Ambo University

Institute of Technology

Hachalu Hundessa Campus

School of Mechanical and Industrial Engineering
Industrial Engineering Department
Bachelor of Science in Industrial Engineering
Research Proposal
Optimization of bottling Manufacturing Process by using
Experimental Design Approach
(A CASE OF AMBO MINERAL WATER SC)

By

Alazer Sintayehu UGR/35787/12

Gebreselassie Dessie A/UR27969/11
Meron Wodajo UGR/36079/12

Submitted to: Mr Birhane D

Date: 16/03/2024Ambo, Ethiopia
i
 Approval Sheet

Ambo University
Institute of Technology, Hachalu Hundessa Campus
School of Mechanical and Industrial Engineering
Industrial Engineering Department
Optimization of bottling Manufacturing Process by using
Experimental Design Approach
(A CASE OF AMBO MINERAL WATER SC)

by

Alazer Sintayehu Abebaw UGR/35787/12

Gebreselassie Dessie A/UR27969/11
Meron Wodajo UGR/36079/12

Department Head (Name) Signature Date

______________________ _____________ _________

Advisor (Name) Signature Date

______________________ _____________ _________

Examiner (Name) Signature Date

______________________ _____________ _________

Examiner (Name) Signature Date

______________________ _____________ _________

ii
 Declaration
We hereby declare that the work which is being presented in this proposal entitled
“Optimization of bottling Manufacturing Process by using Experimental Design Approach
(A CASE OF AMBO MINERAL WATER SC), is original work of our own, has not been
presented by any other institute or university and that all sources of material used for the
proposal have been duly acknowledged.
Alazer Sintayehu Abebaw Signature Date

_____________________ ____________ ________________

Gebreselassie Dessie Signature Date

_____________________ ____________ ________________

Meron Wodajo Signature Date

_____________________ ____________ ________________

This is to certify that the above declaration made by the students is correct to the best of my
knowledge.

Mr Birhane D Signature Date

_____________________ ____________ ________________

iii
 Acknowledgements
We would like to take a moment to express our gratitude to those who were supported and
encouraged us along this project. First of all, we would like to thank almighty GOD for helping
us in every step of our thesis and giving tendency to do and giving chance to arrive this level. We
have great thanks and express our deepest respect for our families, for their encouragement,
financial and moral support in our life to reach this level. Secondly, we would like to forward
and express our sincere appreciation and thank to our advisor Mr. Birhane D. for his continuous
and whole hearted encouragement, support and giving constructive criticism for the completion
of our thesis as planned and needed. In addition to this, we give great thanks for all group
members, for our agreement, encouragement and confidence in this work. The last but not the
least we would like to thank all of our teachers of Industrial Engineering Department.

iv
 Contents
Approval Sheet ii

Declaration iii

Acknowledgements iv

List of tables vii

INTRODUCTION 1

1.1 Background of the study 1

1.2 Problem statement 5
1.3 Research question 6
1.4. Objectives of the Study 6
1.4.1 General objectives 6
1.4.2 Specific objectives 7
1.5. Significance of the Study 7
1.6. Scope of the Study 7
1.7. Limitation of the Study 8
1.8 Organization of the Study 8
Chapter-2 9

RELATED LITERATURE 9

2.1. Introduction 9
2.2. Process 9
2.3 Process Measurement 9
2.4 Factors Affecting Process 10
2.5 Optimization of process 11
2.6. Bottling process 11
2.7 Design of Experiment 12
2.8 Taguchi Method 14
2.9 Literature review Gap 15
Chapter-3 16

RESEARCH METHODOLOGY 16

3.1 Research Design approach 16

v
 3.2 Research Methodology 16
3.3 data collection method 16
3.3.2 Secondary data collection 17
3.4 Data analysis method 17
3.4.1 Integration of Taguchi–ABC and Taguchi–Pareto 17
3.4.2 Taguchi–Pareto 18
3.4.3 Taguchi–ABC 19
3.7 Objective function and procedure of Taguchi method 19
Chapter-4 21

DATA COLLECTION AND ANAYSIS 21

4.1 Introduction 21
4.2 Taguchi’s orthogonal arrays, factors and signal-to-noise ratios 23
4.3 Main effect plots 28
4.4 Analysis of variance (ANOVA) 29
4.5 Taguchi-Pareto Analysis (TPA) 32
4.6 Taguchi ABC 34
References 38

References 43

vi
 List of tables
Table 1: Collected data of NOS, PM and PWH 21
Table 2: the mean, maximum and minimum values of production worked hour, planned maintenance
hour and number of shift. 22
Table 3: Taguchi’s orthogonal arrays, factors and signal-to-noise ratios for mean 23
Table 4: response Table for Signal to Noise Ratios of mean 24
Table 5: Taguchi’s orthogonal arrays, factors and signal-to-noise ratios for maximum 25
Table 6: response table for signal to noise ratios of maximum 25
Table 7: Taguchi’s orthogonal arrays, factors and signal-to-noise ratios for minimum 26
Table 8: response table for signal to noise ratios of minimum 27
Table 9: Taguchi’s orthogonal arrays, factors and signal-to-noise ratios for mean ,maximum and
minimum 27
Table 10: Taguchi S/N ratio response table for mean, maximum and minimum 28
Table 11: Taguchi Pareto for mean 32
Table 12: Taguchi Pareto for maximum 33
Table 13: Taguchi Pareto for minimum 33
Table 14: Taguchi–Pareto experimental design –mean, maximum and minimum (with factors and their
levels) 34
Table 15 : Taguchi–Pareto’s orthogonal arrays, factors and signal-to noise ratios 34
Table 16: Taguchi–Pareto-80–20 S/N ratio response table ratio 35
Table 17: Taguchi AB’s special mix L16 design, factors and signal-to-noise ratios 36
Table 18: Taguchi AB S/N ratio response table 37

vii
 List of figures
Figure 1: 2 An illustration of the experimental design for bottling process plan 20
Figure 2: Main effect plots for mean, maximum and minimum values 29

viii
 Chapter-1

INTRODUCTION
1.1 Background of the study
Nowadays, companies are more competitive and every detail is important if the business wants
to improve its competitiveness’. For this, it is relevant to keep the customers satisfied offering to
them what they are expecting. Moreover, the companies must upload their management
techniques to be able to compete with their rivals, get better performance to do their best for their
customers and improve every day. Depending on product characteristics and production type,
several classical architectures of manufacturing and processing systems (assembly line, flexible
manufacturing system, flexible manufacturing cell, job shop, etc.) can be deployed in a factory
(Luis M sanchaz, 2020). These architectures can be analyzed in conjunction with DOE
applications and factorial algorithms are solved and optimized based on mathematical. In a
Design of Experiments (DoE) study, the factors of interest are varied systematically from their
lowest to highest value and all possible factor combinations are executed in the same set of
experiments.

In order to develop a robust process design or production system, unanticipated events and
system stability must be effectively controlled. By means of parameter design that is offered by
Taguchi, product quality and system robustness can be improved and experimental cost is
reduced (Shang, 2020). Parameter design is one of the important phases for the improved product
quality and can be regarded as the robust design studies. The control factors can be divided as
controllable and non-controllable. In order to determine the effect of these factors, the most
effective way is experimental design suggested by Taguchi.

The optimum parametric setting is a special technique introduced to experimental analysis which
helps a researcher to be able to select the best/significant value out of the parameters that made
up the experimental runs. It is best obtained by finding the maximum value of each factor based
on levels (Ridwan Majekodunmi Adegoke, June 2022) Experimental design methods were
originally developed by (Fisher, 1970). However, classical experimental design methods are too
complex and are not easy to use. Furthermore, a large number of experiments have to be carried
out as the number of process parameters increases. The Taguchi method uses a special design of
orthogonal arrays to study the entire parameter space with only a small number of experiments
1
(Tarng, 2000). To optimize a material flow, one should focus on maximizing or minimizing
specific parameters according with their nature (for instance, cost is a minimizing parameter,
while productivity is a maximizing parameter) (Tianxia, 2012). The modern manufacturing
constraints should produce highly customizable and innovative products; continuously increase
their quality and decrease the time-to-market and costs in order to gain competitive advantages.
All these determined a focus on optimizing the manufacturing and processing systems starting
early from their design phase by using MFT (Material Flow Theory) and specific simulation
software tools (Achraf Ammar, 2019).

The optimization of bottling processes represents a critical endeavor in the beverage industry,
where efficiency, quality, and cost-effectiveness converge. Ambo Mineral Water, renowned for
its purity and premium quality, embarks on a journey to enhance its bottling operations through
the application of experimental design methods. In an era characterized by dynamic consumer
demands and evolving technological landscapes, the quest for operational excellence becomes
paramount. By leveraging experimental design techniques, Ambo Mineral Water seeks to
systematically explore and refine its bottling processes, aiming to maximize throughput,
minimize waste, and ensure product consistency. This initiative underscores Ambo's
commitment to innovation, sustainability, and customer satisfaction in a competitive market
environment. As the beverage industry continues to evolve, the optimization of bottling
processes stands as a cornerstone for Ambo Mineral Water's continued growth and market
leadership.

The bottling process plays a pivotal role in various industries, ranging from beverage
manufacturing to pharmaceutical production. Optimizing this process is essential to ensure
efficiency, minimize costs, and maintain product quality. One effective approach to enhance the
bottling process is through the application of experimental design methods. Experimental design
methods offer a systematic and structured approach to evaluate multiple factors simultaneously
while controlling for potential sources of variation. By systematically altering process parameters
and observing their effects on key performance indicators, experimental design enables the
identification of optimal conditions that maximize output and minimize waste. In this study, we
aim to optimize the bottling process using experimental design methods. By systematically
varying factors such as filling speed, bottle size, temperature, and pressure, we seek to identify

2
the optimal combination of parameters that result in improved throughput, reduced downtime,
and enhanced product quality.

