Hierarchizing the Product Characteristics of Industrial Plain Sewing Machine for Making Best Purchase Decision

. Tousands of readymade garment industries are available in Bangladesh, where thousands of sewing machines are being used. Te productivity of these sewing machines mostly depends on workers but more or less depends on the machine too. Suitable machines should be sourced to increase worker efciency and production. Purchasing these machines needs some information or criteria ranking for choosing the best or most suitable machine. Tis study focuses on identifying and prioritizing those characteristics that fulfl the workers’ (users’) requirements and beneft the industry’s overall income. Tis study intends to investigate the application of the fuzzy Kano model and AHP (analytical hierarchy process) to translate consumer expectations and design into quality products. Customer needs and technical attributes are assessed and prioritized using the AHP. Ten, the optional attributes were prioritized by using the Kano model. RMG industries and their purchasing departments can use these prioritized qualities charts to help them choose a sewing machine. Manufacturers of sewing machines can utilize this information to improve their sewing machine qualities, such as if they need to add or remove any attributes that the consumer requires or does not require.


Introduction
A sewing machine is a mechanical device that uses thread to sew cloth and other soft materials together. During the frst industrial revolution, sewing machines were developed to decrease the amount of manual sewing work done in the apparel industry. Since then, the apparel industry has seen a signifcant increase in efciency and output. Bangladesh's readymade garment (RMG) industry contributes signifcantly to the country's overall economic growth. Tis industry employs approximately two million people, with women accounting for 80% of those employed [1]. It is also important to keep in mind that the earnings from this industry account for more than 80% of Bangladesh's total foreign exchange earnings [2]. A good sewing machine makes the work easier, safer, and comfortable and is therefore more productive for the workers. A good sewing machine also mitigates the chances of musculoskeletal disorders (MSDs) among workers [3]. Ten, how could we choose the "best sewing machine" among all the brands and features? Tat depends on the industry labors' skill level and budget. Before purchasing a sewing machine, the industry must evaluate its demand and skills, considering how people might use the machine in the future as their abilities or technology improves. Day by day, fast-changing technology in this feld and diversifed and rapidly growing competition among marketers drive businesses to seek simple, long-term solutions. As a result, optimizing product characteristics within the context of diferent pricing has become the company's only means towards becoming proftable. Tat is why industry purchasers need proper guidelines and identifcation criteria for selecting sewing machines that will be both productive and benefcial to garment workers.
From the perspective of the sewing machine manufacturers, in this competitive market environment, end users require products with lower prices, high quality, and ontime services. In addition, companies should operate based on consumer needs and attitudes. Terefore, they have to reduce their manufacturing costs. To reduce cost, they need to optimize product characteristics by focusing on consumer demands and their level of satisfaction. Prioritization enables a business to concentrate its people and fnancial resources to provide its customers with the highest quality products. Instead of aimless improvements, targeted eforts are much more aligned to serve maximum value to the customers. For fnding the priority of attributes, the question is "how can we then determine what is most important to the customer?" In this case, a multi-criteria decision-making (MCDM) model will help rank the criteria and assist customers in precisely understanding their priorities.
Tere are several multi-criteria decision-making methods, including AHP (analytic hierarchy process), TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) [4], ELECTRE (Elimination Et Choix Traduisant la Réalité) [5], BWM (Best-Worst Method) [6], DEMATEL (DEcision MAking Trial and Evaluation Laboratory) [7], MAUT (Multiple Attribute Utility Teory) [8], PROMETHEE (Preference Ranking Organization Method for Enrichment Evaluation) [9,10], GRA (Grey Relational Analysis) [11], and SWARA (Stepwise Weight Assessment Ratio Analysis) [12], which have been developed during the previous two decades and are being used practically in every area throughout the world to make decisions. Te most recently developed MCDM method, Basic Uncertain Linguistic Information (IBULI), integrates consistent measures of decision experts that can efectively quantify subjective assessment information provided under uncertainty [13]. In terms of approach, BWM, DEMATEL, and MAUT preference ranking methods are similar to AHP, with a distinction in scale. From an application viewpoint, DEMATEL is ideal for assigning values to infuential aspects based on supplied criteria [14]. MAUT employs the utility function, which tries to replace the value associated with a criterion for a decision maker's level of satisfaction. As TOPSIS is based on distance, it necessitates the defnition of an ideal alternative, which may not be attainable in this scenario. Methods based primarily on statistical data, such as PROMETHEE and ELECTRE, are utilized for complex tasks. Te key advantage of the most advanced AHP conceptualization is that it is easy to implement and allows for hierarchical problem modelling and the ability to make vocal judgments and validate consistency [15]. As a result, the authors are easily persuaded to choose the AHP method over others to achieve the research objectives.
To improve buying decisions' efectiveness, technical and non-technical data, such as customer requirements, must be examined. Rather than market concepts, focusing on potential customers is an excellent technique for identifying their demands. For this purpose, Dr. Kano's two-dimensional quality model is an excellent instrument. Dr. Kano classifed quality elements by using functional and dysfunctional questionnaires in an assessment table. Tis is a fundamental tool for a two-dimensional quality model used in numerous research studies. Te Kano model describes the relationship between user satisfaction and the quality of a product or service. Kano's major advantage is identifying customer needs that may be formed as a customer-specifed quality. Tese inquiries will provide new ideas and a wonderful method to meet customer requirements. Kano models can be applied in every discipline of engineering, science, or social science where the perceptions of customers or users are more important for selecting criteria when designing a product. Te Kano model is widely used for identifying customer requirements. No other method has been as efective as the Kano model in terms of serving its purpose.
Te objective of this study was to determine the hierarchy of plain sewing machine attributes (both core and optional) to make the machine purchase decision optimal. Te authors conducted direct interviews, investigations, and analytical techniques to discover customer needs and product requirements. Te following section contains a literature review of certain studies undertaken to determine the viability of using AHP and Kano for attribute selection, ranking, and best decision making. To the best of the authors' knowledge, there is no literature that addresses the use of AHP and Kano models for fnding, hierarchizing, or developing sewing machine attributes. Te authors sought to solve a fairly typical problem (best sewing machine selection) in RMGs using AHP, which the decision maker accepted and appreciated. Te AHP method was implemented to hierarchize the core attributes and fnd their respective weights. Te fuzzy Kano model was used to determine and hierarchize the optional sewing machine attributes. Sewing machines are used for the same purpose all over the world. As a result, the fndings of this study will be a valuable source of information for both industrialists and manufacturers worldwide. When purchasing sewing machines, the purchaser can observe the hierarchy of product characteristics and customer needs. In contrast, the manufacturer can observe the optional attributes they should focus on or develop machine characteristics for their sewing machines. Following the literature review, the remainder of this paper is organized as follows. Te methodology section explains how AHP and the fuzzy Kano model were implemented step by step. Te results of using the AHP and fuzzy Kano models to determine the best sewing machine were shown in the results and discussion section, and the implications of the fndings were thoroughly discussed. Te conclusion section summarizes and explains how the model was completed in light of our achieved goal. Lastly, the authors speculated on the study's limitations and future scopes of research.

