The Emerald Research Register for this journal is available at www.emeraldinsight.

com/researchregister

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0265-671X.htm

QUALITY AND RELIABILITY CORNER

Quality revolution: leading the innovation and competitive advantages

M.A. Berawi
Civil Engineering and Construction Management Department, Oxford Brookes University, Oxford, UK
Keywords Quality, Value analysis, Knowledge management, Industry competitiveness, Innovation, Competitive advantage Abstract Knowledge management (KM) addresses the critical issues of organizational adaptation, survival and competence in a rapidly evolving environment. KM embodies organizational processes that seek a synergistic combination of the data and information processing capabilities of information and communication technologies (ICT), and the creative and innovative capacity of human beings to improve ICT. In that role, knowledge management has the capacity to improve quality management and avoid or minimize losses and weaknesses that result from poor performance as well as increase the competitive level of the company and inability to maximize its survival potential in the global marketplace. To achieve quality, all parties including the clients, company consultants, contractors, entrepreneurs, suppliers, and governing bodies (i.e. all involved stakeholders) must encourage collaboration and commitment to quality. Design-based organizations have to be quality driven to support healthy growth in todays competitive market. In the march towards globalization (i.e. the one world community) as well as local industries comprising many companies, their design-based organizations need to have superior quality management and knowledge management capabilities to anticipate changes. Intelligence and knowledge is a form of strategic capital, which can be cultivated, created, stored, managed, retrieved and measured to give competitive advantage. To anticipate the needs of the marketplace, a new system that is called quality value model (QVM) is proposed. QVM is a combination of communication and information technologies (ICT) as well as creative and innovative capabilities of human beings to meet challenges and to develop high-quality products. Quality is a reection of the properties in a product that customers use to value and evaluate its economic worth. Furthermore, the mathematical technique within QVM calculates input-factor changes within the product. This enables companies to be more responsive and aids prediction in respect to change and the management of decision uncertainty.

Quality revolution

425
Received October 2002 Revised November 2003 Accepted November 2003

Introduction The combination of intelligence and knowledge is a key factor in the move to globalization. Knowledge management involves the identication and analysis of available and required knowledge, and the subsequent planning and control of actions to develop knowledge assets so as to fulll the organizational objective of protability through higher quality levels as recognized by customers (Berawi, 2002a). Knowledge assets comprise the knowledge regarding markets, products, technologies and organizations that a business owns or needs to own and that enable the enterprise to generate prots. High-quality goals for products, along with a good means to schedule, track and predict costs are the mechanism needed to manage the design from concept

International Journal of Quality & Reliability Management Vol. 21 No. 4, 2004 pp. 425-438 q Emerald Group Publishing Limited 0265-671X DOI 10.1108/02656710410530118

IJQRM 21,4

426

to delivery (Abdul-Rahman and Berawi, 2001). Products that are completed on time and in budget are important parameters for success and quality through QVM. The following are ve denitions and perspectives on quality (Garvin, 1998): (1) Transcendent: Quality is neither mind nor matter, but a third entity independent of the two . . . even though quality cannot be dened, you know what it is (Pirsig, 1974). (2) Product-based: Quality is based on the presence or absence of a certain attribute (Abbott, 1955). (3) Manufacturing-based: Quality is conformation to requirement (Crosby, 1979; Ishikawa, 1985). (4) User-based: Quality means tness for use (Juran, 1974; Deming, 1982). (5) Value-based: Quality means the degree of the excellence at an acceptable price and the control of variability at an acceptable cost (Broh, 1982). QVM adding value through innovation and change The main objective of any management activity is to improve the performance of a company or business to meet strategic objectives. This paper describes a model that can be used to move closer to an elegant solution. This is a tool for management that enables them to anticipate the direction in which competitive advantage through design should head. Therefore, companies will be empowered to seek new design advantages proactively. As customers control revenue the management of quality is the means to inuence the value for money comparison in the purchase decision. What is more, as levels of global competition increase, companies must actively manage their external and internal perceptions by engaging in three areas of activity: (1) Adjusting external perceptions to reect innovativeness and creation. (2) Managing internal capabilities by the appropriate use of ICT and the media and the companys knowledge and ability. (3) Changing the reality, in terms of behaviors, actions, and symbols (supporting society and environment culture). Companies constantly make their commitment to innovation visible through actions that reinforce the need to outperform what already exists. They highlight the importance of having experienced, effective and capable human resources and levels of performance that enable development, improvement and innovation in the move towards globalization (Abdul-Rahman and Berawi, 2002). A vision for a companys future is identied through careful analysis of the system in which the rm exists and the ability of the company to achieve future growth (Woodhead and McCuish, 2003). The key success factors are the specic indicators that are considered reliable measures. They are determined by an analysis of the ability to meet global competition with rapid innovations that reduce the new product development duration of industrial products and projects (Berawi, 2002b). The alternative approach is to be a trend-setter rather than a follower, but the company must know how to play and change its role by using the talents of various cultures and combinations of competition and cooperation. Deploying technology to successfully organize and share knowledge will remain important. A knowledge development