The optimization of bottling processes through the application of Design of Experiments (DOE)
traces its roots back to the mid-20th century. As industries, particularly in food and beverage
production, sought more efficient and cost-effective manufacturing methods, the need for
systematic approaches to enhance bottling processes became evident. Initially, trial-and-error
methods dominated, leading to excessive resource consumption and suboptimal outcomes.
However, in the 1920s and 1930s, pioneers like Ronald A. Fisher laid the groundwork for
experimental design principles, which later formed the basis of DOE. In the following decades,
especially post-World War II, the advent of statistical methods and advancements in industrial
engineering fostered the systematic exploration of bottling processes. The 1950s and 1960s
witnessed significant advancements in the application of DOE to optimize bottling operations.
Industries increasingly recognized the value of DOE in systematically varying process
parameters, such as filling speeds, temperatures, and bottle designs, to identify the optimal
combination for efficiency and product quality. Early adopters, particularly in the beverage
industry, utilized factorial designs and response surface methodologies to understand the intricate
relationships between process variables and bottling outcomes. These endeavors not only led to
tangible improvements in throughput and product consistency but also paved the way for more
sophisticated optimization techniques in subsequent years. By the late 20th century, the
integration of computer technology revolutionized the optimization of bottling processes using
DOE. Industries leverage predictive modeling and machine learning algorithms to anticipate
bottling outcomes and identify optimal process configurations with minimal human intervention.
Moreover, the emphasis on sustainability and environmental stewardship has spurred efforts to
optimize bottling processes not only for efficiency and quality but also for resource conservation
and waste reduction. As industries embrace digital transformation and interdisciplinary
collaboration, the optimization of bottling processes using DOE stands poised for further
innovation and refinement in the years to come.

According to (Noorwali, 2013) lean thinking is “a set of principles and techniques that drive
organizations to continually add value to the product they deliver by enhancing process steps that
are necessary, relevant, and valuable while eliminating those that fail to add value." Furthermore,
in the global engineering research, several authors have adopted optimization methods and
3
integrated other tools with them for performance enhancement. Consequently, the present study
was attempted by focusing on critical variables of the two components of the bottling plant that
have been underrepresented in manufacturing literature in the past optimization of joint
maintenance and production process (Oji, 2021). The Taguchi–Pareto and Taguchi–ABC
Methods have several significant concurrent optimization and prioritization values in the area of
engineering practices, which have not been Extensively validated scientifically (Nwafor S.C.,
2019). The study has proposed a novel idea of joint optimization in maintenance and production
processes for bottling plant using Taguchi–ABC, Taguchi–Pareto and response surface
methodology.

As the bottling process plants face severe competition and threats on operational efficiency and
effectiveness, they are currently at risks of closure, downsizing and extreme profit decline
(Basso, 2019). Thus, innovative performance measures should be installed to intervene
(Abhishek, 2013) Productivity is the main factor on which a business firm's existence depended
upon and the main objective of a business firm is providing consumers’ satisfaction; while
making a profit. Business firms always attempt to maximize profit. But, to maximize profit a
business firm has to be efficient where efficiency is measured by the capacity of the business
firm to raise the productivity of existing resources, so that cost per unit production is reduced
(Derbe, 2023).

One of the critical reasons why process improvement is essential for a bottling plant is that it can
help to increase production. By identifying bottlenecks and inefficiencies in the manufacturing
process and implementing changes to address these issues, a bottling plant can improve the speed
and efficiency of its operations. Experimental helps to test scenarios and identify areas where the
process may be slow or inefficient. And experimental test can help the company identify
potential problems before they occur, and take action to address these issues, preventing costly
delays and downtime and ensuring that the production process runs smoothly and efficiently.
Another way that process DOE can help to increase production in a bottling plant is by
optimizing the manufacturing process for efficiency and cost-effectiveness. Using DOE
software, a bottling plant can experiment with different process configurations and input
variables and determine the most efficient and cost-effective way to produce its products. DOE
also leads to significant cost savings and improved quality and consistency of the final product. It
also reduces the risk of defects and other quality issues, which leads to higher customer
4
satisfaction and a more substantial reputation for the company. Using DOE is particularly
important in the highly competitive bottling industry, where customers have many options.

In optimizing the bottling process for Ambo mineral water, experimental design methods play a
pivotal role in enhancing efficiency and quality. The process involves systematically varying
factors such as bottling speed, downtime, work process, and bottle material to determine their
individual and collective impacts on the final product. Experimental design allows for the
systematic exploration of these variables while controlling for potential confounding factors,
enabling a comprehensive understanding of the process dynamics. By employing techniques
such as factorial design or response surface methodology, researchers can identify optimal
operating conditions that maximize throughput, minimize defects, and ensure product
consistency. Through iterative experimentation and statistical analysis, insights gained from the
optimization process can lead to significant improvements in bottling efficiency, product quality,
and overall cost-effectiveness. This systematic approach not only enhances the competitiveness
of Ambo mineral water in the market but also underscores a commitment to continuous
improvement and customer satisfaction. Traditional optimization methods often rely on trial and
error, leading to prolonged production cycles, increased waste, and suboptimal performance. The
emergence of experimental design methods offers a systematic approach to address these
challenges by systematically exploring the effects of multiple factors on the bottling process
while minimizing experimental runs. Experimental design allows for the identification of key
factors influencing process performance and the determination of optimal parameter settings. By
leveraging the principles of experimental design, this study aims to fill this gap by systematically
investigating the influence of key process parameters on bottling efficiency and product quality.
Through the systematic manipulation of factors such as filling speed, bottle geometry, and
environmental conditions, we seek to identify the optimal combination of parameters that
maximize throughput while minimizing defects and waste.

1.2 Problem statement

Ambo Mineral Water Share Company's total productivity is currently less than its actual
production capacity. The imbalance between planned and actual productivity is attributed to
various factors, such as number of shifts, planned maintenance and production worked hour. The
absence of these optimum parameter setting and long job completion times contribute to a low
production. As a result, the factory faces challenges in meeting the planned of production
5
outcome. Through observation analyzed the main reasons for falling short of the planned
production, which include idle time, breakdown time, poor maintenance planning, and a lack of
maintenance resulted from the absence of optimal setting of planned maintenance, production
worked hour activities. Addressing these root causes is essential to establish a robust planned
maintenance program and ensure the efficient operation of equipment and systems. Unequal
distribution of specific jobs for the workers results from number of shifts i.e. variation time of
production worked hour. Moreover, the overall production is need determine how those
parameters are managed, achieve the company product policies because actual system
performance is not satisfactory. From the company data analyzed the existing parameter level
value in the manufacturing are (PWH= 11, PM= 4, NOS= 6) is very low.

The focus of the study can apply Taguchi design of experiments to identify, quantify, and
analysis manufacturing process parameters that are evident in the manufacturing system and use
the result for maximizing manufacturing of overall production in the case company. Determine
the optimum level settings of parameters that enhance overall productivity by optimizing level
settings of parameters. Identify significant factors such as planned maintenance hours,
production hours worked, and number of shifts that influence production outcomes. The Taguchi
methods, Taguchi Pareto, and Taguchi ABC analyses provided optimal parametric settings for
parameters. The projects contribute to policy making, skill enhancement for maintenance
managers and decision making to sustain overall bottling manufacturing processes of Ambo
Mineral Water Share Company using Experimental Design Approach

1.3 Research question

1 What methodologies should be employed to identify, quantify, and analyze the

manufacturing parameters?

2 What are the main factors that affect the bottling process?

3 What solution can be proposed to alleviate the parameter identified?

1.4. Objectives of the Study

1.4.1 General objectives

6
The general objective of the study is to Optimization of bottling Manufacturing Process by using
Experimental Design Approach.

1.4.2 Specific objectives

 Identify, quantify and analysis manufacturing process parameters

 To identify the factors that hampers the bottling process of AMWSC

 Determine the optimum level settings of parameters

1.5. Significance of the Study

The study outcome benefits Ambo mineral water share Company as well as other bottling
manufacturing firms through the adoption of optimization of bottling manufacturing Taguchi
methods that helps to improve the status of productivity factors available in the factory. As a
company, Ambo mineral water Share Company gained increased daily overall productivity with
bottling productivity process through reduced factors affect their productivity. Also reduced the
gap between target and actual daily output, increased customer satisfaction, and reduced workers
fatigue, idle time with cycle time. Moreover, gives possible solution to existing problem faces in
the overall bottling manufacturing process; remedial measure can be instituted to solve the
problems which increase the productivity and profitability of the company. The thesis will have
advantages to overcome on the problem of overall company manufacturing process waste so that
the company can be productive and more competitive. It is hopefully believed that by optimizing
bottling manufacturing process of the company will have increased competitiveness between
competitors, will decrease the loss of the company in return will increase productivity by mainly
optimizing bottling manufacturing process and also by improving quality of the beverage
produced. The study findings are used as an input for academicians, research institutes, and any
interested party who dares to learn about the optimization of bottling manufacturing process
using design of experiment approach.

1.6. Scope of the Study

The scope of this research is the Optimization of bottling Manufacturing Process by using
Experimental Design Approach (Specific to the case company of Ambo Mineral Water Share
Company) the studies focus area are the overall manufacturing of analyzing and measuring the
production capacity of the company, identifying the root factors that can affect bottling process

7
of the Company, evaluating the existed manufacturing bottling process in the company and
developing improved bottling process for the company.

1.7. Limitation of the Study

The limitation of the study includes the following: -
There was a time limitation (i.e. the time allowed for this thesis was very short it was impossible
to study deeply with in this short period and also, we are preparing our self for Exit exam). There
was no appropriate data related to optimization of manufacturing bottling process of the
company (i.e. we haven’t got enough data and information in order to analyze it deeply). We
have faced some problems while conducting our study’s; some of the major problems are lack of
full information of what is causing hindering manufacturing process. In this company there is
also unwillingness of some respondent, difficult to get volunteer interviews form workers and
concerned managers and also shortage of time and transport cost that restricted us on Achieving
full desired completion of this study.

1.8 Organization of the Study

This study has been organized to comprise the following five chapters.

Chapter-1, which is the introduction part of the study, begins with the research background and
discusses the background of the study, the statement of the problem, basic research questions,
and general and specific objectives, followed by scope and limitation of the study. Finally,
significance of the study and organization of the research reports Chapter-2 contains the
literature review of the study dealing with related works of literature on theoretical review,
empirical review, conceptual framework, hypothesis and the research gap findings for this study.
Chapter-3 contains the detailed research methodology which discusses the research design and
approach of the study, data source, and method of data collection. Chapter-4 contains data
analysis and interpretation, Chapter-5 contains results and discussions of the study in this section
the main findings of the research are summarized and conclusions on major findings are
presented. Recommendations are given based on the research findings. Finally, the study
forwarded some suggestions for further investigations

8
 Chapter-2

RELATED LITERATURE
2.1. Introduction
A literature review encompasses the theoretical review guide of the study which includes all the
relevant information upon which this study is based (Kalpande, 2003). This section literature
review on the basics of process, process optimization, design of experiment and Taguchi method
by different authors, and modeling approaches for process optimization which have been done so
far will be presented in detail. It is also part of this section that describes factors that affecting
process. Finally, the literature reviews on this area have been undertaken a literature summary,
and gaps identified from the literature were presented.

