Selling price of one unit per product.
provided by the factory. However, certain information Profit/unit: Profit earned on one unit of product.
which was not given was calculated using the available Example: The formula (Selling Price ×
data for which we have shown the necessary calculations. Profit Percentage) is used to calculate the profit per unit
Fabric: It is the amount of fabric used in square meters to of a product based on its selling price and profit
manufacture one unit of the product. percentage.
Example: To make one suit we require 5 square meters We calculate the profit per unit for suit as:
of fabric. The selling price of a suit given is 6000, and the profit
Thread: It is the amount of thread used in meters to percentage is 30%.
manufacture one unit of the product. Using the formula, the calculated profit per unit suit is,
Example: To make one suit we require 1400 meters of 30
6000 × = 1800 .
thread. 100
This means that for each suit sold, the factory earns a
Labor cost: The cost of labor required to manufacture one
profit of 1800. Similarly, the selling price and profit per
unit of product.
unit for each product are listed in the Table 2.
Example: The cost of labor for making one suit is
The following Table 3, represents the amount of
Rs.1900.
inventory to be kept for the resources and how much of
Overhead: The cost of indirect expenses incurred in the
that inventory will be consumed after production.
production of one unit of the product.
Table 3. Resource Allocation
Example: The overhead cost incurred while making one
suit is Rs.1300. Measurement Consumption
Resource Type Unit Inventory Value
Average demand: It is the average demand for a product
in a month. Fabrics Meter 378997.5 360950
Example: The average monthly demand of suits is 12000 Threads Meter 79122750 75355000
units.
Profit margin: The percentage of profit earned on the Labor Rupees 62034000 59080000
product after deducting all the costs. Overheads Rupees 47830125 45552500
Example: The profit percentage earned on one unit of suit
is 30%. Consumption value: It is the amount of raw material
Average Current production: It is the average amount of being consumed in a month.
each product being produced in a month. Here we calculate Consumption value by multiplying
Example: On an average 13500 suits are being produced amount of fabric used per meter square used by average
in a month. current production.
Production Capacity: The number of units that can be Example:
produced per day. Consumption Value
Example: The production capacity for suits is 600 units = ∑(Amount of Resources per product
per day. × Average Current Production)
Monthly Capacity: The number of units that can be Consumption value for fabric
produced in a month. = ∑(Fabric Used per Product ×
Example: As the factory operates for 6 days a week and Current Production for the Product) =
a month comprises 4 weeks, the total number of working (5×13500)+(1.85×13000)+(1.3×72000)+
days in a month is 24. Hence, the monthly production can +(1.55×38000)+(1.15×58000)+(0.8×15500)+
be calculated by multiplying the production per day with +(1.35×28000)
24. = 67500 + 24050 + 93600 + 58900 + +66700 +
For suits the production per day is 600, 12400 + 3780
Hence the production per month = 600 × 4 × 6 = = 360950
14400 Inventory: Inventory typically refers to the extra
The following table, Table 2 represents the selling price resources being held in reserve in addition to the
for the products and profit per unit. expected consumption value.
Table 2. Selling Price and Profit Per Unit We are given that the factory wants to maintain a 5%
Product SP/unit Profit/unit buffer inventory, which means that they intend to keep
Suit 6000 1800
an additional 5% of the consumption value as inventory.
Maintaining 5% extra inventory helps to mitigate the
Blazer 3000 900 risks associated with uncertain demand and supply chain
Trousers 550 165 disruptions.
To calculate the inventory, we multiply the consumption
Shirts 400 200
value by 0.05 (5%) and add it to the consumption value
Half t-shirt 350 192.5 itself. So, we get:
Shorts 300 180 Example:
Inventory = Consumption Value ∗ 0.05
Full-sleeve t-shirt 350 157.5
+ Consumption Value