cycle in an organization consists of knowledge creation, knowledge adoption, knowledge distribution, and knowledge review and revision phases (Bhatt, 2000). The challenge for a company is to create an environment that demands and allows generation and processing of information continually as well as facilitating knowledge creation and utilization. This is especially true for rms operating in highly competitive environments (Nonaka and Takeuchi, 1995; Malhotra, 2001). Value added in most industries today is in the form of knowledge, not stuff. The ability of a company to adapt to changes at the right time is important for a company to survive and maintain its competitive advantage. What this paper offers is a tool that will help co-ordinate knowledge. The optimization of intangible assets, such as in-house knowledge are the why and how the company makes progress towards higher level of quality. There is a direct and different inuence on design considerations and market analysis to achieve higher quality. Companies need superior quality to face global competition, yet many classical quality management theories do not promote innovation and change in a dynamic fashion but rather as incremental and continuous tweaks. The attitude to work in businesses and industries drives the attitude to design and the search for innovation. Quality management in business is a system that deals with the procedures of obtaining quality levels that perform intended functions, and to do so within the various human, social, and environmental requirements and constraints. Mathematical model for QVM Given we are not so much interested in achieving a comparable level of quality as offered by rivals, we now explain how we can anticipate quality improvements that enable competitive advantages. Our goal is to identify these factors that really make a difference so that we can ignore insufcient ones. The success of a product can be evaluated by the degree to which it meets or exceeds the customers requirements. Achieving this requires not only the resources of a number of organizations and individuals, but also successful interaction amongst those parties. The concept of QVM has arisen in recognition of the growing customer use of the Internet to source products and the reducing knowledge producers have of what rivals offer. The basic idea for a mathematical approach to an optimal design is to increase all of the products strength factors and at the same time minimize all of its weaknesses. The QVM concept can be applied to all kinds of product and business process innovations as they are essentially different forms of design. The product needs an optimal design and allows us to consider what is as well as what could be (Ilyas et al., 2001). The quality of a product or process is measured in terms of its characteristics or properties that customers value. Weight determined by customers via market research in actual episodes using paired comparison technique (Woodhead and Downs, 2001) allows the explicit link to the customers preferential power over revenue streams to be included in our contemplation. Let me explain the funding theories and move forward to a more detailed explanation of the model. As customers evaluate the worth of an offer the manufacturer must consider the following: . the ability to sell is dependent on the customers perceived worth; and . the customers perceived worth is a product of the amount of quality in a product in relation to the price. From this point, we can show the trends shown in Figure 1.