2.2. Process
Recently, optimization of production has been directly related to the search for models that
ensure the sustain ability of production. It is mainly perceived as the balance of the social,
economic and environmental pillars of development activities (Daneshjo, 2023) one of the goals
of which is to improve the quality of life. Consumers using cosmetic products have become more
aware of environmental issues and it is even more so thus encouraged the industry to develop
greener products and production methods (Bozza, 2022).

2.3 Process Measurement

Among the fundamental principles of productivity improvement is that the productivity of
the existing process should be measured in as much detail as possible before any attempt is made
to improve. Productivity is one of the various measures that are used to evaluate the performance
of an organization (eilon, 2019).

(Darvishi, 2019) studied on optimization of an industrial medium from molasses for bioethanol
production using the Taguchi statistical experimental design method. This study qualitek-4
software was used to optimize the production medium by the Taguchi experimental-design
method and the main factors of the industrial medium including molasses, ammonium sulfate,
urea and ph. at four different levels.
9
Productivity measurement and analysis are helpful in locating the shortages of a particular
factor, which might have caused to decrease in productivity. The basic objective of an analysis of
productivity is to find out some ideas in relation to improving productivity, and these ideas may
be applied in the field of resource utilization. So, at the industry level productivity measurement
is very helpful and comparisons of productivity change within an industry over a period of time
as also between two or more industries can indicate various interesting facts, regarding technical,
economical, and managerial aspects in the industry Generally, the importance of productivity
measures and analysis is to provide little guidance for both public and private economic policy
and measure the significance when the factors associated with the changes (Tadiyos, 2018).
Productivity measures are useful in the following management areas (Maheshwar, 2006) Goal
setting, Estimating resources, requirements, Organizational improvement, Operation control,
Resource reallocation, Responsibility accounting, The motivation for improvement

Productivity measurement has been practiced by almost all types of establishments regardless of
their size and status. productivity can be measured by total productivity is the ratio of total output
to the sum of all input factors. partial productivity is the ratio of output to one class of input. For
example, labor productivity (the ratio of output to labor input), capital productivity (the ratio of
output to capital input), and material productivity (the ratio of output to materials input). multi-
Factor Productivity is Productivity measures that consider more than one variable or factor are
termed multi-factor productivity (MFP) (Dr. Meghna Chhabra, 2018).

2.4 Factors Affecting Process

(D. Annicchiarico, 2014) in this paper author reviewed the factors affecting shrinkage of molded
parts in injection molding. Taking a human-centered approach to the advancement of technology,
it becomes evident that the development of technology takes place in a social setting and is
formed by the operational objectives and processes of usage (L. Norros, 2003). productivity is
any process or procedure developed to transfer a set of input into a specified set of output in
proper quality and quantity of goods/services thus achieving the objectives of industry and
productivity can be affected by different factors such as controllable (internal) factors and
uncontrollable (external) factors (MM. Islam, 2015).

10
2.5 Optimization of process
(Meidute-kavaliauskiene, 2023) Process optimization is a field of science whose goal is to
achieve maximum results and maximum production efficiency with the lowest costs by selecting
production operations and their technological parameters. (C Afteni, 2017) Process optimization
is an activity that selects the best possible solution to a problem, which is evaluated according to
a predetermined criterion, for example, cost of production. (D. Sabadka, 2017) Process
optimization is reducing the production process time as much as possible to increase production
productivity. (G. Tsakalidis, 2017) Process optimization automated process improvement using
predefined performance indicators (goals).

(Y Kuo, T. Y. et al., 2008) used grey based Taguchi method for simulation optimization to solve
a multi-response simulation-optimization problem. Food packaging is defined as a co-ordinated
system of preparing food for transport, distribution, storage, retailing, and end-use to satisfy the
ultimate consumer with optimal cost (Khedkar, D. K. et al., 2020).

An abundance of optimization-based algorithms has been proposed to address the production

scheduling problem. Most of them express the production scheduling problem as a mixed-integer
linear programming (MILP) problem, since it has been proved to be extremely flexible and
rigorous, while ensuring optimality (Georgiadis., 2021). Mathematical programming, especially
Mixed Integer Linear Programming (MILP), because of its rigorousness, flexibility and
extensive modeling capability, has become one of the most widely explored methods for process
scheduling problems (A.Floudas, 2015).

2.6. Bottling process

(Chakraborty K., et al., 2015) proposed a design to feed bottles into the system as a group rather
than one by one. It also describes the filling and capping stages which have been designed to
receive more than one bottle at a time. The operations are controlled by a PLC and a SCADA
system is in place to display the different stages of the system and to control the process from a
remote location. This method saved time to the whole process compared to previous ones.

Maintaining product or process quality involves, among other things, creating processes and
products that operate close to their optimum conditions even when changes occur. More
formally, robust design can be integrated into experimental design. The methods described by
Genichi Taguchi are a well-known approach to integrating DOE and product and process design.
11
While there has been much criticism of Taguchi's statistical approach, there is a broad consensus
that his principles of robust parameter design are both valid and valuable contributions to
engineering science. (mulat, 2018)

The Filling stage is dispensing the intended liquid into the bottles. The filling stage of the
bottling process can be customized to meet operation of one plant to another. One typical
example of filling stage is described: Once the bottles are detected in the input side the conveyor
motor switches ON and it starts moving in the forward direction. The bottles then reach the
desired position for filling and the conveyor stops. The corresponding pumps in process tank
switch ON and filling operation takes place (Kothari, 2004).

A proper and consistent feeding of bottles into the production line plays an important role in the
overall system productivity. Proposed another strategy to make a bottling line more efficient by
designing a special hardware to feed empty bottles automatically into a beer filling production
line (Tianxia Z et al., 2012)

The processing of beverage products involves more and more automated and mechanized
operations in order to produce the expected amount of goods for the market in a limited time.
This is one of the reasons why enterprises are trying to adopt the best automation mechanisms
that would allow them to gain competitive advantages in the market. As technology and
automation evolve, the workforce diminishes in number and technical training becomes more
important (Franson, 2016)

(KC, B., Faruk, O., Agnelli et al., 2015) Presented an application of Taguchi method to optimize
injection molding process parameters of signal/glass fiber hybrid composites. For
experimentation, L18 orthogonal array with a mixed-level design and signal-to-noise (S/N) of
smaller-the-better was used.

2.7 Design of Experiment

The Design of Experiment (DoE) method was used to determine all possible combinations. The
development of mathematical models based on experiment design method, such as Design of
Experiments (DoE), is used as one of the basic tools in process engineering for process planning
and validation (Franceschini, G. et al., 2008).

12
Experiment with Taguchi’s method must observe several steps that are the key for success of the
experiment. The experimental design steps of the Taguchi’s method are as follows (Wahyu Adhi
Yatma et al, 2018):

 Formulation of the problem.

 The purpose of the experiment.

 Determination of reaction (quality characteristics).

 Identification of influencing factors.

Influencing factors are brainstorming, flowcharting, effect diagram, Pareto analysis and Delphi
method.

 Separating of controlled and uncontrolled factors.

 Determination of quantity level and factor level values.
 Identification of interactions between factors.
 Calculation of degrees of freedom.
 Selection of the orthogonal array.
 Assignment factor on the orthogonal array.
 Provision of experimental execution.
 Data analysis.
 Result interpretation
Influencing factors in the experiment along with the level setting are Determined; there are some
things that must be considered, namely (elsevier, 2023):

 Factor levels: Representing the number of levels or attributes given by the influential
factors in the experiment, either controlled factors, noise factors, signal factors, or scale
factors.

 Number of factor levels: The number of levels and settings of the selected level depend
on the extent to which we know the process or product to be examined.

 Range of factor levels: The wider, distance used in the experiment and the higher, the
possibility of finding the effect of factors in determining the quality characteristics.

13
  Feasibility of factor levels: In selection of levels for each factor it is necessary to
consider whether the chosen level allows or can be executed in making a combination of
experiments.

2.8 Taguchi Method

Taguchi methods are statistical methods developed by Genichi Taguchi to improve the quality of
manufactured goods and more recently also applied to engineering (Rosa et al., 2022)

Taguchi Method used Signal to Noise ratio (S/N Ratio) for measure the variance from design
experiment. S/N ratio is the ratio of the mean (signal) to the standard deviation (noise). The
method of calculating the S/N ratio defends at each run of the experiment. In Taguchi Method,
there are 3 (three) characteristics values converted in Signal to Noise Ratio (S/N Ratio). These
three values describe different characteristic of quality according to the purpose of the problem.
S/N Ratio characteristic value are “Nominal is the best”, “Larger is the better” and “Smaller is
the better”. For each level of process parameters, S/N Ratio is calculated based on S/N analysis
smaller is better (Jhumar, 2023).

The Taguchi is a method used for the design of experiments. The first issue is to identify the
factors and the levels. Then, the researcher makes use of the Minitab software as in the present
case, to receive inputs from the configurations. The orthogonal array is for the combination of
the factors and levels and this is called the formation of the design of experiments. In this
context, the Taguchi methods crate a wide surface area for the experiments to take place. So,
fundamentally, after the experiments have been designed, the researcher can conduct the
experiments in several steps such that the study of the experiments can be achievable and in
depth (Darvishi, F., & Moghaddami et al., 2022).

demonstrated the uniqueness of their human centered perspective on Taguchi robust design in
the manual spraying process. They analyzed the process parameters using the stability of
worker’s hand, surface quality, the effect of fatigue, processing time and speed on paint usage.
They developed a way of the design of experiments, regression analysis and multi response
optimization. Their work integrated cost and quality as an objective in a unique perspective of
optimization, statistical analysis and quality management. From the reviewed papers, the absence
of an integrated framework on production function and maintenance function is visible. It is also
clear that prioritization was completely omitted in Taguchi studies, and process engineers were
14
unable to answer the question of which of the studied parameters is critical, important or
unimportant (Almansoori, N., Aldulaijan, S et al., 2022).

Biotechnology marketing and advertising asserted that the Taguchi method helps to improve
process quality at a much-reduced cost. The Taguchi–Pareto and Taguchi–ABC Methods have
several significant concurrent optimization and prioritization values in the area of engineering
practices, which have not been Extensively validated scientifically (Anthony, J., Anand, R.B. et
al., 2006).

discussed the neural network Taguchi scheme and the traditional Taguchi scheme. Taguchi is a
fractional factorial design of experiment based on orthogonal arrays that helps in the evaluation
of maximum number of effects from a minimum number of experimental runs while allowing for
differences in the number of factor levels (Sukthomya W. et al , 2022).

Taguchi method contains system design, parameters design and tolerance design to achieve a
robust design in order to enhance practical product quality. Taguchi design provides a powerful
and efficient method for designing processes that operate consistently and optimally over a
variety of conditions (Athreya1 S et al., 2022).