287
 Enhancing Profitability and Product Mix Optimization Through Linear Programming Problem: A Case Study on the
Indian Garment Industry
Inventory for Fabric = 3. Additivity: The contribution to the objective
consumption value of fabric ∗ 0.05 + function from each decision variable is additive.
+ consumption value of fabric 4. Divisibility: Solutions can have fractional
= 360950 ∗ 0.05 + 360950 values rather than being restricted to whole
= 18047.5 + 360950 numbers.
= 378997.5 5. Non-negativity: All variables and answers are
constrained to be nonnegative, as negative
2.2 Model formulation values are not feasible in the context of the
Few important terminologies in LPP: problem.
Objective Function: Objective function is a linear
function in terms of decision variables. The aim of the
study is to optimize it i.e., either maximization or 3. MATHEMATICAL MODEL
minimization.
Ex : 𝑀𝑎𝑥 𝑍 𝑜𝑟 𝑀𝑖𝑛(𝑍) = 𝑐1 𝑥1 + 𝑐2 𝑥2 + 𝑐3 𝑥3 + To set up the model, the decision variables were set as:
⋯ + 𝑐𝑛 𝑥𝑛 𝑥1 = Number of Suits
Decision Variables: These are used to represent 𝑥2 = Number of Blazer
products, services, projects, or other interrelated 𝑥3 = Number of Trousers
activities that need to be optimized together to make the 𝑥4 = Number of Shirts
best use of resources. These variables are denoted by 𝑥1, 𝑥5 = Number of Half T-shirts
𝑥2, 𝑥3, …, xn and their values need to be determined to 𝑥6 = Number of Shorts
optimize the objective function. 𝑥7 = Number of Full-Sleeve T Shirts
Constraints: The limitation on resources like production Then, the LPP model for maximizing the total profit can
capacity, raw-material, labor, man-hours, machines. be expressed as:
These resources are expressed as linear Z = Total profit during the month in rupees
inequalities in terms of decision variables known as In the LPP model,
constraints. Maximizing the total profit:
Example: 𝑎11 𝑥1 + 𝑎12 𝑥2 + 𝑎13 𝑥3 + ⋯ + 𝑎1𝑛 𝑥𝑛 ≤ 𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑖𝑛𝑔 𝑍 = 1800 𝑥1 + 900𝑥2 + 165𝑥3
= ≥ 𝑏1 + 200𝑥4 + 192.5 𝑥5 + 180𝑥6
Non-Negative Constraints: In a real situation the + 157.5𝑥7
negative values are not possible, thus all decision Now we will form the constraints,
