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: email@example.com *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 companies. 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 codes. 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 immaturity) • 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 components 2 Process standardisation: Standardising common tasks, sequence of tasks and task duration 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 following: • 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 organisation. 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 management. 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 development 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 amount. 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 suppliers. 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 adjustment. 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 gradually. 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 practice. References 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, pp.368–378. 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, pp.352–376. 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.
Pages to are hidden for
"Lean Product Development"Please download to view full document