2.9 Literature review Gap

From the above discussion, the following important gaps in knowledge were revealed:

Numerous studies exist on the manufacturing quality practices but analysis and optimization of
process using design of experiment for bottling plants have been completely omitted in literature.
From literature search, very limited studies relate to optimization of bottling process in the
beverage plant. The optimal parametric settings, global optimized values and specific global
values of the segments were not documented elsewhere in the literature. Most review m, of the
literature used simulation model and linear programing to optimization problem Analysis and
optimization of bottling process where a combination of the above-mentioned techniques could
be implemented (i.e. Taguchi method and SPSS software analysis, Taguchi-Pareto and
techniques among others).

15
 Chapter-3

RESEARCH METHODOLOGY
3.1 Research Design approach
(Kothari C. , 2004) a research design is a road-map for the preparations and executions of an
overall research project. Quantitative research is an approach for testing objective theories by
examining the relationship among variables. A statistical procedure is made to interpret and
analyze the numbered collected data. The final written report has a set structure consisting of
introduction, literature, and theory, methods, results, and discussion. Therefore, a research
framework has been drafted which can clearly show the steps from research problem definitions
to the final parts of the conclusion and recommendation. Defining research problems and
opportunities Review of previous research findings Review of concepts and theories Developing
research questions and defining objectives Research designing and sampling Data collection
analysis and interpretations.

3.2 Research Methodology

Research methodology has been defined as a framework that outlines the methods and
procedures to be followed when collecting and analyzing the required information from
participants (Zikmund, W.G., Babin et al , 2010). This research has followed several methods
and techniques of fact gathering, data collections, and analysis tools. The appropriateness and
selections of various methods and techniques are based on the type and characteristics of the
nature of problems. The quantitative research approach is selected for using numerals to
analyzing, testing objective theories by examining the relationship among variables.

3.3 data collection method

Data collection is an important journey of research work for gathering and measuring
information from the working environment (i.e. variables) which is/are going to be studied and
addressed by the researcher. As a result, to answer the above-stated research questions, this study
encompasses primary and secondary means of data collection techniques. Original and raw data

16
can be easily fetched through a primary data collection source. However, commonly used
materials and references are discovered by secondary data collection sources.
3.3.1 Primary data collection

Primary data collection refers a process of obtaining a desired data (original data) through
observation and Informal interview. In the primary data collection means, different sources will
be used for facilitating the study process. These types of data are very essential to get
information from the primary sources and could help to perform statistical surveys. In general,
Questionnaires, interviews, and direct observations are the key sources for data gathering.

3.3.2 Secondary data collection

Despite its validity, secondary data collection sources are very important when the environment
is rare for information. For this study, the following qualitative and quantitative data sources will
be included. Company reports and records information regards to the case company background,
production status, and facts will be highlighted. Published books whose topics are related to the
objective of this research will be reviewed as a result, to tackle the problem which is going to
solve this study; several kinds of literature (articles, books, journal and etc.) are overviewed with
their strengths and gaps.

3.4 Data analysis method

The data analysis methods perceived in this research works were, MS excel, and table, chart,
and different software were selected. This paper deals with the identification and optimization of
dominant factors of the various segments of a bottling process plant in the perspective of
Taguchi methodical optimization of process factors.
Thus, these data were interpreted through a quantitatively method of data analysis by using MS
Excel, and Minitab software were selected as the appropriate approaches to analyze and verify
the collected data And Taguchi method and by integrating Taguchi pareto and Taguchi ABC was
selected for validation of the proposed solution of make-span and optimization of bottling
process, that achieved by design of experiment and finally, a conclusion and recommendation
were reached.
3.4.1 Integration of Taguchi–ABC and Taguchi–Pareto
By integrating Taguchi with either Pareto or ABC, the benefits of synergy in optimization and
prioritization are earned. The goal of optimization is to attain the best bottling plant performance
17
in the context of the parameters for the joint maintenance and production such as the number of
shifts, planned maintenance hours and production hours worked. Furthermore, prioritizing is a
practice whereby the most essential factors in the association of production and maintenance
functions are established. At the first look, the process engineer may be excited to conclude that
all the factors have elevated and critical priority. However, with the application of either the
ABC analysis or Pareto principles distinctions are made to know those factors that are critical.
The ABC analysis and Pareto technique assist the process engineer to answer the question of
which of the factors being considered is the most critical. It offers the process engineer a proper
technique to assess the need to achieve each parameter in the bottling process. In this way, the
process engineer makes enlightened decisions on what needs to be achieved, what should not be
achieved and the time to attend to particular factors in the production–maintenance interface.
Substantial productive hours may be wasted considering the factors if done intuitively without
the use of ABC analysis or the Pareto principle. With the criteria of critical, necessary and
optional priority according to ABC analysis could be determined. For Pareto analysis, there is
dependence on the 80–20 rule.
3.4.2 Taguchi–Pareto
A significant concern in optimization is the neglect of weight assigned to individual factors in a
Taguchi methodical evaluation exercise. Taguchi–Pareto is a new prioritization approach that is
concurrently activated with optimization while also attempting to optimize the response from the
maintenance and production outcomes in an evaluation exercise. The method was first
introduced in (Ajibade et al., 2019). The assessment, based on the Taguchi–Pareto is hinged on
the 80–20 rule of Pareto, which states that the 80% of success in an endeavor is taken care of by
20% of the group members. In association with the analysis of variance (ANOVA), the variance
of the individual factor is noticed, their cumulative frequency is calculated and roughly at 80% of
the cumulative frequency a cut-off mark is made and this represents the Pareto candidate in the
evaluation. Experts are attracted to the use of Taguchi Pareto given its simplicity and advantages.
A principal advantage of the method is that it reveals which factor(s) control the system with
which attention needs to be focused on for the utmost improvement of performance in the
process. Recently, (Raji, 2019) applied the Taguchi–Pareto scheme to reduce downtime in a
maintenance system, and this has opened a gate to its usefulness in the maintenance field.

Equation (1) states the objective function for the signal-to-noise determination using the
Taguchi–Pareto method (Ajibade et al., 2019)
18
S/N=-10log10 (1/n 80-20 i ) …………………………………………. Equation 1

Where Yi is the evaluated value of the lower-the-better quality feature, n represents the number
of trials, S/N is the signal-to-noise quotient of the system while P80–20 is the obtained ratio from
the analysis of variance.
3.4.3 Taguchi–ABC
The ABC analysis is a powerful classification tool in inventory analysis and has been used to
group items into three classes, based on A (roughly 69%), B (roughly 29%) and C (the balance
from 100%). In the new method, the ABC concept is integrated with Taguchi method such that
prioritization of the factors is done according to the ABC classification while concurrently
optimizing the response of the system using the Taguchi method. Analysis using the Taguchi
ABC framework has been proved as a competent approach to concurrently optimize factors
using ABC scheme and the Taguchi route, as shown in (Raji, 2019). Eqn (2) states the objective
function for the signal-to-noise determination using the Taguchi–ABC method (Ajibade et al.,
2019):

(
S/N =-10log10 1/n (ABC) )…………………………………………………. Equation 2
Where Yi is the evaluated value of the lower-the-better quality feature, n represents the number
of trials, S/N is the signal-to-noise quotient of the system while ABC is the classified index
obtained from ABC analysis to the factor levels.
3.7 Objective function and procedure of Taguchi method
Taguchi method uses the S/N ratio covering the mean and variations of the experimental results
(Anthony, J., Perry, D. et al., 2006a), (Abhishek, K., Datta, S et al., 2013), and aiding in
preparing the ANOVA. Three kinds of signal-to-noise ratios are used (Aksu, 2010): “smaller-
the- better” (Eqn 3), “larger-the-better” (Eqn 4) and “nominal-the-best” (Eqn 5).

S/N =-10log10 (1/n )…………………………………………………………. Equation 3

(
S/N=-10l0g10 1/n i ) ………………………………………………………… Equation 4

S/N=-10l0g10 / ………………………. …………………………………………… Equation 5

19
 Δ =highest - lower ……………………………………………………………………...Equation 6

Literature data from different sources is obtained

Identify parameters and levels of the bottling company

Taguchi method Taguchi-Pareto Taguchi-ABC

method

Obtain the orthogonal array for the problem

Develop the signal to noise ratios for the problem

Obtain the Pareto-reduced Obtain the ABC-based

experimental trials experimental trials

Obtain the response table

Determine the optimal parametric settings, delta values and the ranks of the
parameters

Figure 1: 2 An illustration of the experimental design for bottling process plan

20
 Chapter-4

DATA COLLECTION AND ANAYSIS

4.1 Introduction
In this study, optimization of bottling manufacturing process using the Taguchi design of
experiment (DOE) where Taguchi’s orthogonal array is used. The orthogonal array of L16
design was generated using the Minitab software with the total degree of freedom as 16 and the
larger-the-better method is used for the S/N ratio computations. Production worked hour, number
of shift and planned maintenance hour are selected as input factors. Besides, Taguchi method
design is one of the optimization methods in bottling manufacturing process; however, the
AMWSC could not allow assessing and designing change on layout and other physical parts of
machine during experimental work, so the study was focused on process parameter optimization.
As a result, the optimization of bottling water was produced by considering three control factors
with 4 levels; namely, production worked hour, number of shifts and planned maintenance time
that had significant effects on production. The collected date of production worked hour refers to
the actual hours worked by employee or operators on the production. It excludes breaks,
maintenance, down time and other nonproductive activities. By production worked hour focus on
the actual time spent on productive activities by the workforce.
In the study, an attempt was made to carry out an experiment to gain insight into the effect of
three various autonomous variables, while each variable has four groups of values, called levels.
For this purpose, an L16 orthogonal array was the choice. Table 1 shows the PWH, NOS and the
PM of bottled water for 12 months in an April 2022 to February 2023 year. The mean of three
consecutive production periods was taken to form values under each level table 2 for mean.
However, to obtain the maximum and minimum values, the three consecutive values were also
considered by with attention to picking the maximum and minimum values respectively from the
data. This procedure was also used to generate the mean, maximum and minimum values for the
entire factors concerned table 2.
Table 1: Collected data of NOS, PM and PWH

Months PWH NOS PM

1 384.3 73 9.149

21
2 323.6 54 6
3 323.1 58 31.2
4 320.4 58 15.1
5 303.6 65 4.43
6 347.9 66 1.2
7 425.5 71 6.39
8 379.2 59 3.78
9 415.5 70 10.23
10 400.4 71 14.3
11 423.6 67 7.48
12 428.6 82 9.9

The above table shows that the number of shifts has the greatest influence on production
outcome. Also, for the minimum value, the planned maintenance has the highest effect on the
outcome of production. Furthermore, from the mean values, it is shown that planned
maintenance has the highest effect on the outcome of production.
Table 2: the mean, maximum and minimum values of production worked hour, planned
maintenance hour and number of shifts.

