Lean Product Development by business901

VIEWS: 335 PAGES: 25

More Info
									Int. J. Six Sigma and Competitive Advantage, Vol. 5, No. 1, 2009                            75

The integration of DFSS, lean product development
and lean knowledge management

         Kai Yang* and Xianming Cai
         Department of Industrial and Manufacturing Engineering,
         Wayne State University,
         Detroit, MI 48201, USA
         Email: Kyang1@wayne.edu           Email: xmcai@wayne.edu
         *Corresponding author

         Abstract: Design for Six Sigma, lean product development and lean
         knowledge management are three effective methodologies in improving a
         product development process. The performance of a product development
         process can be measured by product value, product quality, product
         development lean time, efficiency and product life cycle cost. Design for
         Six Sigma can greatly improve product value and quality. Lean product
         development and lean knowledge management can help to achieve better
         product development lead time and efficiency by reducing wastes. Lean
         knowledge management can also help to improve product value and reduce
         product life cycle cost by adopting better technology and practices. These three
         methodologies are complementary and should be integrated.
         Keywords: Design for Six Sigma; lean product development; knowledge
         management; product development metrics.
         Reference to this paper should be made as follows: Yang, K. and Cai, X.
         (2009) ‘The integration of DFSS, lean product development and lean
         knowledge management’, Int. J. Six Sigma and Competitive Advantage,
         Vol. 5, No. 1, pp.75–99.
         Biographical notes: Kai Yang is a Professor in the Department of Industrial
         and Manufacturing Engineering, Wayne State University in Detroit, Michigan.
         His field of expertise includes quality engineering, engineering design
         methodologies, reliability engineering and healthcare management. He is
         author of five books, including his influential book, Design for Six Sigma. He
         received a MS degree in 1985, and a PhD degree in Industry Engineering in
         1990, both of them from the University of Michigan.
         Xianming Cai is a PhD candidate in the Department of Industrial and
         Manufacturing Engineering, Wayne State University in Detroit, Michigan. He
         received the first Master degree in the Design Technology from the National
         University of Singapore in 2005, and the second Master degree in the Industrial
         and Manufacturing Engineering from the Wayne State University in 2009.

1   Introduction

For many industries, products are major revenue generators. Products with good value
and quality will command higher prices and generate more sales, thus creating more
revenues. For most of the products, the ability of a product development process to

Copyright © 2009 Inderscience Enterprises Ltd.
76       K. Yang and X. Cai

accurately capture the market needs, respond quickly to shifting customer demands and
develop and launch the right product at right time is vitally important. Therefore,
the product development process is one of the most important processes for many
     Compared with the other types of processes, such as production processes and
financial transaction processes, the product development process is usually a more
technically sophisticated, costly and time-consuming process.
     Though Lean Six Sigma has achieved great success in many areas, rigidly plugging
in a generic Lean Six Sigma approach to product development process is not appropriate
and it could actually damage an originally effective product development process. One
such example is discussed in a recent cover page story of Business Week (Business Week,
11 June 2007 issue). It described how an inappropriate Six Sigma deployment damaged
3M’s long tradition of innovation culture. A good example of appropriate application of
Six Sigma and lean into a product development process is that of Samsung Group. Many
reputable publications, including Fortune magazine (Yun and Chua, 2002; Lewis, 2005),
described Samsung having developed a very effective Design for Six Sigma approach
that has greatly improved Samsung’s capabilities in innovation, efficiency and quality in
its research and development, and product development processes. An important tool of
DFSS, Theory of Inventive Problem Solving (TRIZ), greatly improved Samsung’s
innovation capability.
     The appropriate implementation of Six Sigma and lean approaches into product
development process is possible and can be extremely rewarding. A good Design for
Six Sigma approach is the right way of implementation of Six Sigma into the product
development process. It is our belief that this Design for Six Sigma approach should have
a strong innovation arm that can greatly enhance, not to stifle, the innovation capability
of the companies who employ them.
     Besides Six Sigma, how can we implement lean operation principles in product
development process in order to greatly improve its speed and efficiency? Again, rigidly
plugging in a generic lean approach in the product development process is also not
appropriate. The lean product development approach is the right approach for it.
Lean product development process is a product development process that is able to
develop products with maximum customer value and minimum wastes in resource and
high speed. The lean product development comes from many sources; they include
Toyota product development system (Kennedy, 2003; Morgan and Liker, 2006),
Don Reinertsen’s work (Reinertsen, 1997) and Yang’s work (Yang, 2008).
     If we closely look at what a product development team is doing every day, we will
find out that they are creating documents, compiling testing reports, doing design
analysis, creating specifications, building prototypes, designing and making tools for
producing the product, and so on. Whenever the product development team generates
enough useful information to produce the product effectively, reliably, economically and
with good quality, and the products shipped to customers are free of after sale problems,
our product development job is done. Clearly, the nature of the product development
process is an information and knowledge generation factory. Therefore, effective
knowledge management is also very important in the product development process.
Unfortunately, in most of the real product development processes, the waste of
knowledge and information is rampant (Kennedy, 2003), and effective management of
knowledge and information with minimal waste is a real challenge. In this paper, we will
present a lean knowledge management frame work which can take this challenge.
         DFSS, lean product development and lean knowledge management                      77

    There are two objectives for this paper. The first objective is to describe the frame
works of DFSS, lean product development and lean knowledge management. The second
objective is to discuss how these three strategies can be integrated to greatly enhance the
product development process.
    In this paper, we will first discuss the performance metrics for the product
development process in Section 2, because the understanding of these performance
metrics is the key to evaluate a product development process and to access the roles of
various methodologies, such as DFSS, in improving the product development process.
Section 3 will give a brief description of DFSS. Section 4 will discuss the lean product
development process. Section 5 will give an overview for our framework in lean
knowledge management. Section 6 will discuss the roles of those methodologies in the
product development process and how these methodologies can be integrated to achieve
the optimal effect.