variables must be non-negative, known as non-negative C1: Fabric Constraint
constraints. This constraint ensures that the total fabric used in
Example: 𝑥1 ≥ 0, 𝑥2 ≥ 0, 𝑥3 ≥ 0, …, 𝑥𝑛 ≥ 0 production cannot exceed the available fabric inventory.
Mathematical Representation:
2.3 Steps for linear programming 5 𝑥1 + 1.85𝑥2 + 1.3𝑥3 + 1.55𝑥4 + 1.15 𝑥5 + 0.8𝑥6
Here are the steps to formulate a mathematical LPP: + 1.35𝑥7 ≤ 378997.5
Step 1: Define the decision variables. C2: Thread Constraint
Step 2: Define the objective function. This ensures that the total thread used in the production
Step 3: Formulate the constraints. of all products does not exceed the thread inventory
Generally, LPP can be written as: available.
𝑀𝑎𝑥 𝑍 𝑜𝑟 𝑀𝑖𝑛(𝑍) 1400 𝑥1 + 550𝑥2 + 270𝑥3 + 380𝑥4 + 140 𝑥5
= 𝑐1 𝑥1 + 𝑐2 𝑥2 + 𝑐3 𝑥3 + ⋯ + 110𝑥6 + 200𝑥7 ≤ 79122750
+ 𝑐𝑛 𝑥𝑛 C3: Labor Cost Constraint
Subject to the conditions, The labor cost constraint limits the total labor cost to a
𝑎11 𝑥1 +𝑎12 𝑥2 +𝑎13 𝑥3 + ⋯ + 𝑎1𝑛 𝑥𝑛 ≤ = ≥ b1 certain amount based on the available budget.
𝑎21 𝑥1 + 𝑎22 𝑥2 +𝑎23 𝑥3 + ⋯ +𝑎2𝑛 𝑥𝑛 ≤ = ≥ b2 1900 𝑥1 + 900𝑥2 + 140𝑥3 + 105𝑥4 + 75 𝑥5
𝑎31 𝑥1 + 𝑎32 𝑥2 +𝑎33 𝑥3 + ⋯ + 𝑎3𝑛 𝑥𝑛 ≤ = ≥ b3 + 60𝑥6 + 85𝑥7 ≤ 62034000
.. C4: Overhead Constraint
𝑎𝑚1 𝑥1 + 𝑎𝑚2 𝑥2 + 𝑎𝑚3 𝑥3 + ⋯ + 𝑎𝑚𝑛 𝑥𝑛 ≤ = ≥ bm This constraint is set to ensure that the total cost of
𝑥1 ≥ 0, 𝑥2 ≥ 0, 𝑥3 ≥ 0, …, 𝑥𝑛 ≥ 0 production does not exceed a certain limit based on the
available budget
2.4 Assumptions 1300 𝑥1 + 720𝑥2 + 120𝑥3 + 90𝑥4 + 65 𝑥5 + 55𝑥6
The LPP model has several assumptions that must be met + 70 𝑥7 ≤ 47830125
to ensure its accuracy and effectiveness. These C5 to C11: Monthly Production Capacity for Each
assumptions are as follows: (Feng et al. 2022) Product
1. Certainty: The objective and constraint values This constraint ensures that the total number of products
are certain and do not change during the period. produced in a month does not exceed the maximum
2. Proportionality: The value of the decision capacity of the production unit.
variable is proportional to the value of the 𝑥1 ≤ 14400 (suits)
objective function. 𝑥2 ≤ 14400 (blazer)