Factors level 1 level 2 level 3 level 4

Mean values
PWH 343.67 323.97 406.73 417.53
NOS 61.67 63 66.67 73.33
PM 15.423 6.91 6.8 10.56
Maximum values
PWH 384.3 347.9 425.5 428.6
NOS 73 66 71 82
PM 31.2 15.1 10.23 14.3
minimum value
PWH 323.1 303.6 379.2 400.4
NOS 54 58 59 67
PM 6 1.2 3.78 7.48

22
From the above table arranged the collected data into the effect of three various autonomous
variables, while each variable has four groups of values in four levels. For this purpose, an L16
orthogonal array was the choice. The mean of three consecutive production periods was selecting
to form values under each level. However, to obtain the maximum and minimum values, the
three consecutive values were also considered by selecting the maximum and minimum values
respectively from the data.

4.2 Taguchi’s orthogonal arrays, factors and signal-to-noise ratios

The orthogonal array is for the combination of the factors and levels and this is called the
formation of the design of experiments. The smaller the better and larger the better signal to
noise criteria are two popular drivers of the Taguchi method, producing the response table, delta
values, rank and optimal parametric settings. However, they are different in some respects. The
smaller the better is used when the specific output that emerges from the system is not desirable
such as energy wastage during the friction stir welding process. However, the larger the better
signal to noise criterion is used when there is a need to maximize a particular output (133b). The
orthogonal array of L16 design was generated using the Minitab with the total degree of freedom
as 11 and the larger-the-better method is used for the S/N ratio computations. The signal-to-noise
ratio measures how the response varies relative to the nominal or target value under different
noise conditions. You can choose from different signal-to-noise ratios, depending on the goal of
your experiment. For static designs, Minitab offers four signal-to-noise ratios. The S/N ratio
(signal-to-noise) analysis is a distinct feature of the Taguchi optimization method that affords a
statistical measure of the ratio between the factors which control the average performance of the
system and the factors which are cumbersome or complicated to control. The contribution of the
control factors on the S/N ratio can be broken down as a result of the statistical independence of
the Taguchi experimental design.

Table 3: Taguchi’s orthogonal arrays, factors and signal-to-noise ratios for mean

orthogonal arrays Factors mean

NO PWH NOS PM PWH_1 NOS_1 PM_1 SNR of

mean

1 1 1 1 343.67 61.67 15.423 28.26289

23
2 1 2 2 343.67 63 6.91 21.5071

3 1 3 3 343.67 66.67 6.8 21.37476

4 1 4 4 343.67 73.33 10.56 25.15133

5 2 1 2 323.97 61.67 6.91 21.50464

6 2 2 1 323.97 63 15.423 28.27253

7 2 3 4 323.97 66.67 10.56 25.13238

8 2 4 3 323.97 73.33 6.8 21.38231

9 3 1 3 406.73 61.67 6.8 21.36771

10 3 2 4 406.73 63 10.56 25.12131

11 3 3 1 406.73 66.67 15.423 28.30226

12 3 4 2 406.73 73.33 6.91 21.52114

13 4 1 4 417.53 61.67 10.56 25.11628

14 4 2 3 417.53 63 6.8 21.36995

15 4 3 2 417.53 66.67 6.91 21.51319

16 4 4 1 417.53 73.33 15.423 28.34093

Table 4: response Table for Signal to Noise Ratios of mean

Larger is better

Level PWH NOS PM

1 24.07 24.06 28.29
2 24.07 24.07 21.51
3 24.08 24.08 21.37
4 24.09 24.10 25.13
Delta 0.01 0.04 6.92
Rank 3 2 1

24
Zareh et al. (2013) noted that irrespective of the quality characteristic used, a higher S/N ratio
indicates the desired quality characteristic is being met. From the Taguchi S/N response table in
Table 4, the optimal setting of parameters for the bottling process is PWH4NOS4PM1.
Table 5: Taguchi’s orthogonal arrays, factors and signal-to-noise ratios for maximum

NO PWH NOS PM PWH_1 NOS_1 PM_1 SNR maximum

1 1 1 1 384.3 73 31.12 33.88278

2 1 2 2 384.3 66 15.1 28.12281

3 1 3 3 384.3 71 10.23 24.87647

4 1 4 4 384.3 82 14.3 27.74199

5 2 1 2 347.9 73 15.1 28.16096

6 2 2 1 347.9 66 31.12 33.7318

7 2 3 4 347.9 71 14.3 27.69819

8 2 4 3 347.9 82 10.23 24.89796

9 3 1 3 425.5 73 10.23 24.8818

10 3 2 4 425.5 66 14.3 27.67401

11 3 3 1 425.5 71 31.12 33.84938

12 3 4 2 425.5 82 15.1 28.20064

13 4 1 4 428.6 73 14.3 27.70975

14 4 2 3 428.6 66 10.23 24.8632

15 4 3 2 428.6 71 15.1 28.15347

16 4 4 1 428.6 82 31.12 34.02766

Table 6: response table for signal to noise ratios of maximum

Larger is better

Level PWH NOS PM

1 28.66 28.66 33.87
2 28.62 28.60 28.16
25
3 28.65 28.64 24.88
4 28.69 28.72 27.71
Delta 0.07 0.12 8.99
Rank 3 2 1

Zareh et al. (2013) noted that irrespective of the quality characteristic used, a higher S/N ratio
indicates the desired quality characteristic is being met. From the Taguchi S/N response table in
Table 6, the optimal setting of parameters for the bottling process is PWH4NOS4PM1.
Table 7: Taguchi’s orthogonal arrays, factors and signal-to-noise ratios for minimum

No PWH NOS PM PWH_1 NOS_1 PM_1 SNR minimum

1 1 1 1 323.1 54 6 20.2795

2 1 2 2 323.1 58 1.2 6.3529

3 1 3 3 323.1 59 3.78 16.3027

4 1 4 4 323.1 67 7.48 22.1932

5 2 1 2 303.6 54 1.2 6.3526

6 2 2 1 303.6 58 6 20.2863

7 2 3 4 303.6 59 7.48 22.1774

8 2 4 3 303.6 67 3.78 16.3066

9 3 1 3 379.2 54 3.78 16.2994

10 3 2 4 379.2 58 7.48 22.1759

11 3 3 1 379.2 59 6 20.2885

12 3 4 2 379.2 67 1.2 6.3534

13 4 1 4 400.4 54 7.48 22.1652

14 4 2 3 400.4 58 3.78 16.3023

15 4 3 2 400.4 59 1.2 6.353

16 4 4 1 400.4 67 6 20.2986

26
 Table 8: response table for signal to noise ratios of minimum
Larger is better

Level PWH NOS PM

1 16.282 16.274 20.288
2 16.281 16.279 6.353
3 16.279 16.280 16.303
4 16.280 16.288 22.178
Delta 0.003 0.014 15.825
Rank 3 2 1

Zareh et al. (2013) noted that irrespective of the quality characteristic used, a higher S/N ratio
indicates the desired quality characteristic is being met. From the Taguchi S/N response table in
Table, the optimal setting of parameters for the bottling process is PWH1NOS4PM4.
Table 9: Taguchi’s orthogonal arrays, factors and signal-to-noise ratios for mean, maximum and
minimum

orthogonal array Factors (mean) S/N ratios

exp no PWH NOS PM PWH_1 NOS_1 PM_1 SNR SNR SNR min
mean Max
1 1 1 1 343.67 61.67 15.423 28.2629 33.8828 20.2795
2 1 2 2 343.67 63 6.91 21.5071 28.1228 6.3529
3 1 3 3 343.67 66.67 6.8 21.3748 24.8765 16.3027
4 1 4 4 343.67 73.33 10.56 25.1513 27.742 22.1932
5 2 1 2 323.97 61.67 6.91 21.5046 28.161 6.3526
6 2 2 1 323.97 63 15.423 28.2725 33.7318 20.2863
7 2 3 4 323.97 66.67 10.56 25.1324 27.6982 22.1774
8 2 4 3 323.97 73.33 6.8 21.3823 24.898 16.3066
9 3 1 3 406.73 61.67 6.8 21.3677 24.8818 16.2994
10 3 2 4 406.73 63 10.56 25.1213 27.674 22.1759
11 3 3 1 406.73 66.67 15.423 28.3023 33.8494 20.2885
12 3 4 2 406.73 73.33 6.91 21.5211 28.2006 6.3534
13 4 1 4 417.53 61.67 10.56 25.1163 27.7097 22.1652
14 4 2 3 417.53 63 6.8 21.3699 24.8632 16.3023
15 4 3 2 417.53 66.67 6.91 21.5132 28.1535 6.353
16 4 4 1 417.53 73.33 15.423 28.3409 34.0277 20.2986

27
 Table 10: Taguchi S/N ratio response table for mean, maximum and minimum

S/R for mean S/R for maximum S/R for minimum

Level PWH NOS PM Level PWH NOS PM Level PWH NOS PM
1 24.07 24.06 28.29 1 28.66 28.66 33.87 1 16.282 16.274 20.288
2 24.07 24.07 21.51 2 28.62 28.6 28.16 2 16.281 16.279 6.353
3 24.08 24.08 21.37 3 28.65 28.64 24.88 3 16.279 16.28 16.303
4 24.09 24.1 25.13 4 28.69 28.72 27.71 4 16.28 16.288 22.178
Delta 0.01 0.04 6.92 Delta 0.07 0.12 8.99 Delta 0.003 0.014 15.825
Rank 3 2 1 Rank 3 2 1 Rank 3 2 1

This method can also be used to find the most optimum combination among the input parameters
which will result in getting the maximum possible output, that is, maximum manufacturing hour.
After completion of table 3- 10 was arrived at by first considering the orthogonal array, L16, and
then following this to produce the actual figures in the next few columns after the orthogonal
array. Finally, the signal-to-noise ratio was computed based on Eqn (4) since the larger-the-better
strategy was desired for all the factors considered. Then, this is repeated by using the maximum
and minimum values instead of the mean to obtain the individual S/N ratios. The mean,
minimum and maximum values are measures of data. So for extensive testing of the data in
Taguchi analysis, apart from using the mean values for analysis, an attempt was made
unconventionally to use the minimum and maximum values in the data array from the bottling
process plant.

4.3 Main effect plots

The main effect plot shows the displays the contributive effect of each parameter to the quality
characteristic represented vertically as the mean S/N ratios. For each of the parameters, the level
which gives the highest S/N ratio is the optimum condition.