Quality revolution

427

IJQRM 21,4

428
Figure 1. Customer-perceived worth and ability to sell

Given price is set by the marketplace, if we hold a view of perfect knowledge, then the task before the manufacturer is to increase quality whilst holding price in a predetermined relationship. Further it is the aim of this paper to enable the value of rival offerings to be positioned in relation to new product development. By overlaying two comparable offerings to the market place we can establish the point at which customers would be indifferent as shown in Figure 2. By accepting point O as the minimal quality level our new product must achieve allows us to consider the range of quality it is within our control to deliver. Figure 3 uses an inuence diagram to make our structure of the interrelationships explicit. In this model we are overwhelmed by variables some of which may turn out to be insignicant. The quality of a product (Q) is a result of the quality of its internal properties (q) that customers use to evaluate value for money (i.e. customers perceived worth). For example the difference between a luxury car and an economy car is to be found through the relationship between price and the quality of secondary properties such as upholstery, comfort, acceleration and so on. Each propertys quality is also the result of various inuential factors (x) such as strength, color, texture, smell, etc. These inuential factors are affected by relationships to other moderating variables (v) such as perishability, unwanted heat gains, poor color stability. As such we can set about searching for advantage and direct our new product development process if we can anticipate which aspects of design are more attractive to customer-perceived worth. This will now be explained through a mathematically based approach named quality value model.

Figure 2. Customers point of difference

Quality revolution

429

Figure 3. The quality value model

A. Deterministic model To begin our explanation of how we seek to measure quality we start with a deterministic approach that will subsequently be changed into a probabilistic one. Our experimentation starts with the properties of a product: Q fq1 ; q2 ; q3 ; . . .; qn ; . . .; qN }; where: qn N = nth property of product; = number of properties considered. 1

Equation (1) shows quality as set of all properties and here we allow for a greater number of properties than are shown in the inuence diagram. A products properties are the result of control factors within: qn {x1 ; x2 ; x3 ; . . .; xn ; . . .; xN }; where: x xn N = customer recognition of feature or function; = ith control factor; = number of control factor; 2

xi min , xi , xi max = the performance range of x. A single property (q) within a product is the realization of phenomena that exist between upper (xi_max) and lower boundaries (xi_min). These phenomena are an aggregation of relationships between other variables (v) that can be positive or negative. These other relationships are caused by variables such as ambient temperature, pressure, reliability, and are termed moderating attributes and

IJQRM 21,4

modeled as v in the inuence diagram in such a way that another boundary set is established: xi v1 ; v2 ; v3 ; . . .; vn ; . . .; vN }; where: 3

430

vn N

= nth moderating attribute; = number of moderating factor;

Vi unfav , Vi , Vi fav = the range of effect caused by moderating variables. Given some attributes are qualitative and others such as weight are in different units we need to translate the attributes into a common format (Woodhead and McCuish, 2003). By positioning statements that correspond to percentiles makes this easily achieved. From equations (1)-(3) we can derive a quality target function that has mathematically represented relationship, which we want to achieve as: Q Opt X ; V : 4

In other words, an optimal quality is achieved when a balance is found for the properties within the boundaries of control and moderating factors. The value before optimization is used to determine which factors are inuential to the performance characteristic of interest given not all factors will inuence quality. Hence: X inf X and V inf # V ; where: Xinf = number of inuential control factors; Vinf = number of inuential moderating factors. Then formula (4) may be rewritten to show that we are only interested in inuential control and moderating factors, and this becomes: Q OptX inf ; V inf : 6 5

The mathematical model for one quality characteristic (i.e. one property) and three inuential control factors that leads to a seventh permutation may be written as:
inf inf inf inf q fa axinf 1 ax2 ax3 ax1 x2 inf inf inf inf inf inf axinf 1 x3 ax2 x3 ax1 x2 x3 }:

We now move on to consider two potential designs (AB) as shown in Parettos area in Figure 4. A and B represent the upper and lower boundaries for hypothetical optimal solutions that reside on the boundary line between these two points. In this scenario we dene optimum as the minimization of both q1 and q2 (see point C). Due to our goal here to maximize quality and minimize cost and we are considering levels to set for two properties. In other situations such as if q1 were design style and q2 were design cost we might seek to maximize q1 and minimize q2. The important point here is that