288
 Journal of Engineering, Management and Information Technology
Vol. 03, No. 04 (2025) 285-294, doi: 10.61552/JEMIT.2025.04.006

𝑥3 ≤ 76800 (trousers) 4. RESULTS

𝑥4 ≤ 43200 (shirts)
𝑥5 ≤ 62400 (half t-shirt) 4.1 Model solution
𝑥6 ≤ 16800 (shorts) We have used a LPP technique Simplex Method. Among
𝑥7 ≤ 31200 (full-sleeve t-shirt) the various software packages, we used Solver add-in
C12 to C18: Average demand and Monthly Capacity from Microsoft Excel 2019 to perform the Simplex
Constraint Method.
We need to include a constraint that ensures the company The optimal solution for an LPP model indicates the
produces at least as much as the average demand or more. optimal number of units of each product that should be
However, in cases where the average demand exceeds the produced in order to maximize profit, subject to the given
production capacity, we cannot produce more than the constraints. The optimal solution is given in the Table 4
capacity. below.
Hence, we take the minimum value between the two as Table 4. Optimal Solution
the production constraint. Units of each product to
Mathematically, this can be represented as: Units of each product to be produced
𝑃𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛 Products be produced (Rounded-up values)
≥ 𝑀𝐼𝑁(𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝐷𝑒𝑚𝑎𝑛𝑑, 𝑃𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛 𝐶𝑎𝑝𝑎𝑐𝑖𝑡𝑦) x1 14095.966 14096
𝑥1 ≥ 𝑀𝐼𝑁(12000 ,14400)
(Suits, In this case we take 12000) x2 14189.77106 14190
𝑥1 ≥ 12000 x3 70000 70000
𝑥2 ≥ 𝑀𝐼𝑁(12000, 14400)
(Blazer, in this case we take 12000) x4 42471.49282 42472
𝑥2 ≥ 12000 x5 62400 62400
similarly,
x6 16800 16800
𝑥3 ≥ 70000 (trousers)
𝑥4 ≥ 35000 (shirts) x7 29804.28124 29805
𝑥5 ≥ 53000 (half t-shirt)
𝑥6 ≥ 14000 (shorts) The solution shows that the company should on an
𝑥7 ≥ 27000 (full-sleeve t-shirt) average produce 14,096 suits, 14,190 blazers, 70,000
To summarize, the LPP Model is given as: trousers, 42,472 shirts, 62,400 half T-shirts, 16,800
𝑀𝑎𝑥𝑖𝑚𝑖𝑧𝑖𝑛𝑔 𝑍 = 1800 𝑥1 + 900𝑥2 + 165𝑥3 shorts, and 29,805 full-sleeve T-shirts to maximize their
+ 200𝑥4 + 192.5 𝑥5 + 180𝑥6 profit. Additionally, the solution provides the maximum
+ 157.5𝑥7 profit that can be earned, which is Rs.7,79,18,005.61.
Subject to,
 5 𝑥1 + 1.85𝑥2 + 1.3𝑥3 + 1.55𝑥4 + Table 5. Profit After Optimization
1.15 𝑥5 + 0.8𝑥6 + 1.35𝑥7 ≤ 378997.5, Profit After Optimization
 1400 𝑥1 + 550𝑥2 + 270𝑥3 + 380𝑥4 +
140 𝑥5 + 110𝑥6 + 200𝑥7 ≤ 79122750, Maximize Z 77918005.61
 1900 𝑥1 + 900𝑥2 + 140𝑥3 + 105𝑥4 +
Table 5 shows us the profit Sunshel Textiles can generate
75 𝑥5 + 60𝑥6 + 85𝑥7 ≤ 62034000, if they were to follow the model presented by this paper.
 1300 𝑥1 + 720𝑥2 + 120𝑥3 + 90𝑥4 +
65 𝑥5 + 55𝑥6 + 70 𝑥7 ≤ 47830125, 4.2 Result analysis
𝑥1 ≤ 14400, We have calculated the current profit earned by Sunshel
𝑥2 ≤ 14400, Textiles on an average in a month by multiplying the
𝑥3 ≤ 76800, profit per unit with the current production of each
𝑥4 ≤ 43200, product.
𝑥5 ≤ 62400, ∑(Average current production × profit per unit)
𝑥6 ≤ 16800, = (13500 ∗ 1800) + (13000 ∗ 900)
𝑥7 ≤ 31200, + (72000 ∗ 165) + (38000 ∗ 200)
𝑥1 ≥ 12000, + (58000 ∗ 192.5) + (15500 ∗ 180)
𝑥2 ≥ 12000, + (28000 ∗ 157.5)
𝑥3 ≥ 70000, = 7,38,45,000`
𝑥4 ≥ 35000, The profit from the current production is calculated to be
𝑥5 ≥ 53000, Rs.7,38,45,000.
𝑥6 ≥ 14000, Using the optimized model, the maximum profit that the
𝑥7 ≥ 27000, Sunshel Textiles can earn is Rs.7,79,18,005.61. This
𝑥1 ,𝑥2 ,𝑥3 ,𝑥4 , 𝑥5 ,𝑥6 ,𝑥7 ≥ 0 indicates that there is a difference of Rs.40,73,005.609
between the current profit and the maximum profit that
can be earned using the optimized model

289
 Enhancing Profitability and Product Mix Optimization Through Linear Programming Problem: A Case Study on the
Indian Garment Industry
On calculating the percentage increase in profit, it is 06/05/18 64 2018 May-23
indicated that the Sunshel Textiles has the potential to
improve their profit margin by 5.5% by implementing the 13/05/18 61 2018 May-23
optimized production plan. 20/05/18 68 2018 May-23
In conclusion, the optimized model provides a production
plan that enables the Sunshel Textiles to increase their 27/05/18 67 2018 May-23
profit margin significantly. Table 6. Google Trends Suits Data (head)

The last 5 rows of the table:

