IT-Based Knowledge Management Systems to Support the Design of Product Development Processes - Information Technology and Product Development - page 48

effect different factors have on the production processes as well as on the prod-
uct outcomes. Knowledge management systems (KMS) can support this process
by encouraging knowledge creation and transfer. There are few KMS that have
focused on the area of knowledge creation, particularly on the use of experiments
for defining development processes.
The objective of this chapter is to explore the potential for experiments as a
knowledge creation exercise in the product development process. Furthermore, it
outlines an IT-based KMS to support the creation, transfer, and use of knowledge
amongst engineers in designing and conducting experiments that lead to robust
product development processes.
This chapter is divided into a number of sections. Section 2, The Role of
Experiments in Product Development, examines the potential of experimentation in
developing robust products and processes. Section 3, Design of Experiments (DOE)
and Robust Design in Product Development, examines traditional approaches to
process design and investigates the use of DOE as an alternative approach. Section
4 focuses on the different types of knowledge used when creating knowledge via
experimentation and relates to Nonaka's knowledge spiral. Section 5 identifies
issues which need to be considered when managing knowledge to support the use
of experiments. In Section 6, we describe a case study that outlines the design of
an IT-based KMS to support the creation of knowledge related to design of product
development processes. Section 7 deals with some of the issues for future research
in this area.
3.2 The Role of Experiments in Product Development
The overall aim of product development is to create a high-quality product that
meets customer's expectations and requirements. To ensure such an outcome, prod-
uct development processes have to achieve a high level of consistency. Often this
would be to very narrow specifications. Consistently producing a high-quality prod-
uct is no accident. Production has associated with it many factors that might affect
the quality of the product - factors such as the raw material, machinery, settings
on the machines amongst others. Each of these factors has the ability to inhibit the
product performance and thus affect its ability to satisfy customer requirements. In
manufacturing, to identify the optimal settings needed to produce high-quality prod-
ucts, often experiments are run. A well-known and structured method of identifying
such settings is the robust design method (RDM).
RDM uses statistical experiments to identify the optimum factors and their set-
tings for product development processes. It allows the control of factors that may
affect the variation of the product quality. It also allows the identification of the rela-
tionships between these factors and the outcomes. The design of robust processes
for product development improves product quality, manufacturability, and reliability
(Leung Tsui, 1992).
Robust design methodology uses an approach called the DOE that has been
proven to be expensive to run in terms expertise and time (Breyfogle, 2003;