28
Figure 2: Main effect plots for mean, maximum and minimum values

4.4 Analysis of variance (ANOVA)

According to Ajibade et al. (2019b), analysis of variances (ANOVA) is a mathematical method
is used to evaluate the individual influence of the governable factors on the quality characteristic.
The Taguchi method provides the details for selecting the optimal setting of parameters for the
system and important information useful for further evaluation. With the use of the ANOVA, the
overall variation from the mean S/N ratio is separated into individual parameters with
corresponding errors. This is made possible as a result of the statistical independence of the
Taguchi method. The total sum of squares (TSS) is first determined as follows (Zareh et al.,
2013):

n is defined as the number of experimental trials in the orthogonal array and (S/N)k is the S/N
ratio of the kth experiment. The sum of squares due to the deviation from the total mean S/N
ratio for each of the parameter is defined mathematically as follows (Zareh et al., 2013):

29
Where m is the number of parametric levels, s is defined as the level number of the parameter A,
(S/N) i is the sum of the S/N ratios involving this parameter A and level s, and p is the repetition
of each level of parameter A. The contribution of the Ath parameter can be obtained in terms of
percentage as follows (Zareh et al., 2013):

For mean
Anova: Two-Factor Without Replication

SUMMARY Count Sum Average Variance

Level 1 3 76.42 25.47333 5.950233
Level 2 3 69.65 23.21667 2.184533
Level 3 3 69.53 23.17667 2.448033
Level 4 3 73.32 24.44 0.3571

PWH 1 4 96.31 24.0775 9.17E-05

NOS 2 4 96.31 24.0775 0.000292
PM 3 4 96.3 24.075 10.92517

ANOVA
Source of
Variation SS Df MS F P-value F crit
Level 10.89687 3 3.632289 0.996067 0.456199 4.757063
Factor 1.67E-05 2 8.33E-06 2.29E-06 0.999998 5.143253
Error 21.87978 6 3.646631

Total 32.77667 11

FOR MAXIMUM
30
Anova: Two-Factor Without Replication

SUMMARY Count Sum Average Variance

Level 1 3 91.19 30.39667 9.048033
Level 2 3 85.38 28.46 0.0676
Level 3 3 82.17 27.39 4.7251
Level 4 3 85.12 28.37333 0.330233

PWH 4 114.62 28.655 0.000833

NOS 4 114.62 28.655 0.0025
PM 4 114.62 28.655 14.19497

ANOVA
Source of
Variation SS Df MS F P-value F crit
Levels 14.25297 3 4.750989 1.005786 0.452566 4.757063
Factors 0 2 0 0 1 5.143253
Error 28.34193 6 4.723656

Total 42.5949 11

FOE MINIMUM
Anova: Two-Factor Without Replication

SUMMARY Count Sum Average Variance

Level 1 3 52.844 17.61467 5.360049
Level 2 3 38.913 12.971 32.84844
Level 3 3 48.862 16.28733 0.000184
Level 4 3 54.746 18.24867 11.57976

31
 PWH 4 65.122 16.2805 1.67E-06
NOS 4 65.121 16.28025 3.36E-05
PM 4 65.122 16.2805 49.79878

ANOVA
Source of
Variation SS Df MS F P-value F crit
Levels 49.81955 3 16.60652 1.000625 0.454491 4.757063
Factors 1.67E-07 2 8.33E-08 5.02E-09 1 5.143253
Error 99.57688 6 16.59615

Total 149.3964 11

4.5 Taguchi-Pareto Analysis (TPA)

The Pareto principle, also called 80/20 rule, states that for a greater number of events
approximately, 80% of the effect is obtained from 20% of the causes (Ajibade et al., 2019).
Applying this principle to the Taguchi method, 20% of the parameters were noticed to represent
80% of the total cumulative, which shows that it is economical. In this study, Taguchi-Pareto
analysis was conducted for the mean, maximum and minimum. The ANOVA result obtained
from the Taguchi mean, maximum and minimum tables enabled us to know the factors that will
be used for Pareto analysis. For the Taguchi mean, all the three factors were considered to be
significant, i.e. PHW, NOS and PM. These factors were used to prepare the signal-to-noise ratio
(S/N ratio). There are three columns as follows: S/N ratio, the cumulative value and the per cent
cumulative. This table enables us to know the 20% that could control the 80%. Table shows the
Taguchi–Pareto experimental design. The term that exceeds 80% was turned to zero since the
last 20% accounts for the 80%. To explain factors and levels, the following details may be
useful. A common procedure implemented in Taguchi scheme is to establish the process design
parameters influencing the bottling plant, namely the planned maintenance, number of shifts and
production hours. These parameters are controllable variables of the bottling plant that influence
the performance measure.
32
Table 11: Taguchi Pareto for mean

No S/N ratio order Mean Cumulative % cumulative

1 21.37 21.37 7%
2 21.51 42.88 15%
3 24.06 66.94 23%
4 24.07 91.01 32%
5 24.07 115.08 40%
6 24.07 139.15 48%
7 24.08 163.23 56%
8 24.08 187.31 65%
9 24.09 211.4 73%
10 24.1 235.5 82%
11 25.13 260.63 90%
12 28.29 288.92 100%

Table 12: Taguchi Pareto for maximum

No S/N ratio order Cumulative % cumulative
1 24.88 24.88 11%
2 27.71 52.59 23%
3 28.16 80.75 35%
4 28.62 109.37 48%
5 28.65 138.02 60%
6 28.66 166.68 73%
7 28.69 195.37 85%
8 33.87 229.24 100%

Table 13: Taguchi Pareto for minimum

No S/N ratio order Cumulative % cumulative

1 16.274 16.274 12%
2 16.279 32.553 25%
3 16.279 48.832 37%
4 16.28 65.112 50%
5 16.28 81.392 62%
6 16.281 97.673 75%
33
7 16.282 113.955 87%
8 16.288 130.243 100%

This table enables us to know the 20% that could control the 80%. Table 14 shows the Taguchi–
Pareto experimental design. The term that exceeds 80% was turned to zero since the last 20%
accounts for the 80%. To explain factors and levels, the following details may be useful. A
common procedure implemented in Taguchi scheme is to establish the process design parameters
influencing the bottling plant, namely the planned maintenance, number of shifts and production
hours. These parameters are controllable variables of the bottling plant that influence the
performance measure.
Table 14: Taguchi–Pareto experimental design –mean, maximum and minimum (with factors
and their levels)

Factors level 1 level 2 level 3 level 4

Mean values
PWH 343.67 323.97 406.73 417.53
NOS 61.67 63 66.67 0
PM 0 6.91 6.8 0
Maximum values
PWH 384.3 347.9 425.5 0
PM 0 15.1 10.23 14.3
minimum value
PWH 0 303.6 379.2 400.4
NOS 54 58 59 0

4.6 Taguchi ABC

In the ABC classification, factors and levels are grouped into A, B and C but limited to A and B in this
case as the factors and levels available will not be able to bring out the factors ABC .Factors A and B
were combined with Taguchi method whereby a special mixed L16orthogonal array was generated to
obtain an optimized result. The revised Taguchi AB S/N ratio response shows that A2B1 is the optimal
parametric setting of the model. The mean values are the mean of each three periods. A special mix L16

34
orthogonal array was designed, and S/N ratio table (Table 14) was constructed from this special mix
table, showing that B
Table 15 : Taguchi–Pareto’s orthogonal arrays, factors and signal-to noise ratios

orthogonal array Factors (mean) S/N ratios

N SNR for SNR
PWH_ NOS_ SNR for
O PWH NOS PM PM_1 maximu minimu
1 1 mean
m m
1 0 38.6741
1 1 1 343.67 61.67 51.69341 34.64788
4
2 1 2 2 343.67 63 6.91 21.5071 26.58314 35.26856
3 21.3747
1 3 3 343.67 66.67 6.8 23.20474 35.41704
6
4 0 0 50.7228 0
1 4 4 343.67 26.11101
3
5 21.5046
2 1 2 323.97 61.67 6.91 50.82909 37.52291
4
6 0 38.8359
2 2 1 323.97 63 26.58167 38.12318
1
7 0 39.3087
2 3 4 323.97 66.67 23.20406 38.26635
7
8 0 19.6585
2 4 3 323.97 6.8 26.10969 49.64604
7
9 21.3677
3 1 3 406.73 61.67 6.8 52.57799 37.57098
1
10 0 38.8941
3 2 4 406.73 63 26.58437 38.17843
4
11 0 39.3737
3 3 1 406.73 66.67 23.2053 38.32346
6
12 0 19.7986
3 4 2 406.73 6.91 26.11212 51.57737
1
13 0 38.7180 0
4 1 4 417.53 61.67 37.57989
5
14 21.3699
4 2 3 417.53 63 6.8 23.57954 38.18867
5
15 21.5131
4 3 2 417.53 66.67 6.91 20.19751 38.33405
9
16 0 52.4137
4 4 1 417.53 23.10672 52.04988
5

Table 16: Taguchi–Pareto-80–20 S/N ratio response table ratio

S/R for mean S/R for maximum S/R for minimum

Level PWH NOS PM Level PWH PM Level PWH NOS

35
1 33.07 30.07 42.32 1 31.9 51.7 1 35.11 36.83

2 29.83 30.15 21.08 2 31.68 25.83 2 40.89 37.44

3 29.86 30.39 20.94 3 32.12 22.45 3 41.41 37.59

4 33.5 35.65 41.91 4 22.29 25.36 4 41.54 51.09

Delta 3.68 5.58 21.38 Delta 9.83 29.25 Delta 6.43 14.26

Rank 3 2 1 Rank 2 1 Rank 2 1

(a mixture of the NOS and PM) has that highest delta value. It means that these two factors will
better influence the outcome of production. Similar to the procedure for the mean values, the
maximum values for every three periods in the 12 periods were taken (Table 16) using the larger
the better method. It is shown that B (a mixture of PM and NOS) has the highest delta value,
showing that the outcome of production will be better influenced by these two factors.
Similarly, for the minimum values of each three periods of the 12 periods, results show that the
PM and NOS will influence the outcome of production greatly. From the Taguchi AB mean,
maximum and minimum value tables and results, it was observed that PM and NOS are the
factors that will influence the outcome of production than the PHW.
Table 17: Taguchi AB’s special mix L16 design, factors and signal-to-noise ratios

Orthogonal array Factors (mean S/N ratios

no A B A B mean Max min
1 1 A1 343.67 61.67 38.67414 40.12281 37.53853
2 1 A2 343.67 63 38.85357 39.27494 38.14112
3 1 A3 343.67 66.67 39.32847 39.8897 38.28489
4 1 A4 343.67 73.33 40.12258 41.09322 39.34895
5 2 C1 323.97 15.423 26.76385 32.8586 18.57163
6 2 C2 323.97 6.91 19.79789 26.58167 4.593857
7 2 C3 323.97 6.8 19.65857 23.20406 14.55946
8 2 C4 323.97 10.56 23.47897 26.10969 20.4857
9 3 A1 406.73 61.67 38.71307 40.15077 37.57098
10 3 A2 406.73 63 38.89414 39.29793 38.17843
11 3 A3 406.73 66.67 39.37376 39.9162 38.32346
12 3 A4 406.73 73.33 40.17701 41.12821 39.39829
13 4 C1 417.53 15.423 26.76776 32.87044 18.57235
14 4 C2 417.53 6.91 19.79867 26.58445 4.593886
15 4 C3 417.53 6.8 19.65933 23.20534 14.55975
16 4 C4 417.53 10.56 23.4808 26.11219 20.48682

36
Table 18: Taguchi AB S/N ratio response table

Mean Max Min

Level A B A B A B
1 39.24 32.73 40.1 36.5 38.33 28.06
2 22.42 29.34 27.19 32.93 14.55 21.38
3 39.29 29.51 40.12 31.55 38.37 26.43
4 22.43 31.81 27.19 33.61 14.55 29.93
Delta 16.86 3.39 12.93 4.95 23.82 8.55
Rank 1 2 1 2 1 2

37
 References

A.Floudas, C. (2015). Mixed Integer Linear Programming in Process Scheduling: Modeling,

Algorithms, and Applications. . Annals of Operations Research, 131-162.