Literature Review
2.1. Literature Review on AHP Model. Te analytic hierarchy process (AHP), a sophisticated computational approach pioneered by Saaty, has been viewed as a major and adaptable modelling tool and has been utilized to solve a wide range of research difculties [16]. Te most signifcant impact of the AHP method has been in resolving supply chain issues. Over the last four decades, researchers working on a variety of AHP techniques have identifed and addressed various complex supply chain problems. Te researchers' common interest was solving supplier ranking and selection problems, which led to positive implications. 2 Mathematical Problems in Engineering Tahriri et al. (2008) implemented AHP for best supplier selection in a steel manufacturing company, deciding how to allocate business and determining where development efort should be applied [17]. Marufuzzaman et al. (2009) implemented AHP to solve the supplier selection in an apparel manufacturing farm in Bangladesh [18]. Wang (2009) implemented AHP to select the right supplier of maintenance and repair parts in the industry [19]. Chan and Chan (2010) implemented AHP to solve the supplier selection problem in the apparel industry with a case example [6]. Ozkan et al. (2011) implemented AHP for the best supplier selection for computer and printer purchasing for the General Directorate of Land Registry [20]. Rouyendegh and Ekran (2012) implemented AHP for supplier evaluation and a selection for a university procurement department [21]. Agarwal et al. (2014) implemented AHP for supplier evaluation and selection in the baby toy manufacturing industry [22]. Azadi et al. (2014) implemented AHP in a car glass company in Turkey to solve the supplier selection problem [23]. Sharma and Rawani (2016) implemented AHP for the selection of the best supplier in the cement industry in India, considering the relevant Green Corporate Social Responsibility (CSR) and Safety criteria [24]. Yadav and Sharma (2016) implemented AHP with a weight cum rating method for selecting the best supplier in a leading car and truck manufacturer company [25]. Environmental challenges have risen in recent decades, and maintaining environmental sustainability has been a primary priority. Researchers have also looked into this feld to uncover the most important ideas that might lead any supply chain system to go "Green." For instance, Govindan et al. (2014) applied AHP to identify barriers to the implementation of green supply chain management (Green SCM) based on procurement efectiveness [26]. Wang et al. (2016) implemented AHP to investigate the pressures for Green Supply Chain Management (GSCM) adoption and to rank the pressures in the mining and mineral industry context [27]. Luthra et al. (2016) implemented AHP to identify and evaluate barriers in implementing a sustainable supply chain with a case example of the Indian plastic industry [28]. Luthra et al. (2013) implemented AHP for ranking of strategies to implement green supply chain management (GSCM) in the Indian manufacturing industry [10]. AHP can also select a thirdparty logistics (3PL) provider in a coordinated supply chain [29].
In the AHP method, the output is ranked using simple scales developed from pairwise comparisons [30][31][32]. Terefore, priorities for some complex models can be simply determined by studying their ranks [33], making it the simplest ranking method to apply in all the possible sectors. AHP is a methodologically sound and user-friendly strategy for single choice decision making [34] to multiple choice decision making [35], for performing resource allocation [36], benchmarking [36][37][38], and quality management [39][40][41]. Te AHP method has been used in various applications to handle the problem of hierarchizing attributes to pick the best service. Many researchers have introduced fuzzy logic into the comparison for this purpose, which provides fuzzy numbers as a pairwise comparison scale.
Tousands of studies have been conducted in which a fuzzy model of AHP has been established, inferred, and claimed to be more valid. However, Saaty (2006), the architect of AHP, claimed that the pairwise judgments are sufciently fuzzy and that the fuzzy technique does not outperform the old one [42]. Te 1-9 scale adopted in traditional AHP is not obsolete; each number represents the judgment as important, moderately important, or more important, making the fuzzy approach superfuous [43]. As a result, the authors approached this study by adhering to the traditional AHP method, inspired by its simplicity and wide range of application records in decision making. Table 1 summarizes notable studies from the previous 15 years. Tese studies have used AHP (without fuzzifcation) as the primary methodology for determining the optimal solution by hierarchizing key attributes.
Te number of attributes used to conform to the criteria comparison matrix in all of this literature ( Table 1) was no more than ten (except in 3 out of 27 cases). It should be noted that consistency is determined primarily by the selection of appropriate attributes rather than the number of attributes used (independent of their quantity). Te criteria comparison matrix in AHP comprises expert opinions and choices rated on Saaty's scale, which ranges from 1 to 9. All of these viewpoints can be distilled into a single choice based on the ratings of all experts, or the mean of all expert ratings can be used to create a pairwise comparison matrix [18]. Among all the references shown in Table 1, researchers involved experts for criteria comparison from various felds. In 25 out of 27 studies, the number of experts who participated in decision making is less than 20. So, the recommended number of experts for an appropriate outcome should be no more than 20 [57]. Involving more experienced and knowledgeable people in that feld could yield impressive results. Te comparison matrix should include input from professionals in several sectors, such as purchase, maintenance, academic, research and development professionals, and also other people involved in purchasing.