2   Product development performance metrics

There are plenty of published literatures that discuss what performance metrics should
be used in the evaluation of a product development process. Clark and Fujimori (1991)
proposed a comprehensive set of performance metrics, which included product
development lead time, cost and total engineering hours. Nichols et al. (1994) proposed
11 major measures that companies should consider for benchmarking in product
development process, in which several measures such as manufacturing ramp up time
were added. Reinertsen (1997) thinks that the ultimate goal of product development and
production is to make profit. Kerssens-van Drongelen and Cook (1997) used a balanced
score card to proposed a set of measures for research and development performance.
Driva et al. (2001) carried out an in-depth analysis of performance measures used for
new product development for 10 companies. In his work, time to market, cost, quality
and sales are among the most frequently used performance metrics. Based on these
existing research literatures and our research, we propose the use of following
performance metrics to measure the performance of a product development processes.
Product Value: Product value is the most important performance metric. Product value
can be measured by the total profit, or total revenue generated by the product over its
entire market life cycle. Unfortunately, it is a lagging performance metric, which means
that we would not know it exactly before the launch of the product. The product value is
related to many factors, which include the following:
1   How well the voice of the customers are captured and deployed: If we develop a hit
    in market place, this product will generate a lot of profit.
2   Creativity and uniqueness: Besides capturing the voice of the customer, if our
    product is also a first of its kind product, and nobody else can provide the similar
    product, we will command the market price.
Product Quality: Product quality is a measure of quality, reliability and robustness.
In a durable consumer goods market, if a high-value product has consistent, repeatable
performance under various usage conditions, and can last a long time, the product will be
very successful. Product quality can be measured before the launch of the product by
product testing or even product performance simulation analysis. For some companies,
78       K. Yang and X. Cai

design score cards are established to measure the product quality. Some lagging
indicators, such as warranty costs, number of initial quality problems etc. can also be
used to measure product quality.
Product Development Lead Time: The product development lead time is the length of
time from the product development kick off time to the launch time of the product.
Product development lead time is a very important metric because it determines the speed
with which new products can be introduced into the marketplace.
Efficiency: In product development, efficiency is the cost of manpower and other
resources required for the product development. In developing same or similar product, a
more efficient product development process will use less resource.
Life Cycle Cost: Life cycle cost is the total cost related to a product. It includes
development costs, production costs, sales and distribution costs, service, support and
warranty costs and disposal costs. A product with high life cycle cost will certainly lower
its profitability. Product development has a particular vested interest in keeping the life
cycle cost for any product as low as possible.
    On a longer time scale, product value, product development lead time, efficiency and
life cycle costs will contribute a great deal to the level of customer satisfaction, market
share and revenues that the company will have. These will in turn translate into
profitability and influence the organisation’s long-term business viability.
    Design for Six Sigma (DFSS), lean product development and lean knowledge
management are effective and complementary approaches which will enhance the
product development performance metrics in different ways. In the subsequent section,
we will discuss these three approaches and their particular contributions in the product
development process.

3    Design for Six Sigma

Design for Six Sigma (DFSS) is a Six Sigma approach which involves designing
or redesigning of products or processes. Design for Six Sigma is a comprehensive
methodology that integrates discipline, training, project management, DFSS road maps
and DFSS tools to lead the product development projects. The goal of DFSS is to design
or redesign the product to intrinsically achieve maximum product value and achieve
superior quality and reliability. Similar to all Six Sigma activities, DFSS activities are
featured by launching and completing carefully selected projects guided by DFSS
roadmaps. There are two popular DFSS roadmaps; IDOV (Banuelas and Antony, 2004)
and DMADV (Cronemyr, 2007).
    From the product design point of view, DFSS is a Six Sigma approach that works on
the early stages of the product life cycle, as illustrated in Table 1. We can see that DFSS
can achieve many different objectives in these early stages in the product life cycle. In
the ideation stage, DFSS can help to accurately acquire the Voice of the Customer (VOC)
to ensure our product will satisfy the needs of the market. In concept development stage,
DFSS can help to successfully deploy VOC into product design, create innovative
product concepts and generate designs free of design vulnerabilities. In product design
stage, DFSS can help to develop optimised product parameter and tolerance design to
achieve high product quality, robustness and reliability. Overall, we can see that DFSS
can help our product development process to achieve high product value.
          DFSS, lean product development and lean knowledge management   79

Table 1     Product life cycle and Six Sigma approach
80          K. Yang and X. Cai

4      Lean product development

Lean operation practices have achieved a great deal of success in both manufacturing
industry and many service industries. Can these lean operation principles achieve the
same drastic results in product development process? The answer to this question is
positive. The birth place of lean manufacturing, Toyota, does have an edge in product
development process compared with North American automobile companies. Table 2
summarises the performance differences between Toyota and North American
Automobile companies near 1990 (Womack et al., 1990).
Table 2        Product development performance comparisons

                                                                        North American
    Measures                                         Toyota          Automobile Companies
    Average engineering hours per new
                                                      1.7                       3.1
    vehicle development (Million hours)
    Average development Time (Month)                  46.2                      60.4
    Employees per team                                485                       903
    Ratio of delayed project                         1 in 6                  1 in 2
    Achieve normal quality after launch               1.4                       11

Toyota’s product development system gained a lot of attention and it is discussed in
literature (Kennedy, 2003; Morgan and Liker, 2006).
     However, there are many distinct differences between product development process
and manufacturing process, so the lean principles have to be modified to work well in the
product development process. For manufacturing processes, what we are going to
produce is very clear in the beginning; the product that we produce has already been
designed so the value of the product is already known. For a product development
process, the value of the product is unknown until it is launched in the market place.
For manufacturing process, rework is treated as a waste; for product development
process, iterative improvement on product design is quite common. Even the goal of lean
operation is different between manufacturing process and product development process.
For manufacturing operation, the goal of lean operation is to minimise the waste and
increase the speed; for product development process, the goal is to simultaneously
maximise the product value and quality, reduce waste and increase development speed.
     Based on these differences, we can give the following definition to the lean product
development process:
The lean product development process is aimed to deliver more value in product by using
less resources through
•      thoroughly capturing the voice of customer and accurately deploying the customer
       value into design
•      accomplishing high product value and quality and low product cost by using the
       most appropriate technology and design
•      effectively transforming the voice of customer into high-quality design with
       high speed and low cost
•      relentlessly decreasing the wastes in the product development process.
        DFSS, lean product development and lean knowledge management                     81