Quality revolution

431
Figure 4. Parettos area (AB) for two quality characteristics, q1 and q2

we accept the relationship being considered is with respect to overall quality. Figure 4 shows that design solution C is better than the design solution D because point C C D D C D C D ( qC 1 ,q2 ) , D (q1 ,q2 ), i.e. q1 , q1 and q2 , q2 . So we now have a deterministic way of nding an optimal level for properties. B. Probabilistic model We now build on the deterministic model by introducing probability. Equations (8) and (9) are adaptations of equations (1) and (2). The reason we do so is to develop a means of predicting sources of advantage from a given product design: Q fP q1 ; P q2 ; P q3 ; . . .; P qn ; . . .; P qN }; P q fP x1 ; P x2 ; P x3 ; . . .; P xn ; . . .; P xN }: 8 9

The optimal value of xi is determined by the optimization of the set v (i.e. moderating factors that detract value). This is dened by: P x fP v1 ; P v2 ; P v3 ; . . .; P vn ; . . .; P vN }: 10

Equation (10) examines the effects of the factors and allows a company to predict the optimal value of the systems performance where P is assumed to be between low and high levels of performance in one property. The predicted values of properties assume a linear relationship in which a customer aggregates benets in the value for money evaluation as shown in equation (11): Y opt e0 e1 x1 e2 x2 e3 x3 . . . en xn =N ; where: e0 = sum of all responses; en = effect of nth factor; xn = coded level of nth factor; N = number of test; n = number of factor. 11

IJQRM 21,4

From equation (11) we can calculate the change in the response value with equation (12) by: Xn where:
new

r 2 r r 2 =2=r 2 r 2 =2;

12

432

Xn_new = new nth factor coded level; r r+ r2 = new level factor natural value; = new high level factor value; = new low level factor value.

The above equations will shortly be demonstrated through a case study. Implementation. Specications and parameters were identied as having most effect on the quality and performance of a product. If the number of the factors is two to three, four tests are required. If the number of the factors is four to seven, eight tests are required. If the number of the factors is eight to 15, 16 tests are required, and so on. The number of tests required is thus twice the lowest number of factors being considered. Furthermore, this paper describes a business performance evaluation model and discusses how quality value model (QVM) can be used to dene the best use of resources to meet the overall business strategy and deal with change and uncertainty within the venture. Tables I and II use three input factors, namely, material costs (f1), man-hours (f2), and number of sub-contractors (f3). We can utilize the intelligence and knowledge based on past experience and experiment as input factors in QVM. The result aims to achieve higher quality, prot and reduce schedule. The mathematical technique in QVM calculates for any input factor changes within the project. This enables companies to respond more quickly because of better ability to predict change and manage decision uncertainty.
Test 1 2 Table I. The calculation of response 3 4 f1 2 (a) + (d) 2 (g) + (j) f2 2 (b) 2 (e) + (h) + (k) f3 + (c) 2 (f) 2 (i) + (l) Response m n o p

Factors Table II. Input factors Cost material Man hours (schedule) Number of sub-contractors

Minimum/low level (2 ) 10 million 50 days 5 sub-contractors

Maximum/high level (+) 40 million 300 days 20 sub-contractors

Case study with three parameters input If the number of any type of factor, e.g. moderating or control factor, is two to three, the calculation of the response is described as shown in Table I. This assumes an interrelationship between the factors which means if one changes the others must also change. Calculation of the effect of the factors (e) is dened by: e0 m n o p; e1 2m n 2 o p; e2 2m 2 n o p; e3 m 2 n 2 o p: The optimum response using equation (11) is: Y opt e0 e1 x1 e2 x2 e3 x3 . . . en xn =N : We will now enter data to examine the approach in operation. Input. Describe value of a whole project experienced or predicted (see Table II). The objective is to reach high quality, prot and as short a time as possible for the project (see Table III). From Tables I-III, the optimum levels and importance of the factors can be determined in order to see the improvement of quality, prot and time and if there are any changes and uncertainty in the input factor in the following cases. Objective function. Objective function 1 is to maximize prot. Calculation of the effects of the factors is as follows: e0 7 4 6 2 19; e1 27 4 2 6 2 27; e2 27 2 4 6 2 23; e3 7 2 4 2 6 2 21: It is used in Table IV to determine the most important factors from data above.
Experiment (test) 1 2 3 4 f1 2 + 2 + f2 2 2 + + f3 + 2 2 + Prot (million) 7 4 6 2 Quality (%) 65 75 55 80 Time (months) 40 30 25 20

