Mining product maps for new product development


Mining product maps for new product development

Shu-Hsien Liao *, Chia-Lin Hsieh, Sui-Ping Huang

Department of Management Sciences and Decision Making, Tamkang University, No. 151, Yingjuan Road,

Danshuei Jen, Taipei 251, Taiwan, ROC

Abstract

Many enterprises have been devoting a significant portion of their budget to new product development (NPD) in order to distinguish
their products from those of their competitors, and to make them better fit the needs and wants of customers. Hence, businesses should
develop products that fulfill the customer demands, since this will increase the enterprise’s competitiveness and it is an essential criterion
to earning higher loyalties and profits. This paper presents the product map obtained from data mining results, which investigates the
relationships among customer demands, product characteristics, and transaction records, using the Apriori algorithm as a methodology
of association rules for data mining. The product map shows that different knowledge patterns and rules can be extracted from customers
to develop new cosmetic products and possible marketing solutions. Accordingly, this paper suggests that the cosmetics industry should
extract customer knowledge from the demand side and use this as a knowledge resource on its supply chain for new product
development.
� 2006 Elsevier Ltd. All rights reserved.

Keywords: New product development; Product map; Data mining; Association rules; Knowledge extraction

1. Introduction

It is well understood that new product development
(NPD) is critical for long-term firm performance and com-
petitive advantages. However, it is difficult to anticipate
what certain customers will value and when changes in
those preferences will emerge. A failure to anticipate these
changes can force suppliers into a reactionary mode, where
success is determined by how quickly one can respond to
new desires as they emerge. Because new products take a
substantial time to develop, suppliers should anticipate
changes in what customers will value as early as possible
(Flint, 2002). Therefore, a strong market orientation and
customer knowledge competence are vital to the success
of new products (Bonner, 2005). Most of the businesses
involved in the production chain such as the manufactur-
ers, suppliers and retailers are aware of the importance

and need for enterprises to acquire and share better knowl-
edge of their customers. But this is easier said than done
since customers’ preferences are concealed in the custom-
ers. It is available but not easily accessible, and there is lit-
tle possibility to explore the full volume of data that should
be collected for its potential value. Inefficient utilization
can render the data collected useless, causing databases
to become ‘data dumps’ (Keim, Pansea, Sipsa, & Northb,
2004). Therefore, how to effectively process and use cus-
tomer data is becoming increasingly important. This calls
for new techniques to help analyze, understand or even
visualize the huge amounts of stored data gathered from
business and scientific applications (Liao & Chen, 2004).
Among the new techniques developed, data mining is the
process of discovering significant knowledge, such as pat-
terns, associations, changes, anomalies and significant
structures, from large amounts of data stored in databases,
data warehouses, or other information repositories (Hui &
Jha, 2000; Keim et al., 2004). In addition, there are many
data mining models in the literature such as classification,
estimation, predictive modeling, clustering/segmentation,

0957-4174/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.eswa.2006.08.027

* Corresponding author.
E-mail address: michael@mail.tku.edu.tw (S.-H. Liao).

www.elsevier.com/locate/eswa

Expert Systems with Applications 34 (2008) 50–62

Expert Systems
with Applications

mailto:michael@mail.tku.edu.tw


affinity grouping or association rules, description and visu-
alization, as well as sequential modeling. Similarly, there
are also many application methods, including association
rules, sequential patterns, grouping analysis, classification
analysis and probability heuristic analysis (Fan, Lu, Mad-
nick, & Cheung, 2002; Hui & Jha, 2000; Goodwin, Dyne,
Lin, & Talbert, 2003; Mehta & Bhattacharyya, 2004;
Musaev, 2004; Tsechansky, Pliskin, Rabinowitz, & Porath,
1999). Knowledge of customers extracted through data
mining can be integrated with product and market-
ing knowledge from research and then provided to both
upstream suppliers and downstream retailers. Thus it can
serve as an important approach for new product develop-
ment. When effectively utilized, such knowledge extraction
can enable enterprises to gain a competitive edge through
production of customer-oriented goods that provide better
satisfaction to consumers and by speeding up the product
development process (Menon & Tong, 2005; Shaw, Sub-
ramaniam, & Tan, 2001).