4.1 Wastes in product development process
All lean operations start with identifying and eliminating wastes in the process. Unlike
the ‘seven wastes’ in the manufacturing process, there are no universally agreed waste
categories for a product development process. Here we have listed the following waste
categories for a product development process:
1   Wasted sale opportunities due to poor product value: The following items are
    included in this waste category:
    •   inability to capture right VOC information
    •   inability to translate VOC into appropriate design
    •   poor choice of technologies
    •   poor innovation capabilities
    •   failure to integrate innovation with VOC
    •   poor quality, reliability and robustness in designed product
    For this category of waste, Design for Six Sigma can help greatly in capturing high
    product value and developing products with high quality, reliability and robustness.
2   Waste in manpower, resource and time: The following items are included in this
    waste category:
    •   waste of manpower and resource in non-value added activities
    •   overburden on the people or resources: excessive work load and unrealistic
        deadlines often lead to half cooked projects and bug-ridden designs, this will
        eventually lead to rework
    •   unproductive meetings: it consumes man hours.
    Lean task management approach that will be discussed in the next section can
    effectively deal with this category of waste.
3   Waste in knowledge and information: The following items are included in this
    waste category:
    •   Reinvention: If someone else has already done this work, reinvention certainly is
        a waste of manpower and resource
    •   Mismatch of subsystems: Many design rework problems happen in the
        unexpected subsystem interactions
    •   Information loss and re-creation: This happens a lot in most of companies
    •   Miscommunication: Miscommunication among product development team
        members often leads to doing the wrong work, and then we have to redo it
    •   Searching for information, waiting for critical information: This is certainly not
        a value added activity.
    Lean knowledge management can effectively deal with this category of waste, and it
    will be discussed in Section 5.
82       K. Yang and X. Cai

4    Waste due to poor design practice: The following items are included in this
     waste category:
     •   Excessive design requirements: such as excessive tolerances, excessive material
         specifications, excessive operator requirements and so on
     •   Excessive complexity in design: The simplest design is the best design, given
         that we can deliver all the product functions
     •   Poor product architecture: Poor product architecture often leads to redesign,
         mismatch, and performance problems.
     Set-based design (Section 4.3) and lean product (Section 4.4) practices can
     effectively deal with this category of waste.

4.2 Lean task management
Lean task management deals with the waste in manpower, resource and time. Lean task
management consists of several approaches from various sources (Reinertsen, 1997;
Mascitelli, 2004; Morgan and Liker, 2006).

4.2.1 Increasing effective value added working time (Mascitelli, 2004)
Based on lean principles, all tasks performed by design engineers can be classified into
the following three categories:
1    Value Added: This category of tasks is the one that really moves the product design
     forward and creates values that external customers are willing to pay for the job
     done. The examples of tasks in this category include drafting new designs,
     conducting design simulation for improvement and creating application software
2    Non-Value Added but Necessary: This category of tasks is the one that may not
     move the product design forward and may not create values that external customers
     are willing to pay, but it is necessary under current circumstances. The examples of
     tasks in this category include design gate reviews, team coordination meeting and
     validation testing.
3    Waste: This category of tasks is the one that does not move the product design
     forward and it has no value for external customers. These tasks can be identified and
     eliminated. The examples of this category of tasks include time spent on moving
     from meeting to meeting, voice mail checking, searching for information and so on.
Mascitelli stressed on the importance of increasing the ratio of value added time, and
decreasing the ratio of non-value added but necessary and the waste, as illustrated in
Figure 1. Mascitelli stated that based on industry survey, in an 8-hour working day, the
average value added time is only 1.7 hours in the Western companies. However, Toyota
claimed that its average value added time is more than 50%.
           DFSS, lean product development and lean knowledge management                      83

Figure 1    Identifying and increasing value added time in product development (see online
            version for colours)

Mascitelli also thinks that it is very important to allocate uninterrupted time slot, 3 hours
or more, to each design engineer for each working day in order to greatly improve the
productivity. From a human factor perspective, when design engineers are doing product
development work, it takes some time to achieve mental focus on the job, and it takes
some time to get even a small task completed.

4.2.2 Smoothing product development job flow
Product development is a complicated process which has multiple stages and involves
many different departments and teams. Each team or engineer will work on several
projects during the whole product development process, so some projects have to wait in
the queue until the current project is finished so the team or engineer can free their hands
to work on them. In this case, the product development process is a queuing network and
how we sequence the jobs and how we assign the work load and timing will make a lot of
difference in overall product development lead time and efficiency. The queuing theory
can provide great help for us as how to reduce the waiting time and how to improve the
throughput (number of projects finished/unit time).
    There are several important results in queuing theory that are relevant to product
development process.
1   Batch Queue is Inefficient: Batch queue means the jobs are coming to queue in big
    groups, or batches. It is a well-known result in the queuing theory that the waiting
    time and queue length are significantly larger in a batch queue system than in a
    queuing system in which the job arrivals are in small pieces, given if the service
    rates and total amount of jobs over a long period of time are the same for both
    queuing systems. The implication for the product development practice is that if
    we load product development team or engineer in big chucks of work, instead of
    one by one, our throughput will be low, and waiting time will be long.
2   Nonlinear relationship between capacity utilisation and queue length: Queuing
    theory states that the relationship between capacity utilisation and average waiting
    time is a non-linear relationship, as illustrated in Figure 2. Capacity utilisation is
84         K. Yang and X. Cai