5. EXTENDING THE MODEL Date Trend Year Month

In order to optimize the production and maximize profits 05/03/23 69 2023 Mar-23
for Sunshel Textiles over the entire year, we have 12/03/23 67 2023 Mar-23
extended our LPP model from a monthly basis to an
annual basis. In order to analyze the data extracted from 19/03/23 70 2023 Mar-23
Google Trends and make the predictions for future 26/03/23 74 2023 Mar-23
trends, we have used the following Machine Learning
technique. 02/04/23 83 2023 Apr-23
Table 7. Google Trends Suits Data (tail)
5.1 GOOGLE trends
Google Trends is a tool provided by Google that allows 5.3 Machine learning
users to explore the popularity of search queries over a Now we use Machine Learning techniques by
given period of time and in a specific geographic incorporating Google trend data to train a Random Forest
location. It provides insights into what people are Regressor model. This model then predicts the average
searching for and how their interests are changing over trend for the time period May 2023 to April 2024. The
time. implementation was done using Python programming
It's useful for businesses and researchers to understand language.
demand for products or services in different regions. It Random Forest Regressor is an ensemble learning
provides real-time data and has a large sample size. algorithm used for regression tasks. It builds multiple
While it is true that Google Trends does not directly decision trees and combines their predictions to predict
measure actual purchases or sales, it does provide continuous numerical values. The randomness in
valuable insights into consumer behavior and interests. building the trees reduces overfitting and improves
Google Trends has gained popularity as a prominent accuracy. It is used for predicting prices, sales, and other
source for big data research and applications, owing to continuous numeric quantities. (Jain and Kumar 2020).
several key factors. Its user-friendly interface makes it
easy to use, providing valuable insights into consumer Month Average demand
interests and behavior for various applications and
May’23 78.53
studies (Jun et al. 2018).
This can be indicative of potential demand, and can help
June’23 68.59
companies like Sunshel Textiles to make informed
decisions about their production and marketing July’23 66.28
strategies.
August’23 68.16
5.2 Calculation for suits
Table 6 and Table 7 are the trends for suits which help us September’23 66.38
determine the relative change in interest among the
general population with regarding the item, Suits. These October’23 82.98
datapoints are based on the visits on e-commerce
websites via Google for the products in question. November’23 81.9
Following are a few snippets of the front 8 rows and the
last 5 rows of the table obtained from Google Trends December’23 74.35
which has been referred for further analysis,
The first 8 rows of the table: January’24 72.36
Date Trend Year Month
February’24 84.01
08/04/18 73 2018 April -23
March’24 70.19
15/04/18 71 2018 April-23
22/04/18 72 2018 April-23 April’24 80.95
29/04/18 65 2018 April-23 Table 8. Average trends for Suits in May 2023 to April 2024

290
 Journal of Engineering, Management and Information Technology
Vol. 03, No. 04 (2025) 285-294, doi: 10.61552/JEMIT.2025.04.006

The table shows the average demand for suits over a 12- 5.4 Model solution
month period from May 2023 to April 2024 and the The Table 10 shows us the optimal solution as calculated
corresponding column displays the average demand for using model-2. Table 10 is based on the same idea as
suits in each of those months. Table 4, but different in the sense that it is only applicable
Similarly, we calculated the average demands for blazers, for the month of May’23.
trousers, shirts, half t-shirts, shorts and full-sleeve t-
shirts. Now we will see the month wise demand for each Units of each Units of each product
of the products for the month of May 2023, in Table 9: product to be to be produced
We use the values from the Table 8 and calculate their Products produced (Rounded-up values)
deviation from their average. And then use this number
to multiply the “Average Demand” we have from the x1 14146.55684 14147
Table 1 to the respective columns. From the above x2 13851.64517 13852
calculation we get the following table with May’23
appropriate values for “Average Demand”. x3 73556 73556
For example, say we need to predict the demand for Suits
for the month of May’23. We refer to Table 8 which will x4 39495.71756 39496
give us average trend for suits for the all months. From
that we use the value for the month of June’23, which x5 62400 62400
comes out to be 68.59. We have used the value for the
month of June and not May, as the goods that are to be x6 16800 16800
ready by June should be manufactured in the month of x7 30072.6 30073
May. Similar pattern has been used for all further
calculations for the following months. Now in order to Table 10. Optimal Solution for May’23
find the deviation of this trend value from the annual
average, we divide 68.59 with 74.04 which is the The Table 10 shows us the optimal solution for the month
calculate average for all the months. And the value we of May’23 as predicted by the model 2. It shows the exact
get is 0.926. This states that the demand for the month of number of units that have to be produced to make the
May’23 will be 0.926 times the average demand or a maximum profit possible according to the current
7.74% decrease. Hence to calculate the final value, we demand for the products.
multiply average demand for suits, which we get from
Table 1 as 12000 with 0.926. The final value we get will Max Z
be 11115.6. We use the exact same logic to predict the 77738600.98
demand for all products for the month of May’23, as seen
in Table 9 below. Profit At
Current
Current Production