Literature Review on Kano Model.
Tere are numerous instances where the Kano model has been successfully implemented. It was used to develop product features by obtaining new ideas from consumers, fnding ergonomic standards, fnding the best attributes of community cultural products, fnding product quality attributes, and so on ( Table 2). Te Kano model was efectively applied to these engineering options. Surprisingly, the Kano model has also been employed in non-engineering applications to determine user satisfaction. Education, healthcare, and tourism may all be good examples ( Table 2). Te nature of the users/ consumers and the area of use of that specifc product or service determine Kano questionnaire sampling. Tere is no established sampling direction that can ensure the validity of the sample use in the Kano model.
Te Kano model with fuzzy approaches has also been widely used to solve diferent issues. In terms of how the attributes deal with the fuzziness of human ideas, Lee and Huang nicely illustrated the distinctions between the two Mathematical Problems in Engineering types of surveys [72]. Te fuzzy domain is concerned with mathematical and statistical approaches, making it more efective in the research. Te traditional Kano survey requires individuals to make a single decision about a certain feature, ignoring the individual's uniqueness and uncertainty. When it comes to product design when gaining customer satisfaction is the major goal, a single point of view does not produce the best results. Te fuzzy Kano concept encourages people to express themselves in fve distinct ways and with fve alternative options. For example, survey respondents may select one of fve options [72], which include "I like it that way," "It must be that way," "I am neutral," "I can live with it that way," and "I dislike it that way." Te survey becomes more refective of respondents' opinions than the traditional one. Many researchers have addressed the diferences between the classical and fuzzy Kano theories, produced new fuzzy concepts, and emphasized the need to incorporate the fuzzy concept into implementation.
Fuzzy Kano approaches were accepted and implemented by researchers in product development [73][74][75], the manufacturing industry [76], the renewable industry [77], and the supply chain [78], and all researchers agreed that the results of their implementations were more acceptable than those of traditional Kano models. For instance, Jain et al. implemented the fuzzy Kano method to identify customer attributes in an Indian iron and steel plant industry and then incorporated the SWARA method to assign weight and rank them with the Complex Proportional Assessment of Alternatives (COPRAS) method for supplier selection [79]. Jain and Singh used the fuzzy Kano method to identify the "must-be" attributes among 36 criteria and then performed supplier selection using the SWARA and Range of Values (ROV) methods [80]. Kilaparthi and Sambana integrated the Kano model with the VIKOR analysis, where the Kano model provided weightage for the supplier selection attributes, and the VIKOR resolved the uncertainty while selecting the best supplier of a forging frm in India [81]. Kuo et al. developed a supplier hierarchy in the Indonesian palm oil industry by combining the fuzzy AHP and Kano approaches. Te Kano model was used to fnd the most important sustainability criteria, and the AHP model was used to fnd the weight of each criterion and create a ranking [82]. Te Kano model, particularly its fuzzy approach, can also provide an efective solution for managing customer service logistics [83]. Referring these examples, introducing the fuzzy Kano approach to determining optional sewing machine attributes would serve to achieve half of our research's goal.
Customer needs may vary across categories, and today's "must-be" features will become core elements in the future. Te Kano model is used for more than just identifying customer needs. It also implies the signifcance of the characteristics in fve distinct categories. Te frst priority should be to deliver what is expected (must-be needs); the Combined AHP and multi-expression programming method 5 No second priority should be to deliver what is specifed (onedimensional needs), and the last priority should be to provide attractive needs [84]. Identifcation of the sewing machine optional attributes using the Kano model will also result in a hierarchy of those identifying attributes, which should be the primary consideration for product developers to seek and focus on. Every sewing machine has some features in common that are widely used all over the world. However, by adding some optional criteria to these globally launched products, they may be able to satisfy their customers more than before. Te Kano model was used in this study to determine which extra or optional features customers expect in current sewing machines and to what extent they do not. To identify optional features, consumers of sewing machines must be interviewed, which cannot be represented through simple yes/no responses. For this purpose, collecting, recording, and calculating information using fuzzy logic should be more feasible.