     defined as the percentage of time that the server is busy. What this relationship
     indicates is that when the server is about partially loaded, the waiting time will be
     very low; however, if we increase the capacity utilisation towards 100%, then
     queue length will grow very fast and the waiting time will be extremely high. The
     implication for the product development process is that overburdening the product
     team or engineers will make the product development lead time much longer.
3    Constant Job Arriving Rate vs Variable Arriving Rate: In queuing theory, it is also a
     well-known result that given the same mean job arrival rate, the queuing system with
     higher variance in arrival rate will result in longer waiting time and queue length
     than that of the queuing system with lower variance in arrival rate. The implication
     for the product development process is that if we load jobs to engineers evenly, then
     the throughput will be higher.
4    Uneven Job Size vs Similar Job Size: Again, we are comparing two queuing systems.
     Suppose that in both queues, the arriving time patterns are the same. The first
     queuing system is such that every incoming job takes about same amount of time to
     be finished by the server. The second queuing system is such that every incoming
     job has different sizes, even if the average job processing time is the same as the
     first queue system. Then the waiting time and queue length of the first queuing
     system will be shorter than those of the second one. The implication for the product
     development is that loading engineers with similar job sizes for each task will
     increase the job throughput.

Figure 2    Queue length vs Capacity utilisation

Toyota product development system (Kennedy, 2003; Morgan and Liker, 2006) has
developed some well-established practices to ensure smooth product development job
flows. Specifically, one of the key principles in Toyota product development system is to
create levelled product development process flow. This principle calls for synchronising
activities across different functional departments in a product development organisation.
           DFSS, lean product development and lean knowledge management                  85

It also calls for evenly distributed workload to various departments and engineers. This
approach creates steady work load and job flow so the tasks will flow through the
organisation smoothly and waiting lines will be unlikely to occur.

4.3 Set-based design
In Toyota’s product development process (Morgan and Liker, 2006), one of its design
principles calls for ‘front-load the product development process to explore thoroughly
alternative solutions while there is maximum design space’. The technical approach for
this ‘front-loading’ principle is the Set-based design practice.
    Set-based design approach should be used in concept design stage. In regular design
practice, for each design entity, be it a subsystem or a component, we will start with a
small number of alternative design concepts, select one seemingly good concept, and
move into detailed design and conduct evaluation test. If the test shows the concept is
acceptable, we will move this concept to the parameter design, and prototyping stage. If
the test shows this concept is not acceptable, we will start another raw concept and do
another round of development; we may iterate this process until an acceptable design is
found. This regular design practice is illustrated in Figure 3.

Figure 3    Regular concept design process

On the other hand, the Set-based design will simultaneously start several concepts for the
same design entity. For example, in automobile door design, door lock is a subsystem, in
regular design practice, we will pick one door lock design concept and move to detailed
design, build a prototype and do testing, if the testing is not satisfactory, we will either
fine tune the current design, or pick another concept and start over. For Set-based design
practice, we will form several design teams simultaneously, and each team will start
a different concept for the door lock and grow its concept by detailisation, design
evaluation and tests. As the concept design for each team moves forward, we will
86         K. Yang and X. Cai

evaluate all the concepts simultaneously, by slowly eliminating the weakest design, and
combining the advantages of several better design concepts; hence we will eventually
end up with a superior design. It is very important that the Set-based design stay in the
concept design stage, we can test the concepts through CAD (Computer-Aided Design)
simulation, Alpha prototype, small-scale lab test and so on. This will ensure that the
Set-based design approach will not be expensive and time consuming. Figure 4 illustrates
how Set-based design works.

Figure 4    Set-based concept design

The Set-based design has been studied by many researchers. Sobek et al. (1999) pointed
out that Set-based design practice will enable product development team to gain much
wider and deeper knowledge about design entity by exploring more design alternatives.
This improved knowledge will help the designer to come up with a better final design.
Costa and Sobek (2003) concluded that the advantages of Set-based design include
discovering better concepts early in the design process. Zhang et al.’s (2009) work
concluded that Set-based design practice will enhance product concept flexibility, which
is the ability to revise design quickly and accurately with the changing market needs.

4.4 Lean product
In Yang’s recent book, Voice of the Customer Capturing and Analysis (Yang, 2008), it is
stated that the product development process is an information creation process. The
consumer of this information is the product design. Axiom 2 of the axiomatic design
principles (Suh, 1990) states that the best design is the design that delivers all product
         DFSS, lean product development and lean knowledge management                   87

functions satisfactorily and has the lowest possible information content, or complexity
level. In Yang’s book, it is stated that the ideal product development process is such that
it creates information and knowledge at the highest efficiency, speed and quality, but the
consumption of information for each product is minimum.
     The information consumption is proportional to the complexity of the product design.
So we need to trim all unnecessary complexities out of product design. A product design
that is free of unnecessary complexities is called a lean product.
     The complexity in engineering design is related to the following:
•   number of functions and parts
•   complexity in product architecture (How different modules and design parameters
    are related to each other)
•   uncertainty (such as uncertainty caused by variation, quality and technical
•   complex relationship between design parameters and product performance.
Based on the work of Huthwaite (2004), the following approaches can be used to reduce
the unnecessary complexities in the product design and create lean products:
•   reducing unnecessary product functions and parts
•   loosening up unreasonable tolerances
•   using standard/out of shelf parts
•   controlling technical immaturity
•   avoiding complicated user/operator requirements
•   avoiding complicated interface requirements.
In Toyota product development system, standardisation is extensively used to reduce the
information consumption in product design. One of Toyota’s product development
principles is called utilising rigorous standardisation to reduce variation and creating
flexibility and predictable outcomes. This principle calls for applying the following four
kinds of standardisation over the product development organisation:
1   Design standardisation: Engineering checklist, standard architecture, share common
2   Process standardisation: Standardising common tasks, sequence of tasks and task
3   Skill set standardisation
4   Standardised skill inventories
This principle uses the fact that standardisation will reduce complexities in product
design and reduce confusions in communications among engineers. Standardisation will
make each job more transparent and uniform, so you can have more predictable
outcomes. Standardisation will also reduce the waste caused by reinvention, mismatch,
information loss and re-creation.
88       K. Yang and X. Cai