Production 73845000
Average Demand

Production Cap

Profit Percent
Profit Margin

Monthly Cap
Fabric Used

Labor Cost

Overheads
Product

Thread

Difference In
Profit After
Optimization 38,93,601
Table 11. Comparing Current and Optimal Profits
Suit 5 1400 1900 1300 11115.6 30% 600 14400 30 13500
The Table 11 shows us the difference in Profit before and
Blazer 1.85 550 900 720 8197.2 30% 600 14400 30 13000 after using the model presented by this paper.The profit
Trousers 1.3 270 140 120 73556 30% 3200 76800 30 72000 from the current production is calculated to be
Rs.7,38,45,000.
Shirts 1.55 380 105 90 35941.5 50% 1800 43200 50 38000
Using the optimized model for the month of May’23, the
Half t- maximum profit that the Sunshel Textiles can earn is
1.15 140 75 65 59031.4 55% 2600 62400 55 58000
shirt
Rs.7,77,38,600.98. This indicates that there is a
Shorts 0.8 110 60 55 19027.4 60% 700 16800 60 15500 difference of Rs.38,93,601 between the current profit and
Full- the maximum profit that can be earned.
sleeve t- 1.35 200 85 70 30072.6 45% 1300 31200 45 28000 Calculating the percentage increase in profit, we get an
shirt increase of 5.27% from the current profit. This indicates
Table 9. Projected Apparel Production Metrics for that the Sunshel Textiles has the potential to improve
May’23 their profit margin in the month of May’23 by 5.27% by
Note: Monthly cap and current production are same as in implementing the optimized production plan.
Table 1. In conclusion, the optimized model enables Sunshel
Similar in section 3, we develop a LPP model from Table Textiles to increase their profit margin.
9, Table 2 and Table 3. The only difference between them
being the “Average Demand” column.

291
 Enhancing Profitability and Product Mix Optimization Through Linear Programming Problem: A Case Study on the
Indian Garment Industry