Defning Key Consumers.
Te relationship between product and service capabilities and consumer or user needs must be examined during the optimization phase. To make these judgments valid, it is critical to defne clear and consistent defnitions as well as the most important underlying principles addressing the consumer or user. Te frst step in identifying the key consumer is identifying the sets of all potential customers. Following the identifcation of many client segments, the second step focuses on the primary clients. Once the user groups have been determined, selecting the consumers could be easier. Sewing machines are available as both home and industrial appliances. However, our priority was on sewing machine industry appliances, where thousands of sewing machines are used in the garment industry. Sewing machine operators and purchasers are the consumers here. For this Improving the ergonomics and user needs of adolescent workstation design in a school workshop using an integrated Kano and QFD approach No 255 Hwang et al. [62] Identifying the relationship between user emotional experiences and satisfaction with community culture products, by using the Kano model and Kano's regression method

No 62
Salahuddin and Lee [63] Identifying the key quality features of wearable technology embedded products that have the most infuence on consumer satisfaction Matı´as-Guiu et al. [68] Te Kano methodology was used to assess the expectations of patients receiving symptomatic migraine treatment and the neurologists treating them

No 204
Huang [69] Improving mobile healthcare service quality by implementing the Kano model and the IPA (importance performance analysis) method by confrming service quality attributes for priority improvement No 300

Tourism industry
Yang et al. [70] Proposes a strategic pricing model for a 5-star hotel in Taiwan that  can provide customers with increased value by charging less while  maintaining reasonable features   No  115 Zobnina and Rozhkov [71] Identifying the factors that infuence tourist satisfaction in the context of a Russian hotel No 1238 Mathematical Problems in Engineering reason, our primary goal was to interview consumers (workers and supervisors) and purchasers in the RMG industry.

Identifying Core and Optional Attributes of the Sewing
Machine. Weber et al. examined 74 articles on supplier selection from 1966 to 1990 and identifed 23 common attributes used as vendor selection criteria in those studies [85]. Quality, delivery, performance, warranty, capacity, and price were the most important selection criteria among all 23 attributes. In this study, the sewing machine characteristics were separated into two categories. Tese are (i) core attributes and (ii) optional attributes. Te core characteristics are common among sewing machines of all brands and types. Te core attribute data mainly consisted of technical information such as motor speed, warranty period, expenses, noise level, efciency, and so on. An industrial visit was undertaken to a reputed readymade garment industry in Gazipur, Bangladesh. It is a woven cloth manufacturing industry that produces both tops and bottoms. Te purpose of the industrial visit was to learn about the core and optional attributes of sewing machines and their use and expectations in the industry. Te best way to obtain this information is to interview technical personnel involved with these machines, such as those in maintenance, industrial engineering, and procurement team, who directly purchase or assist in purchasing and preserving technical data. A team of eight industrial engineers, one procurement ofcer, and two maintenance managers was formed to collect core attribute data and get comparable information to establish a pairwise comparison matrix. Te authors discovered some core attributes from discussion with the experts by analyzing machine designs, drawings, and functioning. Some attributes were added that are not connected to machine performance but are crucial variables to consider when purchasing, such as warranty period, pricing, and durability. Ten sewing machine core qualities were identifed in fve categories, as indicated in Figure 1. Almost all articles analyzed consider "delivery" as one of the most crucial selection criteria. Tis paper solely discusses hierarchizing machine attributes, not vendor selection. When it comes to supplier selection, "delivery" could be a critical consideration. Te optional attributes are those that people wish to have to help them with their current machines. Another discussion with experts was conducted to evaluate which unique characteristics of sewing machines they desired to have in order to increase user satisfaction. As the Kano method implies, sewing machine users (operators and supervisors) were surveyed to determine the optimal strategy for collecting optional attributes. Before conducting the survey, verbal consent was obtained from the factory's operations head. To collect the satisfaction level of users, 150 sewing machine operators and 30 supervisors were randomly selected for the survey from a total of approximately 4000 workers and 78 supervisors in that industry.