5    Lean knowledge management

5.1 Overview of knowledge management
Knowledge management is a process that an organisation can use to identify, create,
represent and distribute knowledge. Knowledge management has always existed
throughout the history. Apprenticeship, on-job training and corporate libraries are
examples of knowledge management practices. With the advancement of computer
technology, more and more computer-based systems, such as knowledge base, best
practice database, and expert systems, have been introduced in the knowledge
management system. There are numerous literatures in the area of knowledge
management; Kotnour (1999), Nonaka and Takeuchi (1995), Liao (2003) and Stankosky
(2005) are among the most informative literatures.
    In their book, The knowledge creating company, Nonaka and Takeuchi (1995) stated
that a successful knowledge management and knowledge creation system was crucial for
successes of many Japanese companies, in which it is really important to convert the
individual’s knowledge from employees into organisational knowledge. Nonaka and
Takeuchi discussed the distinction between explicit knowledge and tacit knowledge.
Explicit knowledge is knowledge that can be explicitly expressed and communicated.
Written reports, presentations, research papers, books, instructional manuals, procedures,
sound and visual records and softwares are examples of explicit knowledge. On the other
hand, tacit knowledge is knowledge that cannot be expressed explicitly in a stand-alone
form. Tacit knowledge is buried deeply inside people and it is difficult to express
and transmit it explicitly. However, it usually can be transmitted only through a lot of
people-to-people interaction. The simplest example of tacit knowledge is that nobody can
learn NBA basketball skills by reading or by watching audio/video media. Nonaka and
Takeuchi stated that there is a big portion of knowledge within an organisation that is in
the form of tacit knowledge, and that tacit and explicit knowledge can be transformed
from one to another. A successful knowledge management programme should be able to
convert tacit knowledge into explicit knowledge in order to share it.
    Thought it already becomes a multibillion dollar business, the traditional knowledge
management programmes do have a few pitfalls:
1    Based on the research of Morgan and Liker (2006), many computerised knowledge
     management systems in Western companies suffer from obsolete and inadequate
     contents due to lack of constant inflow and updating of fresh knowledge.
2    Traditional knowledge management programmes are based on managing explicit
     knowledge. There is little technical means to capture, convert and share tacit
     knowledge (Fahey and Prusak, 1998).
3    Most of the computerised knowledge management systems are developed and driven
     by the advancement of IT technologies, few of them are designed based on the need
     of real product development practices, with the exception of Toyota (Morgan and
     Liker, 2006).
We will first describe the role of knowledge and information in the product development
process in Section 5.2. In Sections 5.3 and 5.4, we will discuss our framework of a lean
knowledge management system.
           DFSS, lean product development and lean knowledge management               89

5.2 Knowledge and information in product development
In Yang’s recent book (Yang, 2008), Voice of the Customer Capture and Analysis, he
described in detail that there are three types of knowledge and information activities
in the product development process; they are information mining, information
transformation and information and knowledge creation.

5.2.1 Information mining
Information mining is the extraction of valuable information from information sources.
For example, in research and publication work, the literature survey is an information
mining work.
    In the product development process, there are two major types of information mining
work, they are the mining of the voice of customer, and the mining of the technological
information, as illustrated in Figure 5.

Figure 5    Information mining in the product development process

1   Information Mining of the Voice of Customers: According to Figure 5, the product
    development is a sequence of ‘mapping’ processes proposed by Suh (1990). The
    very first step is the mapping from the customer domain to the function domain; this
    step is really a rigorous product definition step. The whole value of the product is
90       K. Yang and X. Cai

     largely determined by whether the product will be welcomed by the potential
     external customers, or buyers. It is not an exaggeration that accurately capturing the
     voice of customer is like a strike in the gold. There are numerous examples about the
     importance of the mining of the voice of the customer and success stories, such as
     Federal Express, Starbucks and iPod.
2    Information Mining of the Technology Information and Knowledge: In product
     development process, we need to obtain the technological know-how and
     information to transform the voice of customer into reality. In this information
     mining work, the quality and speed of the information mining are very important.
The quality of this information mining means that we are getting the best technology in
terms of performance, the cost of technology and the robustness of the technology.
    The speed of this information mining is also critically important; the ideal technology
information mining process is featured by the abilities to pull the right technology
information to the right people at high speed.

5.2.2 Information transformation
Information transformation is similar to ‘mappings’ in Suh’s axiomatic design theory
illustrated in Figure 5. Another kind of information transformation is the hierarchical
design deployment, as illustrated in Figure 6. Most of the information transformation
work deals with existing knowledge. For example, in automobile product development,
body design and assembly is a big chunk of design work. Body styles have to change to
make cars catch the fashion trend. But there is very little new knowledge needed in this
design work. The types of work in this information transformation usually include the
•    given a need, ‘pull’ design solutions from a known source
•    design of interfaces
•    shape and form design
•    system flow down and integration
•    design analysis, simulation
•    testing
•    prototype building.
For complicated products with thousands of parts or more, the scope of this information
transformation work can be very large. We need a whole organisation with many people;
they have different tasks, different knowledge background and experience to work
together. The information and knowledge need to flow smoothly through the whole
           DFSS, lean product development and lean knowledge management                 91

Figure 6    Hierarchical design deployment (see online version for colours)