Abhishek. (2013). Simulation Modelling Practice and Theory. Journal of minimally, 31-40.

Abhishek, K., Datta, S et al. (2013). “Taguchi approach followed by fuzzy linguistic reasoning
for quality-productivity optimization in machining operation: a case study”. Journal of
Manufacturing Technology Management, Vol. 24 No. 6, pp. 929-951.

Achraf Ammar, B. S. ( 2019). performance optimization of a flexible manufacturing system

using simulation. Ppuplisher online, Volume 95, Issue 11.

Ajibade. (2019). “Optimisation of water absorption parameters of dual-filler filled composites

using Taguchi and moderated Taguchi techniques”. Kufa Journal of engineering.

Ajibade et al. (2019). “Optimisation of water absorption parameters of dual-filler filled

composites using Taguchi and moderated Taguch techniques”. Kufa Journal of
engineering.

Ajibade, O.A., Agunsoye, J.O et al. (2019). Optimisation of water absorption parameters of dual-
filler filled composites using Taguchi and moderated Taguchi techniques”. Kufa Journal
of, Vol. 10 No. 2, pp. 134-151.

Aksu, B. (2010). Shainin and Taguchi Methods and Their Comparison on an Application. 1st
International Symposium on Computing in Science and Engineering, June 3-5, pp. 801-
809.

Almansoori, N., Aldulaijan, S et al. (n.d.). “Manual.

Almansoori, N., Aldulaijan, S et al. (2022). “Manual spray painting process optimization using
Taguchi robust design”. International Journal of Quality and Reliability Management.,
10.1108/IJQRM-07-2019-0248.

Anthony, J., Anand, R.B. et al. (2006). “Multiple response optimization using Taguchi
methodology and neuro fuzzy based model”. Journal of Manufacturing Technology
managment, Vol. 17 No. 7, pp. 908-925.

Anthony, J., Perry, D. et al. (2006a). “An application of Taguchi method of experimental design
for new product design and development process”. Assembly Automation, Vol. 26 No. 1,
pp. 18-24.
38
Athreya1 S et al. (2022). Application of Taguchi Method for Optimization of Process Parameters
in Improving the Surface Roughness of Lathe Facing Operation. International Refereed
Journal of Engineering and Science (IRJES).

Basso. (2019). Optimizing perioperative outcomes with selective bowel resection following an
algorithm based on preoperative imaging for bowel endometriosis. Journal of minimally.

Belavendram, N. (1995). Quality by Design Taguchi Techniques for Industrial Experimentation.

internatinal journal.

Belavendram, N. (1995). Quality by Design Taguchi Techniques for Industrial Experimentation.

internatinal journal.

Belavendram, N. (1995). Quality by Design Taguchi Techniques for Industrial Experimentation.

international journal.

Belavendram, N. (n.d.). Quality by Design Taguchi Techniques for Industrial Experimentation.

international journal for enginer.

Bozza. (2022). Current regulatory and market frameworks in green cosmetics. The role of
certification. Sustainable Chemistry and Pharmacy.

C Afteni, G. F. (2017). A review on optimization of manufacturing process performance.

International Journal of Modeling and Optimization.

Chakraborty K., et al. (2015). controlling Process of a Bottling Plant using PLC and SCADA.
Indonesian Journal of Electrical Engineering and Informatics (IJEEI), 39~44.

D. Annicchiarico, J. A. (2014). Review of factors that affect shrinkage of molded part in

injection molding. Materials and Manufacturing Processes.

D. Sabadka, V. M. (2017). The use of lean manufacturing techniques – SMED analysis to

optimization of the production process. Advances in Science and Technology Research
Journal.

Daneshjo. (2023). Assessment of environmental performance and development sustainability of

systems product. Ecological Engineering &Environmental Technology.

Daneshjo, N. (2023). Assessment of environmental performance and development sustainability

of systems product. Ecological Engineering &Environmental Technology.

Daneshjo, N. (2023). Assessment of environmental performance and development sustainability

of systems product. Ecological Engineering &Environmental Technology. international
journal.

Daneshjo, N. (2023). Assessment of environmental performance and development sustainability

of systems product. Ecological Engineering &Environmental Technology.
39
Daneshjo, N. (2023). Assessment of environmental performance and development sustainability
of systems product. Ecological Engineering &Environmental Technology.

Daneshjo, N. (n.d.). Assessment of environmental performance and development sustainability

of systems product. Ecological Engineering &Environmental Technology.

Darvishi, F. &. (2019). Optimization of an industrial medium from molasses for bioethanol
production using the Taguchi statistical experimental-design method. International
Journal of Quality & .

Darvishi, F., & Moghaddami et al. (2022). Optimization of an industrial medium from molasses
for bioethanol production using the Taguchi statistical experimental-design method.
Fermentation. industrial journal.

Derbe. (2023). Impact of dairy production on smallholder households Food Security.

International Journal of industrial.

Dr. Meghna Chhabra, D. R. (2018). Productivity in manufacturing industries. International

Journal of Innovative Science and Research Technology.

eilon. (2019). productivity improvement of garnment. jobshop scheduling, 5-7.

elsevier. (2023). Design of experiment for engineers and scientsts third edition. jiju antony.
netherlands: mattew peans.

Fisher, R. (1970). s wide-ranging account of methods of statistical inference, and also for the
wealth of applications made to biology. Statistical methods for research workers.

Franceschini, G. et al. (2008). Model-based design of experiments for parameter precision: state
of the art. Chem. Eng. Sci.

Franson, D. (2016). Encyclopaedia of occupational health and safety. Beverage Industry .

G. Tsakalidis, K. V. (2017). Towards a comprehensive business process optimization

framework. Proceedings 2017 IEEE 19th Conference on Business Informatics (CBI) (pp.
pp. 129–134). Thessaloniki, Greece: IEEE.

Georgiadis., G. P. (2021). Food and Bioproducts Processing. New Innovations in Food

Packaging in Food Industry, 204-221.

Jhumar, S. (2023). Parametric Optimization of Turning Process Using Taguchi Method and
ANOVA Analysis. International Journal of Advances in Engineering & Technology
(IJAET), 289-301.

Kalpande, S. a. (2003). Reliability analysis and hypothesis testing of critical success factors of
total productive maintenance. Journal of Operations Management, Vol. 26 No. 4, pp.
468-489.
40
KC, B., Faruk, O., Agnelli et al. (2015). Optimization of injection molding parameters using
Taguchi method for reducing shrinkage. industrial engineering journal.

Khedkar, D. K. et al. (2020). New Innovations in Food Packaging in Food Industry. Emerging
Technologies in Food Science.

Kothari, C. (2004). research methodology methods and techniques. new age international .

Kothari, C. R. (2004). Research Methodology, Methods and Techniques. international

conference. New Delhi: New Age International Publishers.

L. Norros, E. K. (2003). Human-technology interaction research and design. VTT roadmap.

Luis M sanchaz, R. n. (2020). A review of agile manufacturing systems. International Journal of

Production Research, vol. 39, no. 16.

LUIS M. SANCHEZ, R. N. (2020). A review of agile manufacturing systems. International

Journal of Production Research, vol. 39, no. 16.

Maheshwar. (2006). Financial and Management accounting. New Delhi: MDS Books. New
Delhi: MDS Books.

Meidute-kavaliauskiene. (2023). OPTIMISATION MODEL OF PRODUCTION PROCESSES

AS AN EXAMPLE. 13th International Scientific Conference. Vilnius, Lithuania: Faculty
of Business Management, Vilnius Gediminas Technical University,.

MEIDUTĖ-KAVALIAUSKIENĖ et al. (2023). OPTIMISATION MODEL OF PRODUCTION

PROCESSES AS AN EXAMPLE. 13th International Scientific Conference. Vilnius,
Lithuania: Faculty of Business Management, Vilnius Gediminas Technical University, .

MM. Islam, M. H. (2015). Line balancing for improving apparel production by operator skill
matrix. International Journal of Science, Technology and Society.

Montgomery, D. (2000). Design and Analysis of Experiments. New York: Wiley.

mulat, A. ( 2018). EVALUATION AND OPTIMIZATION OF PROCESS PARAMETERS

USING SIMULATION TAGICHI APPROACH. DSpace Institution's institutional
repository, 19.

N, D. (2023). Assessment of environmental performance and development sustainability of

systems product. Ecological Engineering &Environmental Technology.

Noorwali, A. (2013). Apply Lean and Taguchi in different level of variability of food. The
Manufacturing Engineering Society International Conference, MESIC 2013 (pp. 728 –
734). Elsevier Ltd.

41
Nwafor S.C., O. S. (2019). Taguchi optimization of cast geometries for A356/organic particulate
aluminium alloy composites using a two-phase casting process. Journal of Applied
Science and Process Engineering, Vol. 6.

Oji, B. &. (2021). Optimisation of bottling process using “hard” total quality management
elements. The TQM Journal, Vol. 33 No. 2, pp. 473-502.

Raji, A. a. (2019). “Enhancement of maintenance downtime using Poisson motivated Taguchi

optimisation method”. Al-Nahrain Journal for Engineering Sciences,, Vol. 22 No. 4,.