Implementing AHP for Ranking the Core Attributes.
Te authors employed the extraction method to determine the maximal eigenvalue/eigenvector and judgment matrix in AHP. Applying AHP for ranking core attributes, the following steps and equations were followed.
(i) Te frst step in implementing AHP is to create the numerical pairwise comparison matrix. Saaty proposed evaluating or comparing the various criteria on a scale of 1-9 [16] (Table 3). Te scales were used to rate the comparative importance. Construct a pairwise comparison matrix where weights are given based on comparative importance of the given criteria. In this matrix, where a ij is the element of i th row and j th column of the comparison matrix.
It is also the weight vector for the AHP priority matrix evaluation.
We now have the proper attribute hierarchy. However, another important aspect of developing AHP is guaranteeing its consistency. Te following steps should be followed to verify that the decisions in the pairwise comparison are consistent: (i) Multiply the complete comparison matrix [Aij] by the eigenvector column to get the Row matrix: (ii) Calculate the lambda value (λ) for each row by diving the Row matrix value by diving its respective eigenvector: (iv) Ten, the following equation can be used to get the Consistency Ratio (CR): Here, CI denotes Consistency Index� λ max − n/n − 1, where n � no. of criteria; and RI stands for the Random Index, and it can be found in Table 4.
If the calculated Consistency Ratio (CR) ≤ 0.1, then the criteria matrix is within the consistency limit.

Using the Kano Model to Optimize the Optional Attributes.
On the x-axis, Figure 2 shows the Kano model's quality element. Te quality element is ofered more to the right as the arrow moves to the right, and the quality element is ofered less to the left as the arrow moves to the left. Te yaxis shows the consumer's level of product satisfaction. Customers are satisfed when the arrow is pointing upward; when it is pointing downward, customers are dissatisfed. As a result, the following quality attributes can be classifed as "Kano customer need attributes." Te Kano model segregates the attributes of a product or service into fve separate categories. Te fve categories of Kano's quality attributes are as follows: As part of the initial phase, the actual users of this machine were surveyed. Te survey generated a few optional attributes. Te objective and purpose of this research were communicated to the participants before the commencement of the survey so that they could relate to the research purpose and confdently deliver their responses. A Kano   Higher importance than moderate 5 Strong importance 6 More strong importance 7 Very strong importance 8 Very, very strong importance 9 Extreme importance questionnaire (Table5) was prepared and provided to the participants for data collection. Te authors prepared a training session to help workers understand the terms, writing, and form-flling method. Te entire document was translated into Bengali for the workers' beneft. In collaboration with the industrial engineering team, the authors directed all participants to provide ratings based on their experiences and perceptions. A total of 180 employees (150 machine operators and 30 supervisors) were randomly chosen for data collection and were provided a data collection form (Table 5). A total of 102 forms were received from respondents (response rate of 68%), all of which were fully completed and readable. After getting all of the data forms from the participants, the authors used the method developed by Lee and Huang [72], which is explained below. Let us defne two universal set of positive and negative responses U and V. P = {P1, P2,. . ., Pp} and N = {N1, N2,. . ., Nn} represents the sets of p and n linguistic variables on U and V, respectively, which together construct a p × n matrix of a two-dimensional quality model, and Fuzzy Sample {FSk, k = 1,2,. . ., r} is random fuzzy sample on U and V.
We assign linguistic variables P i and N j to normalized membership m (P) ki ( p i�1 m(P) ki � 1 and m (N) kj ( n j�1 m(N) kj � 1, and let S ij = r k�1 m(P) ki ⊗ m(N) kj and T h be the sum of S xy which (x, y) cell corresponds to the h th attribute in the evaluation matrix for each sample FS k.
Te fuzzy Kano's mode (FKM) of this quality element is the largest value of {T h } α, which is in the range of α signifcant classifcation level. Te dataset is referred to as multi-fuzzy mode if more than two sets of Kano's fuzzy quality attribute categorization have the same value of T h . A fnal score of equal indicates that M > O > A > I which has the most impact on the product [72].