5.2.3 Information and knowledge creation
In the product development process, there are some design tasks that no ready solutions
can be pulled for from somewhere. These tasks involve creating new information and
new knowledge. Here are some of the scenarios:
•   Resolution of some technical bottlenecks that nobody has accomplished before:
    For example, the fuel efficiency of internal combustion engine is low. With the
    increasing petroleum price, this technical difficulty needs to be resolved.
•   Development of the new generation of product: we want to drastically improve
    our product’s performance, cost and so on to move ahead of competition. This
    improvement is not merely a fine tuning of existing product.
•   Develop a product with new marketing concept.
•   Technology push product development: Many research results are coming out from
    universities, research institutions and so on, and many patents are created every year.
    Manufacturers are bringing these new technologies into their products.
92       K. Yang and X. Cai

5.3 Lean knowledge and management
As we discussed in Section 5.2, the product development process consists of information
mining, information transformation and knowledge creation. An ideal product
development process should be such that it creates information and knowledge at the
highest efficiency, speed and quality. At the same time, the waste of information and
knowledge in the product design should be at a minimum. In actual product development
processes, however, the waste of information and knowledge is difficult to see and is
running rampant in many companies. An effective knowledge and information
management system is crucial in a lean product development process. In this subsection,
we are going to discuss the following approaches in lean knowledge and information

5.3.1 Knowledge and information supermarket
As we discussed earlier, the lean production system is based on the ‘pull’ concept. The
pull concept is derived from supermarket inventory replenish practice. In knowledge and
information management, the concept of ‘supermarket’ is also very appealing. In today’s
electronic information technology, we do not need to consume and store multiple
copies of information. However, knowledge and information could become outdated and
obsolete from time to time. The meaning of knowledge and information supermarket is
as following:
1    the information and knowledge is always fresh and up to date
2    the information and knowledge is sufficient to serve all the needs of the product
3    we know where each information and knowledge is stored
4    the information is ready to be pulled at the right time, the right kind and right
A good information and knowledge supermarket will greatly reduce the time and
resource spent in retrieving and search information, misalignments in the product
development projects, knowledge recreation and reinvention.

5.3.2 Several knowledge management practices in Toyota
There are several knowledge management practices that are derived from the real needs
of the product development practice and they are proven to be very effective. They are
V-Comm system and Visible knowledge.
V-Comm System:
Toyota’s V-Comm system (Morgan and Liker, 2006) is a knowledge and information
management system that is close to our ‘information and knowledge supermarket’
concept. V-Comm stands for ‘Visual and Virtual Communication’ for Toyota. V-Comm
was initially launched in 1996, as a ‘digital build’ software, and it is improved
continuously. In 2001, V-Comm became more mature, and helped Toyota to slash the
car development time from 18 month to 11 Months. It is currently a software platform for
design review and communication.
         DFSS, lean product development and lean knowledge management                    93

    V-Comm’s main functions include:
•   virtual design prototyping and production prototyping
•   design review and visual communication
•   model-based design simulation and error checking
•   knowledge database and communication
V-Comm’s knowledge database is very comprehensive and always kept up to date, it has
the following contents:
•   Best Practice Files
•   Past Issues, quality hazards
•   Recommended key specifications
V-Comm is a great success in lean knowledge and information management because it is
right on the main traffic points of the product development process; and it is a system
that is ready to pull information when needed at the right place. It is updated constantly
and contains comprehensive amount of information and it is easy to search.
Visible Knowledge and A3 Report:
Visible knowledge is a very important component of Toyota’s product development
system. Visible knowledge is a practice to capture and illustrate knowledge so that it is
easily shared with other people within the organisation.
    A3 report is a kind of report where A3 size papers (11’ × 17’) are used. Each
report will use exactly one A3 size paper, or equivalent to a two pager report for regular
8’ × 11’ size paper. Based on the objectives, there are primarily three kinds of A3 report:
•   Knowledge sharing
•   Problem Solving
•   Project Status Report
In Toyota, A3 report is one effective tool to convert individual employees’ tacit
knowledge into explicit knowledge and let the whole organisation to capture and share
this knowledge. In many other western companies, half finished, ill placed, ill explained
and incomplete reports and documents are very common; only the people who created
them can understand them, and a lot of knowledge is in employees’ head. Employees’
turn over and retirement will create huge loss of knowledge for their employer.
    In Toyota, before the meeting takes place, even if it is an one to one meeting, the
participants usually email an A3 report to each other. In this way, the participants can get
enough information on the subject from each other before the meeting, and serious
discussion can take place very quickly, thus resulting in very productive meetings.

5.4 Computerised lean knowledge management system
From our early discussion, a well-structured knowledge and information supermarket that
can provide all the right kind of knowledge and information for a product development
process at right time and right place is essential to effectively create product design with
94       K. Yang and X. Cai

minimum waste. This knowledge and information supermarket will only be possible with
the aid of computer technology, and we will call it a computerised lean knowledge
management system.
    The main functions of this computerised lean knowledge management system should
include the following:
1    Capture the relevant external explicit knowledge base for product development,
     such as relevant patents, technology information obtainable from open sources,
     such as university, internet, competitors’ products benchmarking and so on.
2    Capture, maintain and update the marketing and voice of customer information
     (Omar et al., 1999).
3    Capture, maintain and update corporation’s internal explicit knowledge base, such as
     computer files of standardised or custom made design modules, testing procedures,
     design check lists, technical reports, testing results and so on.
4    Capture, convert and share tacit knowledge of company employees, and possibly
5    Facilitate easy access of all information for all kind of internal users, regardless of
     their professional affiliation and location.
6    Manage the knowledge contents to ensure that the contents are well organised,
     constantly updated and free of errors.
Based on these main functions, our proposed computerised lean knowledge management
system has the following main modules. It is an integration of a traditional ECM and a
centrally managed CMS with a corporate Wiki.
Enterprise Content Management (ECM): It is a technical system that can capture,
manage, store, preserve, update and deliver contents and documents based on
organisational needs. ECM can be enabled by Enterprise Content Integration (ECI)
technology. ECI is a middleware software technology that connects together all computer
systems that manage documents and digital contents (Fisher, 2006; Bridges, 2007).
Content Management System (CMS): It is a system that used to manage the contents.
Content management systems are deployed primarily for the interactive use by a
potentially large number of contributors (McQueen, 2004; Huseyin and Yavuz, 2005).
For example, wiki software is the content management system for the website Wikipedia.
Actually, we believe that Wikipedia is an excellent example of massive collection of
information and knowledge from many individual users and massive transformation of
tacit knowledge of web contributors into a great wealth of explicit information loaded on
Wikipedia ready to be shared by people all over the world. Wikipedia is also a good
example of information supermarket, because it is well organised, easy to find and
always up to date (Hoehndorf et al., 2006). We believe that a content management
system similar to wiki software can play a great role of capturing employees and
suppliers tacit knowledge and converting it into a great knowledgebase.
           DFSS, lean product development and lean knowledge management                 95