Map display is a powerful tool with the ability to con-
vey a large amount of information in a limited space, and
it also provides an interactive tool to allow the user to
interact with the underlying information (Lin, 1997).
Thus, the mapping approach, which focuses on the use
of IT, can be used as a tool to support new product
development. Holmlund and Strandvik (1999) proposed
perception configuration as a new concept, and intro-
duced configuration maps as tools for analyzing percep-
tions in business relationship studies. Tülin and Russell
(1998) presented market maps with a probabilistic spatial
panel data model that allows the positions of products
sharing the same name to be correlated across product
categories. In a business setting, the combination of per-
ceptions by two parties (such as buyers and sellers) can
be represented as a perception configuration. All the per-
ceptions are depicted on the horizontal and vertical axes
of the map. This map can be used to capture both the
composition and the dynamics of perception configura-
tions, and it is generically applicable to dyadic perception
studies. Daniel, Wilson, and McDonald (2003) utilized a
marketing map to represent the best practice in market-
ing and also used the process map to understand how
IT can be deployed in order to support a marketing
information system. The map mainly illustrates the links
between various stages of the marketing process. Based
on this idea, this research implements a product map to
illustrate that new product development is essentially
the function that matches the enterprise’s offers to cus-
tomers’ demands.

This paper investigates the following research issues in
the development of new cosmetic products: What exactly
are the customers’ ‘‘needs’’ and ‘‘wants’’ for cosmetic prod-
ucts? Can product design and planning for product lines/
product collection be integrated with the knowledge of cus-
tomers? Can the knowledge of customers be transformed
into knowledge assets of the enterprises during the stage
of new product development (NPD)? To investigate these

research issues, the Apriori algorithm is a methodology
of association rules for data mining, which is implemented
to mine customer knowledge. Knowledge extracted from
data mining results is illustrated as knowledge patterns
and rules on a product map in order to propose possible
suggestions and solutions for NPD and marketing. The rest
of this paper is organized as follows. In Section 2, we pres-
ent research design, which discusses research problems and
introduces the proposed data mining system, including sys-
tem framework, relational database design, and physical
database design. Section 3 presents the data mining pro-
cess, including the Apriori algorithm, and the knowledge
extraction process. Section 4 illustrates results analysis
and a product map for NPD. Discussions and future works
are presented in Section 5; and Section 6 presents a brief
conclusion.

2. Research design

2.1. Research problems

Firstly, as the perfume, cosmetic, and toiletry prepara-
tions industry entered the 1990s, it faced many challenges
including regulatory changes, product safety concerns, calls
for scientific data to document product claims, increasing
environmentalism, desire for natural ingredients, pressure
from the growing animal rights movement, as well as
economy and market channels for product distribution
(Kumar, 2005). Lawmakers in USA began investigating
possible revisions to the traditional ‘drug’ and ‘cosmetic’
definitions established under the Food, Drug and Cosmetic
Act. Since the distinction between cosmetics and drugs was
sometimes vague. According to FDA guidelines (FDA,
2001), the term ‘cosmetic’ means (1) articles intended to
be rubbed, poured, sprinkled, or sprayed on, introduced
into, or otherwise applied to the human body or any part
thereof for cleansing, beautifying, promoting attractive-
ness, or altering the appearance, and (2) articles intended
for use as a component of any such articles; except that
such term shall not include soap. However, the products
claiming to offer medical benefits or physiological effects
were called over-the-counter (OTC) drugs. Examples of
the latter included antiperspirants, sunscreen products, hair
care products, shampoos to cure or remove dandruff, etc.
Therefore, cosmetics product development is not only con-
cerned with market reasons but also with safety concerns
(Katušin-Razem, Mihaljevic, & Razem, 2003; Morohoshi
et al., 2005; Pauwels & Rogiers, 2004). Secondly, cosmetic
products are a kind of biomedical and truly personal prod-
uct, which is popular with both woman and man on differ-
ent age scales (Chang & Chang, 2003). Therefore, cosmetic
products contain diversified product lines and collections in
order to fit the needs and wants to specific customer seg-
ments. This means that customer-orientation is the main
concern for investigating market segmentation according
to customers’ attributes, preferences, and personal factors.
Thus, customers’ needs and wants are important resource

S.-H. Liao et al. / Expert Systems with Applications 34 (2008) 50–62 51



https://isiarticles.com/article/2725