Implementation of AHP and Fuzzy Kano Model
3.5.1. Implementing AHP for Hierarchizing Core Attributes. An expert team, comprising eight industrial engineers, one procurement ofcer, and two maintenance managers, was formed to provide comparative information on the attributes when assessing for the creation of a pairwise comparison matrix. Table 3 shows the ratings based on the priority after the ranking. Table 6 shows the criteria comparison matrix for all of the attributes, along with their ratings. Tese ratings were determined by experts agreeing on a single judgment that one measure is more or less essential than another. For example, when comparing machine durability with cost, the attribute "machine durability" was given more importance and rated fve since experts believe that cost should not be given higher priority at the expense of machine durability. Te cost function was graded in inverse to that rating, which became 0.2 (1/5).
Te frst and most crucial stage was accomplished, and the next step was to determine the n th root and related eigenvector weights, which will determine the importance hierarchy of these attributes. Te n th root value and eigenvector weights were determined using the formulae stated in the methodology section (Table 7).
Te attribute hierarchy is obtained. A step-by-step procedure detailed in the Methodology section was used to check the consistency. Te decision matrix's consistency was tested by calculating the Consistency Ratio (CR) ( Table 8). Te Row matrix was frst obtained by multiplying the comparison matrix by the eigenvector column. Te Row matrix was computed using the Excel function "MMULT." Te Lambda Value (λ) for each row was computed by dividing the Row matrix by the eigenvector for that row. By averaging all of the Lambda values, the Lambda Average (λ max ) was calculated to be 10.89. Te Consistency Index (CI) was then obtained, and the Consistency Ratio (CR) was found by using the Random Index (RI) value of 1.49 from Table 4. Ten, the Consistency Index (CI) was calculated: For calculating the Consistency Ratio (CR), the value of Random Index (RI) was found to be 1.49 (Table 4).
According to Saaty if the calculated Consistency Ratio (CR) < 0.1, then the comparisons are valid in the consistent limit. We may now specify the hierarchy of a sewing machine's core attributes based on their weights (eigenvector) as our obtained CR is 0.066, less than 0.1 [16].
Te hierarchy is shown below (top most attributes rank higher):

Using the Fuzzy Kano Model to Hierarchize Optional
Attributes. Following a thorough discussion and brainstorming session with the experts, 9 attributes were identifed that could add value to working with the sewing machine by increasing its efciency, making any task easier, eliminating extra time for searching or any unnecessary motion, or adding any ergonomic feature that is required for worker safety. Te following are the nine optional attributes: (i) Wiper facility (ii) Under machine thread trimmer facility   Functional answer Dysfunctional answer Folder attachment Functional answer Dysfunctional answer

Automatic needle feed
Functional answer Dysfunctional answer

Programmable
Functional answer Dysfunctional answer Knee lever facility Functional answer Dysfunctional answer

Button powering system
Functional answer Dysfunctional answer

Measuring scale along with the table
Functional answer Dysfunctional answer

Convenient drawer for storing sewing accessories
Functional answer Dysfunctional answer Table  7: Hierarchy matrix with relative weight (eigenvector). All sewing machines do not share these attributes. Based on identifying all optional attributes, the next step was to determine the level of satisfaction the worker would feel while working with this machine. Te questionnaire form (Table 5) was circulated and collected from participants, and the results were recorded on separate Excel sheets for each attribute. Each attribute was calculated separately. Te calculation of identifying the degree of worker satisfaction for the attribute "wiper facility" is shown in this article as a sample calculation, and the detailed calculation of other attributes was not shown because the formula and calculations for each attribute are the same. Table 9 shows the aggregated data sheet comprising the percentage responses for the attribute "wiper facility" from 10 participants out of 102. Te calculation for other attributes was the same. Tat is why the authors only showed the calculation for the frst attribute, "wiper facility," instead of presenting the calculation for the other eight attributes in this article.
After S is obtained, it contains a 5 × 5 matrix which is then compared with the two-dimensional attribute classifcation 5 × 5 table based on the Matzler and Hinterhuber [88] model, as shown in Table 10.
When matrix S and Table 10 are compared, it is possible to deduce that participant 1 has 0.60 of the attractive quality attributes and 0.4 of the indiferent quality attributes. Te summary of the satisfaction level for all 102 respondents was calculated using the same approach. Table 11 shows the satisfaction percentages for the 10 participants out of 102.
Te α-cut common consensus standard notion is utilized to obtain {T h } α to fnd a more accepting and suitable category. Te cutof value of α ≥ 0.4 was used. When the quality attribute level is equal to or greater than α, "1" will be displayed; when it is less than α, "0" will be displayed in their perspective values depicted in Table 12.
Te "wiper facility" quality attribute for the frst responder can be divided into two categories: attractive and indiferent (Table 12). In a bigger context, the sum of responses for each column (attribute) can be identifed as the highest level of satisfaction that the consumer seeks. Table 12 shows that while the attribute wiper facility was classifed as attractive 22 times, the majority (frequency 38) of the responses were in favour of its optional use. Te requirement for all quality attributes will be obtained by repeating the above processes for estimating the satisfaction level, as indicated in Table 13.