Figure 7    The structure of proposed lean knowledge management system (see online
            version for colours)

6   The integration of DFSS, lean product development and lean
    knowledge management

Design for Six Sigma, lean product development and lean knowledge management are
three effective methodologies related to the product development process. These three
methodologies improve the product development processes in different ways.
    Table 3 provides a summary about how these three approaches will help to improve
the product development performance metrics described in Section 2.
    Table 3 clearly indicates that merely adoption of one of these strategies, be it Design
for Six Sigma, lean product development or lean knowledge management, will not be
adequate to improve all major aspects of product development performance. The
integration of these three approaches is very important. Before we outline an integration
strategy of these three approaches, we will conduct a brief analysis of the pros and cons
of these three approaches.

6.1 Analysis
6.1.1 Design for Six Sigma
Design for Six Sigma is centred at applying effective tools to strengthen every stage
of the product development process (Yang, 2003; Cronemyr, 2007). But Design for
Six Sigma will not usually be used to analyse and improve the product development
process itself; there is no literature or reported cases of using DFSS to design the
product development process, to the best of our knowledge. Therefore, if the product
development process itself is not efficient, there is no direct experience of using DFSS to
remove wastes and inefficiencies in the product development process. Also, the DFSS
tools used in product development practice are not related to knowledge and information
management. Application of DFSS tools will not ensure that the design engineers will
use the most appropriate technologies and practices.
96        K. Yang and X. Cai

Table 3     The effects of DFSS, lean PD, lean knowledge management and product development
            performance metrics
        DFSS, lean product development and lean knowledge management                 97

6.1.2 Lean product development
Lean product development is centred at analysing and improving the product
development process itself by removing wastes. However, lean product development
practice does not address the issues of quality, reliability and robustness in product
design directly. Lean product development does not deploy innovation or statistical
tools to strengthen specific design activities. In a recent literature review by Baines
et al. (2006), it is also found that there are vast differences in implementing
knowledge/information management in real-world lean product development practices.
For example, there is very little emphasis in knowledge management in several lean
product development approaches (Huthwaite, 2004; Mascitelli, 2004; Oppenheim, 2004).
On the other hand, Toyota’s lean product development approach has a strong lean
knowledge management emphasis and a workable approach.

6.1.3 Lean knowledge development
Lean in knowledge management and research development is a new and challenging
area. Peter Drucker’s (1994) vision of knowledge economy is mostly a reality now. Goh
(2002) stated that the real challenge for Six Sigma is how to adapt to knowledge-based
environment. Toyota’s successful A3 approach and V-Comm experience showed that
lean knowledge management is a very powerful tool. For product development process,
lean knowledge management can reduce the waste of information and knowledge, and
ensure the product development team to access the most updated, most appropriate
technical, business and process knowledge. Lean knowledge management cannot reduce
the waste of manpower and resource. Lean knowledge management also cannot play the
roles of DFSS tools.
    Obviously, with more and more information and knowledge becoming digital,
effective computerised lean knowledge management systems will have to be our future.
However, developing, implementing and perfecting such systems is still extremely
challenging due to huge cost, enormous manpower commitment and difficult work habit

6.2 Integration strategy
Our analysis clearly showed that DFSS, lean product development and lean knowledge
management will all help the product development processes significantly, but none of
these approaches is self-sufficient. We need to implement all of these to achieve the
best result.
    From our analysis, DFSS and lean product development improve different aspects of
the product development process. DFSS improves product value and product quality,
whereas lean product development improves product development lead time, efficiency,
flexibility and product development cost. Also, DFSS and lean product development
work in very different ways; DFSS applies tools to improve the value and quality of
designed product, whereas lean product development speeds up design process by
removing wastes. DFSS and lean product development are mutually complementary and
they are not interfering with each other. The relationship between DFSS and lean product
development is similar to that of Six Sigma and lean. DFSS and lean product
development can be implemented in parallel. The introduction and implementation of
98       K. Yang and X. Cai

lean knowledge management will improve product development process and benefit
design team tremendously. However, any computerised knowledge management system
will involve big undertaking; it is easier to start with low-cost approach, such as
A3 report, visible knowledge, and build up our lean knowledge management system

7    Summary

Design for Six Sigma consists of roadmap and tools that can improve the value and
quality of designed products by better capturing the voice of customer, adopting better
design practices, fostering innovation and achieving better performance and robustness
through parameter and tolerance design.
    Lean product development is a custom-made lean operation principle for the product
development process that can reduce product development lead time, improve product
development efficiency and reduce product development cost by reducing wastes in the
product development process.
    Lean knowledge management is a knowledge management system that supports lean
product development practices and preserve and improve organisational knowledge.
Lean knowledge management will reduce product development lead time, increase
product development efficiency by greatly reducing the wastes of knowledge and
information. Lean knowledge management may improve the product value by adopting
better knowledge and technology in the product. Lean knowledge management may also
reduce product’s production costs by adopting better technology and practices.
    Designs for Six Sigma and lean product development are complementary of each
other and they can be implemented in parallel. Lean knowledge management system
should be built gradually in order to be seamlessly integrated into our daily working