Similarly, the demand for the rest of the months will be:
Product Jun’23 Jul’23 Aug’23 Sep’23 Oct’23 Nov’23 Dec’23 Jan’24 Feb’24 Mar’24 Apr’24
Suit 10741.2 11046 10758 13448.4 13273.2 12049.2 11726.4 13615.2 11374.8 13119.6 12727.2
Blazer 6927.6 7111.2 7269.6 9648 17996.4 20308.8 16894.8 17440.8 9781.2 10246.8 10393.2
Trousers 72569 72051 72464 74613 73542 71771 68467 73710 70056 77448 75936
Shirts 34601 36491 35693 39434.5 32984 31216.5 30061.5 37250.5 37247 41030.5 39669
Half t-shirt 57075.7 59269.9 59672.7 52496.5 44657.8 43258.6 41832.9 51335.8 62333.3 63865 59715.1
Shorts 18804.8 18508 18006.8 17767.4 17684.8 17785.6 13651.4 14180.6 16595.6 19027.4 19027.4
Full-sleeve
t-shirt 29076.3 30194.1 30399.3 26743.5 22750.2 22037.4 21311.1 26152.2 31754.7 32535 30420.9
Table 12. Projected Demand for the Rest of the Months
The Table 12 shows us the projected demand for each product the sell across the next 11 months.
The decision variables for the rest of the months will be:
Jun’23 Jul’23 Aug’23 Sep’23 Oct’23 Nov’23 Dc’23 Jan’24 Feb’24 Mar’24 Apr’24
x1 14146 14122 14126 14196 13800 13884 14042 13792 14069 13120 13708
x2 13911 14004 13971 13664 14400 14400 14400 14400 14258 12240 14400
x3 72569 72051 72464 74613 73542 71771 68467 73710 70056 76800 75936
x4 41123 40551 40055 40627 41563 42105 43114 41512 40734 41031 39669
x5 62400 62400 62400 62400 62400 62400 62400 62400 62400 62400 60093
x6 16800 16800 16800 16800 16800 16800 16800 16800 16800 16800 16800
x7 29077 30195 30400 27834 28248 29019 30457 28175 31755 31200 30421
Table 13. Optimal Solution for the Rest of the Months
Similarly, we calculate the maximized profit for each Through this table we can see that by using our optimized
month using our mathematical model: mathematical model over the entire year, this company
Current could considerably increase its profit by
Months Maximize z profit Difference Rs.4,48,00,758.09.
May’23 77738600.98 73845000 3893600.984
Jun’23 77796312.26 73845000 3951312.262 6. CONCLUSION

Jul’23 77812511.6 73845000 3967511.6 This study is centered on Sunshel Textiles Ltd. The
profits comparison between the actual production and
Aug’23 77790180.74 73845000 3945180.745 suggested production using LPP models show
Sep’23 77705187.18 73845000 3860187.179 considerable differences. It can be concluded that the
apparel company should use LPP to determine their
Oct’23 77730729.86 73845000 3885729.861 optimal product mix.
Thus, it will be possible to obtain the following results:
Nov’23 77820283.53 73845000 3975283.529
Dec’23 77987355.99 73845000 4142355.99 6.1 Model
Using the optimized model, the maximum profit that the
Jan’24 77722234.65 73845000 3877234.65 Sunshel Textiles can earn is Rs.7,79,18,005.61. This
indicates that there is a difference of Rs.40,73,005.609
Feb’24 77898467.99 73845000 4053467.99 between the current profit and the maximum profit that
Mar’24 75458516.49 73845000 1613516.486 can be earned using the optimized model.
On Calculating the percentage increase in profit, it is
Apr’24 77480376.82 73845000 3635376.817 indicated that the Sunshel Textiles has the potential to
improve their profit margin by 5.5% by implementing the
Total 930940758.1 886140000 44800758.09 optimized production plan.
Table 14. Profits Comparison for the Rest of the
Months

292
 Journal of Engineering, Management and Information Technology
Vol. 03, No. 04 (2025) 285-294, doi: 10.61552/JEMIT.2025.04.006

Conflict of interests: There is no conflict of interests.

6.2 Extended model Author contributions: All the authors contributed
By using our optimized mathematical model over the equally to the writing of this research paper and have
entire year this company could significantly increase its accepted responsibility for the entire content of this
profit by Rs.4,48,00,758.09. manuscript and approved its submission.
There is a 5.05% increase in the profit by implementing Acknowledgement: All the authors gratefully
the optimized production plan. acknowledge Sunshel Textiles India Pvt Ltd.,
Umbergaon, Vapi 396171, Gujrat, India for providing the
data to conduct this research work.