Discussion
Two of the most critical components of an AHP application are a clearly stated purpose and a set of criteria and alternatives [20]. According to Cheng and li, AHP has two key applications: frst, it is used to create a decision-making matrix by weighting and comparing a set of traits to the alternatives; second, it is used to decide from several options [33]. Te generation of comparison matrices is critical in AHP applications. Managers from diferent departments have diferent perspectives on the same issue based on the focus of their areas. When the whole weight of their decisions is added together, it is not easy to anticipate when the overall weight will stabilize. In this instance, decision makers who are directly or indirectly involved in purchase decisions must be particularly relevant. Te minimum sample size for a reasonably acceptable analysis is 30, and for AHP, it is likely to be considerably lower [25]. Having many experts undertake the comparison adds greater value [25]. However, the greater the number of experts is, the more difcult the decision is to make. A team of eight industrial engineers, one procurement ofcer, and two maintenance managers       Te higher the frequency in a category is, the more likely the attributes will meet that user satisfaction level.
developed the decision matrix. Industrial engineers provided data on the efciency, durability, and performance of the machines utilized in production. Procurement ofcers gave pricing and warranty information, while the maintenance manager ofered information on the machine's design, ergonomics, and the lubrication system available with the machines under consideration. To make their decisions in this study, the authors relied on real data and the purchase experience of the decision makers. Several research studies aggregated expert evaluations by taking the geometric mean of the experts' individual ratings [18].
Te major goal of this study was to weigh and rate the various sewing machine attributes to make the best machine choice possible. Te frst stage in implementing AHP is identifying and ranking the major and subcriteria relevant to the decision problem. Te criteria comparison matrix, created based on the expert's viewpoint, was used to hierarchize the 10 core attributes under fve groups. Te criteria comparison matrix's consistency was determined to be feasible (Table 8), and the matrix's rankings were assessed to be acceptable. Te AHP implementation outcomes show that the attribute "separate servo motor" was given the highest weighting (eigenvector of 0.24), indicating that it should be the most signifcant. In contrast, the criteria "warranty period" and "table lamp" were given the lowest priority (eigenvector of 0.02) ( Table 7). Pairwise comparisons of these ten attributes were also performed. Te pairwise comparison matrix established here is the result of expert decisions. When creating the comparison matrix, 45 direct and 45 reverse decisions were made. Tese choices were consistent throughout all comparisons, making the fnal judgment more accurate (Table 8). Tere is a wide range of plain sewing machines with varying features and pricing available on the market. When a purchaser intends to purchase a machine, he should frst confrm the machine's core attributes. To the client, the hierarchy created by applying AHP should be referred to as a priority list. He can prioritize which machine properties should be prioritized frst and which should be prioritized later. For instance, suppose the purchaser has two goals to achieve: he wants to acquire a sewing machine that is both durable and inexpensive. Because the machine's durability is ranked higher than its cost, he can make a priority decision by purchasing the more durable machine while sacrifcing a bit of cost, by making a simulation decision.
Te second portion of this study is concerned with categorizing the optional sewing machine features that the user desired. Some automated sewing machines allow all the operations to be programmed, although most machines do not. Some sewing machines have measurement scales attached to the machine bed, while others do not. A discussion with experts was held again to determine which unique characteristics ofered in a sewing machine satisfed the users' desires to have them in all of the machines they want to purchase. Tis is how these attributes were discovered to determine the extent to which these features satisfy the users. It also answers whether all sewing machines should have these functions. Ten, they must decide which features should be prioritized for product development. According to the hierarchy decision made by implementing the Kano model in this research, the "thread trimmer facility" and the "machine's button powering mechanism" are the 2 most important attributes that most users look for as "must-be" attributes (Table 13), which means that users no longer want these two attributes to be optional, but rather a "must-be" included feature in the sewing machines. Te absence of either of these two features in a machine will result in user dissatisfaction. Te four facilities or attributes such as "wiper facility," "programmable machine feature," "knee lever," and "measuring tape" as attachments with the machine were mostly found to be one-dimensional attributes, which means that providing these facilities with the machines will result in higher satisfaction and avoiding them may result in lower satisfaction. Te attractive attributes found were "folder attachment," "automatic need feed facility," and "drawer facility" for storing sewing equipment, which can produce higher user satisfaction. Still, in the case of higher costing decisions, these can be denied without generating dissatisfaction for the user. If a supplier wants to satisfy more clients and expand their target market, they should concentrate on these attributes from the Kano model.
Te weighting technique for the comparison matrix is based on a range of criteria, including personal experience, expertise, linguistic facts, and a judge's subjective assessment [89]. Te experts likely have a bias or inclination towards their previously observed decision. Tat is why it is critical to check the matrix's consistency. Te consistency of the criteria comparison matrix is dependent on the interviewers' ability to derive the optimal decision from the responses they obtain from those being interviewed. All the people who made decisions were part of the brainstorming process, not just the author. Te author asked them if their conclusion aligned with what the authors had obtained. Chen et al. addressed two major challenges in adopting MAGDM (multi-criteria group decision making): the frst is the information loss all through the decision-making process, and the second is how to scientifcally incorporate decision makers' perspectives into the MAGDM [90]. Te authors involved experts from various felds to create the decision matrix, allowing them to view the problem from multiple perspectives and contribute their diverse attitudes to make it more accurate. In this study, the authors formed a team of 11 experts from three diferent departments who are directly or indirectly related to the purchase of these machines (such as the procurement department) and their functionality (such as the industrial engineering and maintenance departments), which primarily eradicates the frst issue of information loss. Te authors attempted to aggregate all decisions made by all decision makers into a single decision to assign a single priority ranking through negotiation and discussion among themselves, which alleviated some of the challenges associated with adopting an irrelative perspective.
However, some aspects could be improved. When AHP is integrated with other methods deployed in the upstream or downstream parts of AHP for the same objective, it can provide excellent support to users (Table 1). Users can use AHP ranking matrix inputs by the DEA (data envelopment analysis) method to make it more efcient in application [45]. Trough focusing on individual observations, linear programming can also be utilized to maximize the performance of individual observations [45]. Te modifed Delphi approach can also serve the same purpose [35]. Te software can be developed to determine the most suitable alternative as well as to simplify iterations. In addition to building the comparison matrix and determining eigenvalues, the expert choice software can run a sensitivity analysis to see how the alternatives vary as the criterion's importance varies [54]. Quality function deployment (QFD) is a well-known tool for converting customer requirements (CRs) into engineering characteristics (ECs). QFD can be applied to the extracted customer attributes to uncover the voice of the customer whose supplier's attributes signifcantly impact attaining the defned objectives [91]. Since both the AHP and the Kano Model involve linguistic words, addressing some fuzzy linguistic term sets will examine the experts' complex language expressions to portray their emotions [92] appropriately. According to Chen et al. two major issues may arise when experts attempt to employ the language assessment paradigm [92]. Te frst is that the experts' opinions may be dubious. Tey may occasionally provide multiple linguistic terms to demonstrate their hesitancy during evaluation rather than just one linguistic term to indicate their preferences. Te second point is that when solving the MCDM problem, the process of creating decision matrices should be done in a single decision-making stage. However, in most circumstances, it is not practised or recognized.
Tis research might guide machine manufacturers and suppliers to fnd the best possible solution to satisfy their customers and users. Customers' perceptions and expectations fuctuate over the time. What users think now could be better product development advice to manufacturers and suppliers. Tis research may provide improved messages or suggestions to sewing machine manufacturers in this manner. Another issue with purchasing is that the process becomes more complicated when there are multiple decision makers. Te purchasing process should be made as simple as possible. Te decision to buy the best sewing machine is no longer difcult. It is now based on simple metrics and rankings that are easy to understand and do not require the buyer to be very technical.

