Combining Kano Model and Adaptive Neuro Fuzzy Inference System

(ANFIS) to Optimize an Interaction Design of Tea Powder Packaging

INTRODUCTION

When first introduced to Kanos ideas, people are usually excited by them. Kanos ideas
about types of quality are often a revelation, and his method for sorting the features of a product
into various quality categories based on a questionnaire filled out by customers offers an
apparently straightforward process for gaining deep understanding of customer requirements.
However, as with so many tools, successful application of Kanos method requires skill and
experience.

LITERATURE REVIEW

Kanos Model
In planning a product or service, one makes a list of potential customer needs that the
product or service should perhaps try to satisfy. Going to see current and potential customers
(voice of the customer visit) is one good way to get ideas for what should be on the list of
potential customer requirements.
There are many instruments and methods designed to help organizations understand the
customer requirements. Among these methods, Kano Model is an instrument that has been
widely used to perceive the voice of customer in order to influence the customer satisfaction
[1]. Kano et al developed a model to classify the attributes of services or products which studies
the relationship between customer satisfaction and the function of product or services [2]. Kano
model classifies the customer requirements into 6 categories, based on the extent to which they
have been able to satisfy customers. The first class includes attractive quality elements and
includes some attributes of services or products the presence of which leads to customer
satisfaction; however the absence of them will not result in dissatisfaction. These are the
attributes of products or services that will distinguish the organization from its rivals. The
second class includes one-dimensional quality elements that have a positive and linear
relationship with customer satisfaction. So, the more these attributes are realized, the more
satisfied the customers will be and vice versa, the lesser these attributes are realized, the lesser
satisfied the customer will be. The third category is the must-be quality elements. These
attributes are the basic criteria for a product or service the lack of which will make the
customers quite dissatisfied. However, the fulfillment of these attributes will not increase the
customer satisfaction. Another class belongs to the indifferent quality elements the presence or
the absence of which will not influence the satisfaction or dissatisfaction of customers. The
fifth class in Kano Model is the reverse quality elements the presence of which causes the
dissatisfaction of customers and the absence of which causes their satisfaction. There is another
class in Kanos classification, other than these elements, that is called questionable elements.
This is when customer has not perceived the question or the information provided by the

question is not sufficient, or in other words, the customer is doubtful about this criterion [1, 2,
3, 4, 5]. The classification of quality attributes in Kano Model will be possible by Kanos twodimensional questionnaire. In this questionnaire, any one of these customer requirements is
analyzed by a pair of functional (positive) question and dysfunctional (negative) question.
There are 5 points or answers for every question: like, must-be, neutral, live with and dislike.
Then, the answers of customer are combined into two categories of functional and
dysfunctional questions and so the customer requirements are classified into 6 categories. In
Kano Model, an indicator that is marked as mode or the most frequent, based on customers
answers, is used as the final classification of customer requirements [1, 6]
Adaptive Neuro Fuzzy Inference System (ANFIS)
Fuzzy set theory is a perfect mean for modeling uncertainty (or imprecision) arising
from mental phenomena, which are neither random nor stochastic. Human beings are heavily
involved in the process of decision analysis. A rational approach toward decision making
should take into account a human subjectivity, rather than employing only objective probability
measures. This attitude towards the uncertainty of human behavior led to the study of a
relatively new decision analysis field: Fuzzy decision making [7]. Fuzzy inference systems are
one of the most applied and popular systems developed for fuzzy reasoning. Fuzzy reasoning,
also known as approximate reasoning, is an inference procedure that derives conclusions from
a set of fuzzy if-then rules and known facts. The fuzzy inference system is a popular computing
framework based on the concepts of fuzzy set theory, fuzzy if-then rules, and fuzzy reasoning.
It has been applied successfully in a wide range of science and engineering such as control,
function approximation, signal processing, simulation, data clustering and data mining and
decision support systems. In literature, we can find some other names such as fuzzy rule-based
system, fuzzy expert system, fuzzy model, fuzzy associative memory, fuzzy logic controller,
and simply (and ambiguously) fuzzy system. Fuzzy IF-THEN rules are the basic on fuzzy
inference systems [8]. A fuzzy rule is in the following form:
IF premise (antecedents), THEN conclusion (consequents).
In application cases, there is always more than a one rule. There has been developed
some inference methods such as Mamdani inference system [9], and Takagi and Sugeno
inference systems [10]. Conventional inference systems, in spite of their applicability, have
some shortages such as: there is no systematic procedure for rule generation and validation,
and also there is no systematic approach for optimizing the system parameters such as
membership function parameters to achieve more accurate solutions. Jang [11] developed a
network-based inference system which was capable to be trained from experiments. He called
that ANFIS, which is the abbreviation of adaptive neuro fuzzy inference system. He applied
least square estimator (LSE) and back-propagation (BP) to train the system.

REFERENCES

[1] Wang, Ting. and Ji, Ping. 2010. Understanding customer needs through quantitative analysis of Kanos
model. International J. Quality and Reliability Manage., 27(2): 173-184.
[2] Garibay. Cecilia, Gutirrez, Humberto. and Figueroa, Arturo. 2010. Evaluation of a Digital Library by Means
of Quality Function Deployment (QFD) and the Kano Model. The J. Academic Librarianship, 36(2): 125132.
[3] Rashid, Md Mamunur. Tamaki, Jun'ich. Ullah, Sharif. and Kubo, Akihiko. 2010. A Kano Model based
Computer System for Respondents Determination: Customer needs analysis for Product Development
Aspects. Management Ssienc and Engineering. 4(4): 70-74.
[4] Mirfakhrodini, S.H., M. Saleholia and R. Jamali, 2009. Quality Management Reengineering (Case Study:
Graduate Students of University of Yazd) Higher Education Planning and Research Quarterly, pp: 53.
[5] Chen, Ling-Hsiu. Kuo and Ying-Feng. 2011. Understanding e-learning service quality of a commercial bank
by using Kanos model. Journal of Total Quality Manage., 22(1): 99-116.
[6] Chen, Ling Hsiu. Lin and Hsiang Chih, 2007. Integrating Kanos Model into E-learning Satisfaction.
Proceedings of the 2007 IEEE IEEM.
[7] Kahraman, C., & Bykzkan, G. (2008). A combined fuzzy AHP and fuzzy goal programming approach
for effective six-sigma project selection. Journal of Multiple-Valued Logic and Soft Computing, 14(6).
[8] Jang, R., Sun, C. T., & Mizutani, E. (1996). Neuro-fuzzy and soft computing. Prentice Hall.
[9] Mamdani, E. H. (1977). Application of fuzzy logic to approximate reasoning using linguistic systems. Fuzzy
Sets and Systems, 26, 11821191.
[10] Takagi, T., & Sugeno, M. (1985). Fuzzy identification of systems and its application to modeling and control.
IEEE Transactions on Systems Man and Cybernetics, 15(1), 116132.
[11] Jang, R. (1993). ANFIS: Adaptive-network-based fuzzy inference systems. IEEE Transactions on Systems,
Man, and Cybernetics, 23, 665685.