References:

Aliu, T. O., Gbolagade, M. G., & Adeshola, A. D. (2023). Application of Linear Programming on Cost Minimization of
Pig Feeds. Earthline Journal of Mathematical Sciences, 13(1), 267-289.
Feng, Z., Wang, R., Huang, X., Jiang, X., & Liu, Y. (2022, December). Analyzing Textile Industry by Linear
Programming. In 2022 2nd International Conference on Economic Development and Business Culture (ICEDBC
2022) (pp. 1459-1463). Atlantis Press.
Jain, S., & Kumar, V. (2020). Garment categorization using data mining techniques. Symmetry, 12(6), 984.
Jun, S. P., Yoo, H. S., & Choi, S. (2018). Ten years of research change using Google Trends: From the perspective of big
data utilizations and applications. Technological forecasting and social change, 130, 69-87.
Koschmann, A., & Sheth, J. (2018). Brand line extensions: creating new loyalties or internal variety-seeking? Journal of
Product & Brand Management, 27(4), 351-362.
Lin, C. W. R., Chen, M. T., Tseng, M. L., Jantarakolica, T., & Xu, H. (2020). Multi-objective production programming
to systematic sorting and remanufacturing in second-hand clothing recycling industry. Symmetry, 12(7), 1161.
Magnanini, M. C., Terkaj, W., & Tolio, T. A. (2022). A novel algorithm for optimal buffer allocation in automated
asynchronous unreliable lines. The International Journal of Advanced Manufacturing Technology, 122(11-12),
4405-4419.
Majumdar, A., & Sinha, S. (2018). Modeling the barriers of green supply chain management in small and medium
enterprises: A case of Indian clothing industry. Management of Environmental Quality: An International Journal,
29(6), 1110-1122.
Mallick, P., Muduli, K., Biswal, J. N., & Pumwa, J. (2020). Broiler poultry feed cost optimization using linear
programming technique. Journal of Operations and Strategic Planning, 3(1), 31-57.
Render, B., Stair Jr, R. M., Hanna, M. E., & Hale, T. S. (2023). Quantitative Analysis for Management, Global Edition--
(Perpetual Access). Pearson Higher Ed.
Shamsuzzaman, M., Mashud, M., Rahman, M. M., Rahman, M. M., Hoq, E., & Das, D. (2023). Management and
Maintenance of Textile Machinery. In Advanced Technology in Textiles: Fibre to Apparel (pp. 31-63). Singapore:
Springer Nature Singapore.
Syifa, E. A., Istiqomah, T. N., Puspita, N. P., Ratnasari, L., Khabibah, S., Anggoro, P. W., & Bawono, B. (2023). The
Application of Linear Programming for the Optimal Profit of Pt. Naruna Using the Simplex Method. Management
Systems in Production Engineering, 31(2), 138-143.
Tesfaye, G., Berhane, T., Zenebe, B., & Asmelash, S. (2016). A linear programming method to enhance resource
utilization case of Ethiopian apparel sector. International Journal for Quality Research, 10(2), 421–432.
Woubante, G. W. (2017). The optimization problem of product mix and linear programming applications: Case study in
the apparel industry. Open Science Journal, 2(2), 1-11.

Siddharth Ghuge Tanishaa Gupta Harsh Sharma

Nilkamal School of Mathematics, Nilkamal School of Mathematics, Nilkamal School of Mathematics,
Applied Statistics & Analytics, Applied Statistics & Analytics, Applied Statistics & Analytics,
SVKM’s NMIMS Deemed to be SVKM’s NMIMS Deemed to be SVKM’s NMIMS Deemed to be
University, Mumbai 400056, University, Mumbai 400056, University, Mumbai 400056,
Maharashtra, India Maharashtra, India Maharashtra, India
[email protected] [email protected] [email protected]

293
 Enhancing Profitability and Product Mix Optimization Through Linear Programming Problem: A Case Study on the
Indian Garment Industry

Purvi Shetty Debasmita Mukherjee

Nilkamal School of Mathematics, Nilkamal School of Mathematics,
Applied Statistics & Analytics, Applied Statistics & Analytics,
SVKM’s NMIMS Deemed to be SVKM’s NMIMS Deemed to be
University, Mumbai 400056, University, Mumbai 400056,
Maharashtra, India Maharashtra, India
[email protected] [email protected]
ORCID 0000-0003-1735-0421

294