Conclusion
Several strategies have been used to solve the supplier selection problem. It is now widely accepted that price alone should not be the primary consideration when making a purchase; other factors impact product performance more than price. When choosing a supplier, both customer and engineering factors are considered. AHP is perhaps the frst method for resolving the supplier selection problem and making the optimal purchasing decision. It is well known for its simplicity of use and wide range of applications. Te AHP technique is founded on three basic principles: model structure, alternative comparison, and priority setting. Tis research begins by determining the core characteristics of a sewing machine, which are then prioritized using the AHP method. Using the AHP and fuzzy Kano models, two categories of product attributes, core and optional, were discovered and hierarchized. Te core attributes were hierarchized using AHP. Te consistency approach was used to assess the consistency of the AHP process and the output result.
Te Kano model was used to identify if users expected more features from the sewing machines they intended to buy. Tousands of sewing machines are used across the readymade garment industries of Bangladesh. After infrastructure, most of an RMG manufacturing company's investment is earmarked for purchasing sewing machines. To purchase these machines, they must frst determine which machine will serve them best and be the most productive. Tey require some hierarchy or purchasing criteria. Before purchasing sewing machines, buyers can study the product characteristics and hierarchy of user needs. Te manufacturer can see the optional features that need to be prioritized or developed as machine characteristics for their sewing machine. Purchasers and users will proft from making the best purchase decisions, and manufacturers will beneft from product development insights from this research. Tis research will also contribute to creating a competitive sewing machine market by providing numerous features at a lower cost and analyzing those aspects. Tat is how the primary objectives of this research were met by benefting both suppliers and manufacturers.

Limitations and Future Scopes of This Study
Tere are always some risks associated with multi-criteria decision making, especially when multiple experts' viewpoints are involved. Both the AHP and Kano models require expert input in creating a decision matrix and user satisfaction scores. Since numerous elements might infuence a decision maker's behavior towards a product or service, in future studies, some risk factor calculation methods could be integrated with both methods to improve the accuracy of the results. Researchers may contact purchasers from various industries in diferent areas to see if our study's rating difers from theirs. Following the Kano model implementation procedures, questionnaires must be prepared and instructed for them to be flled out completely and accurately to gain the maximum possible benefts and outcomes. More optional attributes could be found if the sample size and demography were increased. In future research, the authors may look into any system that ranks suppliers based on how they want them to be ranked.

Data Availability
Te data used to support the fndings of this study are available from the corresponding author upon request.

Conflicts of Interest
Te authors declare that they have no conficts of interest.