Ridwan Majekodunmi Adegoke, S. A. (June 2022). Analyzing The Effect of Aspect Ratios on
Optimal Parametric Settings Using Taguchi, Taguchi-Pareto, and Taguchi-ABC method.
International Journal of Industrial Engineering and Engineering Management (IJIEEM),
Page 32, 31.

Rosa et al. (2022). An overview on Taguchi method. International Journal of Engineering and
Mathematical Sciences. international journal.

Shang, J. (2020). Operational design of a supply chain system using the Taguchi method,
response surface methodology, simulation, and optimization. International Journal of
Production Research.

Sukthomya W. et al . (2022). Taguchi experimental design for manufacturing process

optimisation using historical data and a neural network process model. International
Journal of Quality & Reliability Management.

Tadiyos. (2018). Productivity Analysis And Improvement In Ethiopian Metal Manufacturing

Industries incase of Yesu Metals Manufacturing. ethiopian journal.

Tarng, Y. (2000). The use of fuzzy logic in the Taguchi method for the optimisation of the
submerged arc welding process. The International Journal of Advanced Manufacturing
Technology, 688-694.

Tianxia. (2012). Application of PLC for Arranging Bottle in Beer Filling Production Line. 24th
Chinese Control and Decision Conference (CCDC), (pp. 1215 – 1220). Chine.

Tianxia Z et al. ( 1215 – 1220.). Application of PLC for Arranging Bottle in Beer Filling
Production Line. 24th Chinese Control and Decision Conference (CCDC), (p. 2012).
chiena.

Tianxia Z et al. (2012). Application of PLC for Arranging Bottle in Beer Filling Production Line.
24th Chinese Control and Decision Conference (CCDC), (pp. 1215 – 1220.). chiena.

Wahyu Adhi Yatma et al. (2018). Reduction Bottle Cost of Milkuat LAB 70 ml Using Optimal.
Journal of Applied Research on Industrial, 223–238.

42
Y Kuo, T. Y. et al. (2008). a grey-based Taguchi method for optimizing multi-response
simulation problemsEngineering Optimization. Engineering Optimization.

Z, T. (n.d.). Application of PLC for Arranging Bottle in Beer Filling Production Line. 24th
Chinese Control and Decision Conference (CCDC).

Zikmund, W.G., Babin et al . (2010). Business Research Methods. Mason: South Western
Cengage Learning.

References

A.Floudas, C. (2015). Mixed Integer Linear Programming in Process Scheduling: Modeling,

Algorithms, and Applications. . Annals of Operations Research, 131-162.

Abhishek. (2013). Simulation Modelling Practice and Theory. Journal of minimally, 31-40.

Achraf Ammar, B. S. ( 2019). performance optimization of a flexible manufacturing system

using simulation. Ppuplisher online, Volume 95, Issue 11.

Basso. (2019). Optimizing perioperative outcomes with selective bowel resection following an
algorithm based on preoperative imaging for bowel endometriosis. Journal of minimally.

Belavendram, N. (1995). Quality by Design Taguchi Techniques for Industrial Experimentation.

internatinal journal.

Belavendram, N. (1995). Quality by Design Taguchi Techniques for Industrial Experimentation.

internatinal journal.

Belavendram, N. (1995). Quality by Design Taguchi Techniques for Industrial Experimentation.

international journal.

Belavendram, N. (n.d.). Quality by Design Taguchi Techniques for Industrial Experimentation.

international journal for enginer.

Bozza. (2022). Current regulatory and market frameworks in green cosmetics. The role of
certification. Sustainable Chemistry and Pharmacy.

C Afteni, G. F. (2017). A review on optimization of manufacturing process performance.

International Journal of Modeling and Optimization.

43
Chakraborty K., et al. (2015). controlling Process of a Bottling Plant using PLC and SCADA.
Indonesian Journal of Electrical Engineering and Informatics (IJEEI), 39~44.

D. Annicchiarico, J. A. (2014). Review of factors that affect shrinkage of molded part in

injection molding. Materials and Manufacturing Processes.

D. Sabadka, V. M. (2017). The use of lean manufacturing techniques – SMED analysis to

optimization of the production process. Advances in Science and Technology Research
Journal.

Daneshjo. (2023). Assessment of environmental performance and development sustainability of

systems product. Ecological Engineering &Environmental Technology.

Daneshjo, N. (2023). Assessment of environmental performance and development sustainability

of systems product. Ecological Engineering &Environmental Technology.

Daneshjo, N. (2023). Assessment of environmental performance and development sustainability

of systems product. Ecological Engineering &Environmental Technology. international
journal.

Daneshjo, N. (2023). Assessment of environmental performance and development sustainability

of systems product. Ecological Engineering &Environmental Technology.

Daneshjo, N. (2023). Assessment of environmental performance and development sustainability

of systems product. Ecological Engineering &Environmental Technology.

Daneshjo, N. (n.d.). Assessment of environmental performance and development sustainability

of systems product. Ecological Engineering &Environmental Technology.

Darvishi, F. &. (2019). Optimization of an industrial medium from molasses for bioethanol
production using the Taguchi statistical experimental-design method. International
Journal of Quality & .

Derbe. (2023). Impact of dairy production on smallholder households Food Security.

International Journal of industrial.

Dr. Meghna Chhabra, D. R. (2018). Productivity in manufacturing industries. International

Journal of Innovative Science and Research Technology.

eilon. (2019). productivity improvement of garnment. jobshop scheduling, 5-7.

Fisher, R. (1970). s wide-ranging account of methods of statistical inference, and also for the
wealth of applications made to biology. Statistical methods for research workers.

Franceschini, G. et al. (2008). Model-based design of experiments for parameter precision: state
of the art. Chem. Eng. Sci.

44
Franson, D. (2016). Encyclopaedia of occupational health and safety. Beverage Industry .

G. Tsakalidis, K. V. (2017). Towards a comprehensive business process optimization

framework. Proceedings 2017 IEEE 19th Conference on Business Informatics (CBI) (pp.
pp. 129–134). Thessaloniki, Greece: IEEE.

Georgiadis., G. P. (2021). Food and Bioproducts Processing. New Innovations in Food

Packaging in Food Industry, 204-221.

Kalpande, S. a. (2003). Reliability analysis and hypothesis testing of critical success factors of
total productive maintenance. Journal of Operations Management, Vol. 26 No. 4, pp.
468-489.

KC, B., Faruk, O., Agnelli et al. (2015). Optimization of injection molding parameters using
Taguchi method for reducing shrinkage. industrial engineering journal.

Khedkar, D. K. et al. (2020). New Innovations in Food Packaging in Food Industry. Emerging
Technologies in Food Science.

Kothari, C. R. (2004). Research Methodology, Methods and Techniques. international

conference. New Delhi: New Age International Publishers.

L. Norros, E. K. (2003). Human-technology interaction research and design. VTT roadmap.

Luis M sanchaz, R. n. (2020). A review of agile manufacturing systems. International Journal of

Production Research, vol. 39, no. 16.

LUIS M. SANCHEZ, R. N. (2020). A review of agile manufacturing systems. International

Journal of Production Research, vol. 39, no. 16.

Maheshwar. (2006). Financial and Management accounting. New Delhi: MDS Books. New
Delhi: MDS Books.

Meidute-kavaliauskiene. (2023). OPTIMISATION MODEL OF PRODUCTION PROCESSES

AS AN EXAMPLE. 13th International Scientific Conference. Vilnius, Lithuania: Faculty
of Business Management, Vilnius Gediminas Technical University,.

MEIDUTĖ-KAVALIAUSKIENĖ et al. (2023). OPTIMISATION MODEL OF PRODUCTION

PROCESSES AS AN EXAMPLE. 13th International Scientific Conference. Vilnius,
Lithuania: Faculty of Business Management, Vilnius Gediminas Technical University, .

MM. Islam, M. H. (2015). Line balancing for improving apparel production by operator skill
matrix. International Journal of Science, Technology and Society.

Montgomery, D. (2000). Design and Analysis of Experiments. New York: Wiley.

45
mulat, A. ( 2018). EVALUATION AND OPTIMIZATION OF PROCESS PARAMETERS
USING SIMULATION TAGICHI APPROACH. DSpace Institution's institutional
repository, 19.

N, D. (2023). Assessment of environmental performance and development sustainability of

systems product. Ecological Engineering &Environmental Technology.

Noorwali, A. (2013). Apply Lean and Taguchi in different level of variability of food. The
Manufacturing Engineering Society International Conference, MESIC 2013 (pp. 728 –
734). Elsevier Ltd.

Nwafor S.C., O. S. (2019). Taguchi optimization of cast geometries for A356/organic particulate
aluminium alloy composites using a two-phase casting process. Journal of Applied
Science and Process Engineering, Vol. 6.

Oji, B. &. (2021). Optimisation of bottling process using “hard” total quality management
elements. The TQM Journal, Vol. 33 No. 2, pp. 473-502.

Ridwan Majekodunmi Adegoke, S. A. (June 2022). Analyzing The Effect of Aspect Ratios on
Optimal Parametric Settings Using Taguchi, Taguchi-Pareto, and Taguchi-ABC method.
International Journal of Industrial Engineering and Engineering Management (IJIEEM),
Page 32, 31.

Shang, J. (2020). Operational design of a supply chain system using the Taguchi method,
response surface methodology, simulation, and optimization. International Journal of
Production Research.

Tadiyos. (2018). Productivity Analysis And Improvement In Ethiopian Metal Manufacturing

Industries incase of Yesu Metals Manufacturing. ethiopian journal.

Tarng, Y. (2000). The use of fuzzy logic in the Taguchi method for the optimisation of the
submerged arc welding process. The International Journal of Advanced Manufacturing
Technology, 688-694.

Tianxia. (2012). Application of PLC for Arranging Bottle in Beer Filling Production Line. 24th
Chinese Control and Decision Conference (CCDC), (pp. 1215 – 1220). Chine.

Tianxia Z et al. ( 1215 – 1220.). Application of PLC for Arranging Bottle in Beer Filling
Production Line. 24th Chinese Control and Decision Conference (CCDC), (p. 2012).
chiena.

Tianxia Z et al. (2012). Application of PLC for Arranging Bottle in Beer Filling Production Line.
24th Chinese Control and Decision Conference (CCDC), (pp. 1215 – 1220.). chiena.

Wahyu Adhi Yatma et al. (2018). Reduction Bottle Cost of Milkuat LAB 70 ml Using Optimal.
Journal of Applied Research on Industrial, 223–238.
46
Y Kuo, T. Y. et al. (2008). a grey-based Taguchi method for optimizing multi-response
simulation problemsEngineering Optimization. Engineering Optimization.

Z, T. (n.d.). Application of PLC for Arranging Bottle in Beer Filling Production Line. 24th
Chinese Control and Decision Conference (CCDC).

47