Baines, T., Lightfoot, H., Williams, G.M. and Greenough, R. (2006) ‘State of the art in lean
     engineering: a literature review on white collar lean’, Proceedings of the I MECH Part B
     Journal of Engineering Manufacture, Vol. 220, No. 9, pp.1539–1547.
Banuelas, R. and Antony, J. (2004) ‘Six Sigma or Design for Six Sigma’, The TQM Magazine,
     Vol. 16, No. 4, pp.250–263.
Bridges, J.D. (2007) ‘Taking ECM from concept to reality’, Information Management Journal,
     Vol. 41, pp.30–32.
Clark, K. and Fujimori, T. (1991) Product Development Performance: Strategy, Organization and
     Management in the World Auto Industry, Harvard Business School Press, Boston, MA.
Costa, R. and Sobek II, D.K. (2003) ‘Iteration in engineering design: inherent and unavoidable or
     product of choices made?’, Proceedings of DETC'03, 2–6 September, Chicago, IL, USA.
Cronemyr, P. (2007) ‘DMAIC and DMADV – differences, similarities and synergies’,
     International Journal of Six Sigma and Competitive Advantage, Vol. 3, No. 3, pp.193–209.
Driva, H., Pawar, K.S. and Menon, U. (2001) ‘Performance evaluation of new product
     development from a company perspective’, Integrated Manufacturing Systems, Vol. 12, No. 5,
Drucker, P. (1994) ‘The age of transformation’, The Atlantic Monthly, pp. 53–80.
         DFSS, lean product development and lean knowledge management                        99

Fahey, L. and Prusak, L. (1998) ‘The eleven deadliest sins of knowledge management’,
     California Management Review, Vol. 40, No. 3, pp.265–275.
Fisher, G. (2006) ‘Enterprise content management’, Information Management & Technology,
     Vol. 39, pp.162–163.
Goh, T.N. (2002) ‘A strategic assessment of Six Sigma’, Quality and Reliability Engineering
     International, Vol. 18, pp.403–410.
Hoehndorf, R., Prüfer, K., Backhaus, M., Herre, H., Kelso, J., Loebe, F. and Visagie, J. (2006)
     ‘A proposal for a gene functions wiki’, Springer-Verlag, Montpellier, France, pp.669–678.
Huseyin, S. and Yavuz, A. (2005) ‘Overcoming scormification difficulties in implementing a
     learning content management system’, ITHET 6th Annual International Conference,
     IEEE, Santo Domingo, Dominican Republic, pp.3–11.
Huthwaite, B. (2004) The Lean Design Solution, Institute for Lean Design, Mackinac Island, Mich.
Kennedy, M. (2003) Product Development for the Lean Enterprise, The Oaklea Press,
     Richmond, VA.
Kerssens-van Drongelen, I.C. and Cook, A. (1997) ‘Design principles for the development of
     measurement systems for research and development process’, Research and Development
     Management, Vol. 27, No. 4, pp.345–357.
Kotnour, T. (1999) ‘A learning framework for knowledge management’, Journal of Project
     Management, Vol. 3, No. 2, pp.32–38.
Lewis, P. (2005) ‘A perpetual crisis machine’, Fortune, 19 September.
Liao, S. (2003) ‘Knowledge management technologies and applications – literature review from
     1995 to 2002’, Expert System with Applications, Vol. 25, pp.155–164.
Mascitelli, R. (2004) The Lean Design Guidebook, Technology Perspectives, Woodside, CA.
McQueen, H. (2004) ‘Plan ahead for your content management system’, Information Outlook,
     Vol. 8, pp.15–17.
Morgan, J. and Liker, J. (2006) The Toyota Product Development System: Integrating People,
     Process and Technology, Productivity Press, Portland, OR.
Nichols, K., Pye, A. and Mynott, C. (1994) UK Product Development: A Comprehensive Study of
     Development Performance in the UK Industry, Design Council, London.
Nonaka, I. and Takeuchi, H. (1995) The Knowledge-Creating Company: How Japanese Companies
     Create the Dynamics of Innovation, Oxford University Press, New York.
Omar, A.R., Harding, J.A. and Popplewell, K. (1999) ‘Design for customer satisfaction: an
     information modelling approach’, Integrated Manufacturing Systems, Vol. 10, pp.199–209.
Oppenheim, B.W. (2004) ‘Lean product development flow’, Systems Engineering, Vol. 7, No. 4,
Reinertsen, D. (1997) Managing the Design Factory, Freedom Press, New York.
Sobek II, D.K., Ward, A. and Liker, J.K. (1999) ‘Toyota’s principles of set-based concurrent
     engineering’, Sloan Management Review, Vol. 40, No. 2, pp.67–82.
Stankosky, M. (Ed.) (2005) Creating the Discipline of Knowledge Management: The Latest in
     University Research, Butterworth-Heinemann, Burlington, MA.
Suh, N.P. (1990) The Principles of Design, 1st ed., Oxford University Press, New York.
Womack, J.P., Jones, D.T. and Roos, D. (1990) The Machine that Changed The World, Rawson
     Associates, New York.
Yang, K. (2003) Design for Six Sigma, A Roadmap for Product Development, McGraw-Hill,
     New York.
Yang, K. (2008) Voice of the Customer Capture and Analysis, McGraw-Hill, New York.
Yun, J. and Chua, R. (2002, November) ‘Samsung uses six sigma to change its image’, Six Sigma
     Forum Magazine, Vol. 2, No. 1, pp.13–16.
Zhang, Q., Vonderembse, M. and Cao, M. (2009) ‘Product concept and prototype flexibility in
     manufacturing: implications for customer satisfaction’, European Journal of Operational
     Research, Vol. 194, pp.143–154.

To top