Quality revolution

433

Table III. Results from past experience

Importance range 1 2 3

Factor f1 f2 f3

Effect (e) 27 23 21

Optimum level 10 million 50 days 5 sub-contractors Table IV. The most important factors on prot

IJQRM 21,4

Table IV shows that the most inuential factor on prot output is material costs, followed by man-hours and the number of sub-contractors. The optimum response using equation (11) is: Y opt e0 e1 x1 e2 x2 e3 x3 . . . en xn =N ;

434

where: Yopt = prot optimization; e x = effect of the factors; = boundary level from Table I.

To predict the result from the test 2 and compare with the historic data (see Table III), then: Y opt2 19 2 7x 2 3x 2 1x=4 19 2 71 2 321 2 121=4 16=4 4 million The value of prediction is the same as the historic value in Table III, which processes the validity of the equation. Next we will predict the optimum level for prot that is achievable. Y opt 19 2 721 2 321 2 121=4 30=4 7:5 The value of the optimum prot output level that can be achieved in this project is 7.5 million based on the assumed constraints. Now we consider the effect of uncertainty and predict based on changed factor. If there are any changes bringing uncertainty to the input factor, in this case the value of factor 2 changes to 100 days, the transformation from previous value for factor 2 gives: Xn
new

r 2 r r 2 =2=r 2 r 2 =2;

100 2 300 50=2=300 2 50=2 275=125 20:6

Using this one gets:

19 2 721 2 320:6 2 121=4 28:8=4 7:2

Quality revolution

When factor 2 changes to 100 days, the prot output reduces from 7.5 million to 7.2 million. By understanding and managing the factor changes and uncertainty in outcome, the company can predict the sum of prot that can be obtained through the manipulation of factors. Using probability as in equations (8)-(10) we can assess design targets a priori. Quality. We now apply the same logic to quality (see Table I). Here quality is measured in percentiles of customer satisfaction gleaned from market research. Calculation of the effects of the factors is: e0 65 75 55 80 275; e1 265 75 2 55 80 35; e2 265 2 75 55 80 25; e3 65 2 75 2 55 80 15: It is used in Table V to determine the most important factors from data above. Table V shows that the most inuential factor on prot output is material costs, followed by the number of sub-contractors and nally man-hours. We will now predict the optimum level of quality: Y opt 275 351 2 521 151=4 330=4 82:5 percent The value of the optimum quality output level that can be achieved in this project is 82.5 percent. We will now consider the sensibility to change and its impact on quality. If there are any changes bringing uncertainty to the input factor, in this case the value of factor 1 changes to 20 million, the transformation from previous value for factor 1 gives: X 20 2 40 10=1=40 2 10=1 230=30 21

435

Importance range 1 2 3

Factor f1 f3 f2

Effect +35 +15 25

Optimum level 40 million 20 sub-contractors 50 days Table V. The most important factors on quality

IJQRM 21,4

We will now use this quality optimum equation to get: Y 275 3521 2 521 151=4 260=4 65 percent

436

When the changes from factor 1 change from 40 million to 20 million, the quality output will reduce from 82.5 percent to 65 percent. This illustrates the difference between quality and value and the relationship between product, purpose and cost. Time constraint. Objective function 1 is to improve time constraint. Calculation of the effects of the factors is: e0 40 30 25 20 115; e1 240 30 2 25 20 215; e2 240 2 30 25 20 225; e3 40 2 30 2 25 20 5: It is used in Table VI to determine the most important factors from data above. Table VI shows that the most inuential factor on prot output is man-hours, followed by material costs and the number of sub-contractors. We will now predict the optimum level of time: Y opt 115 1521 2521 2 51=4 70=4 17:5 The value of the optimum time output level that can be achieved in this project is 17.5 months. We will now consider the sensibility to change and its impact on time. If there are any changes bringing uncertainty to the input factor, in this case the value of factor 3 changes to ten sub-contractors, the transformation from previous value for factor 3 gives: X 10 2 20 5=3=20 2 5=3 1:66=5 20:332
Importance range Factor f2 f1 f3 Effect +25 +15 25 Optimum level 300 days 40 million 5 sub-contractors

Table VI. The most important factors on quality

1 2 3

We will now use this time optimum equation to get: Y 115 1521 2521 2 50:332=4 73:35=4 18:33

Quality revolution

437
When the changes from factor 3 change from 20 sub-contractors to ten sub-contractors, the time schedule output will change to 18.33 months. Conclusion The QVM in this paper is offered as a means to help a company to face the globalization and one world community. QVM produces a synergy between technology and behavioral issues and human innovation that is necessary to compete and survive in the challenging global marketplace of the new world. The optimization of intangible assets, such as in-house knowledge are the why and how the company makes progress towards high quality. The basic idea for a mathematical approach to an optimal design is to increase all of the products strength factors and at the same time minimize all its weaknesses. Furthermore, the mathematical technique in QVM also calculates for any input factor changes within the project. This enables companies to be more responsive and predictable in respect to change and the management of decision uncertainty. Once having set up the model it becomes a routine simplex process to establish the optimal levels in individual qualities to maximize the quality. Therefore, the concept of QVM has arisen to ensure that companies can plan their quality improvement, anticipate their rivals development plan and nally develop competitive advantage by better knowledge management.
References Abbott, L. (1955), Quality and Competition, Columbia University Press, New York, NY. Abdul-Rahman, H. and Berawi, M.A. (2001), Developing Knowledge Management for Construction Contract Management, Prolog Association of Japan, Tokyo, pp. 358-78. Abdul-Rahman, H. and Berawi, M.A. (2002), Managing change in construction contracting, Contract Management, Vol. 42, pp. 10-16. Berawi, M.A. (2002a), Developing a knowledge system, Proceedings of the 20th Annual Association of Management and International Association of Management (AoM/IAoM) Conference, Canada. Berawi, M.A. (2002b), A new system for quality management in globalization towards innovation and competitive advantages, paper presented at the 3rd International Conference on Decision Making in Urban and Civil Engineering, London. Bhatt, G. (2000), Organizing knowledge in the knowledge development cycle, Journal of Knowledge Management, Vol. 4 No. 1, pp. 15-26. Broh, R.A. (1982), Managing Quality for Higher Prots, McGraw-Hill, New York, NY. Crosby, P.B. (1979), Quality Is Free, New American Library, New York, NY. Deming, W.E. (1982), Quality, Productivity and Competitive Position, MIT/CAES, Boston, MA. Garvin, D.A. (1998), Managing Quality: The Strategic and Competitive Edge, The Free Press, New York, NY.

IJQRM 21,4

438

Ilyas, T., Tamjis, M.R. and Berawi, M.A. (2001), Optimizer, Applied Science Fair, Kuala Lumpur. Ishikawa, K. (1985), What Is Total Quality Control?, (Lu, D.J. (trans.)), Prentice-Hall, Englewood Cliffs, NJ. Juran, J.M. (1974), Quality Control Handbook, McGraw-Hill, New York, NY. Malhotra, Y. (Ed.) (2001), Knowledge Management and Business Model Innovation, Idea Group Publishing, Hershey, PA. Nonaka, I. and Takeuchi, H. (1995), The Knowledge-Creating Company. How Japanese Companies Create the Dynamics of Innovation, The Oxford University Press, New York, NY. Pirsig, R.M. (1974), Zen and the Art of Motor-cycle Maintenance, Bantam Books, New York, NY. Woodhead, R.M. and Downs, C.G. (2001), Value Management: Improving Capabilities, Thomas Telford Ltd, London. Woodhead, R. and McCuish, J. (2003), Achieving Results: How to Create Value, Thomas Telford Ltd, London.