Application of consistent fuzzy preference relations in predicting the success of knowledge management implementation

Decision Support

Application of consistent fuzzy preference relations

in predicting the success of knowledge

management implementation

Tien-Chin Wang

, Tsung-Han Chang

a

Institute of Information Management, I-Shou University, 1, Section 1, Hsueh-Cheng Road, Kaohsiung 840, Taiwan, ROC

b

Institute of Information Engineering, I-Shou University, 1, Section 1, Hsueh-Cheng Road, Kaohsiung 840, Taiwan, ROC Received 19 October 2005; accepted 6 September 2006

Available online 4 December 2006

Abstract

The implementation of knowledge management (KM) involves innovation and reformation for organizations. KM implementation requires not only a substantial investment, but also changes the organization culture and structure. Before embarking on KM, thorough planning is crucial to ensure the implementation achieves the intended objectives of accruing profit and enhancing competitiveness for organisations. Therefore, this study proposes an analytic hierarchical prediction model based on the reciprocal additive consistent fuzzy preference relations to help the organizations become aware of the essential factors affecting the KM implementation, forecasting the chance of successful KM initiative, as well as identifying the actions necessary before implementing KM. Pairwise comparisons are used to determine the priority weights of influ-ential factors and the ratings of two possible outcomes (success and failure) amongst decision makers. The subjectivity and vagueness in the prediction procedures are dealt with using linguistic terms quantified in an interval scale [0, 1]. By mul-tiplying the weights of influential factors and the ratings of possible outcomes, predicted success/failure values are obtained to enable organizations to decide whether to initiate knowledge management, inhibit adoption or take remedial actions to increase the possibility of successful KM project. This proposed approach is demonstrated with a real case study assessed by eleven evaluators solicited from a Liquid Crystal Display (LCD) manufacturing corporation located in Taiwan.  2006 Elsevier B.V. All rights reserved.

Keywords: Decision analysis; Knowledge management; Reciprocal additive consistent fuzzy preference relations; Analytical hierarchy process

1. Introduction

With the advent of the knowledge economy, knowledge itself has become a strategic asset, as well as the main source of organizational competitive predominance. The tangible assets reduce by degrees, whereas the intangible assets increase to improve the organizational innovation capability[22]. As stated in[21], knowledge derived from power is a resource for learning things, reserving valuable heritage, solving problems, creating core competitiveness,

0377-2217/$ - see front matter  2006 Elsevier B.V. All rights reserved. doi:10.1016/j.ejor.2006.09.039

* _{Corresponding author. Tel.: +886 7 6577711x6568; fax: +886 7 6577056.}

E-mail addresses:tcwang@isu.edu.tw,123@king.idv.tw(T.-C. Wang),joan@kyvs.ks.edu.tw(T.-H. Chang).

and initiating new situations for both individuals and organizations now and in the future. During recent decades, the core of the organization is moving from being labour or capital intensive to being technology intensive and cur-rently is moving towards becoming knowledge intensive. Knowledge related to manufacturing, customers, previous success and failure, service, and products are assets that can produce a long-term and sustained competitive advan-tage for organizations[20,23,28]. Numerous enterprises expect to gain the ability to manage their knowledge and strengthen existing advantages by initiating knowledge management in their business operations[26,27,29,32]. The decision regarding whether to implement KM is difficult for many organizations. The sustainable subsistence or downfall of an organization could be based on this decision, and consequently it is important to consider all imple-mental perspectives of an organization before a consensus is achieved regarding implementation. Although success-ful KM implementation cases have been widely reported, such as Microsoft, Samsung[4], Buckman Lab, Boeing 707, Oticon, Affaers-Vaeriden, GE, etc., several examples of failure have also occurred around the world.

Many influential factors determine the success of KM implementation. The factors that require consideration include not only financial issues, but also organizational culture and harmony, problems in integrating the new operational process and old, human relationships, effectiveness of strategic management, CEO’s character and vision, definition of new roles in the organization and many others[1–3,7,8,18,19,25,33]. The main challenge of KM implementation is integrating the above factors with organizational and personnel constraints and capabil-ities. The major goal of KM implementation is frequently to accrue maximum benefit and achieve competitive-ness[24], therefore KM acts as a stimulus forcing the organization to change its practices. Specifically, before the organization can realize the benefits associated with knowledge management initiatives, a fundamental question must be asked: ‘‘What influences must be considered to ensure that knowledge management implementation can successfully help the organization increase revenues, improve profit and achieve competitive competency?’’ Addressing this question help the organizations determine what critical points should be covered.

KM implementation takes time for its impact on the organization to be fully felt. Therefore, a predicted chance of successful implementation and an effective decision approach can facilitate decision making on KM implementation. The proposed prediction model based on the reciprocal additive consistent fuzzy pref-erence relations[17] in this work can help organizations identify key factors affecting KM implementation, and remedial actions then can be taken to ensure successful implementation. This model not only provides a checking mechanism, but also helps in analyzing the organizational ability by considering key success factors for KM implementation. Many internal and external factors affecting the success of KM implementation indi-cates that the issue is a multiple attribute decision making problem. The reciprocal additive consistent fuzzy preference relation is applied to solve such a problem in this study.

The next section discusses the reciprocal additive consistent fuzzy preference relation. An analytic hierarchy framework based on the additive reciprocity transitivity for predicting knowledge management implementa-tion is derived in Secimplementa-tion3. In Section4, an empirical case of KM initiative in Taiwan is presented. Finally, discussion and conclusions are given in Sections5 and 6.

2. Reciprocal additive consistent fuzzy preference relation

Herrera-Viedma et al.[17]proposed the consistent fuzzy preference relations for establishing pairwise com-parison preference decision matrices using the so-called reciprocal additive transitivity property. This method not only enables decision makers to express their degree of preference for a set of attributes or alternatives, but also avoids checking the inconsistency in the decision making process.

The following briefly describes some definitions and propositions presented in[5,6,9–17,30,31,34]. The basic definitions and propositions below are used throughout this study unless otherwise specified.

2.1. Multiplicative preference relation

Definition 2.1. A multiplicative preference relation A on a set of alternatives X is indicated by a matrix A X · X, A = (aij), aijdenotes the ratio of the preference intensity of alternative xito that of xj, A is assumed

multiplicative reciprocal, that is

2.2. Reciprocal additive transitivity fuzzy preference relation

Definition 2.2. Suppose a fuzzy preference relation P on a set of alternatives X is denoted by a matrix P X · X, this is presented by a membership function: lp: X· X ! [0,1], P = (pij), pij= lp(xi, xj)

"i, j2{1, . . . , n}. pij is interpreted as the preference degree of the alternative xi over xj. If pij¼1₂ implies there is no difference between xi and xj (xi xj), pij= 1 indicates xi is absolutely preferred to xj, similarly

pij= 0 indicates xjis absolutely preferred to xi, pij>21indicates that xiis preferred to xj(xi> xj). P is assumed

additive reciprocal, that is

p_ijþ pji ¼ 1 8i; j 2 f1; . . . ; ng: ð2Þ

Proposition 2.2. Suppose there is a set of alternatives X = {x1, . . . , xn}, which is associated with a multiplicative

preference relation A = (aij) with aij2 ½1₉;9. Then the corresponding reciprocal additive fuzzy preference relation

P = (pij) with pij2 [0, 1] to A = (aij) is defined as follows:

p_ij¼ gðaijÞ ¼

1

2 ð1 þ log9aijÞ: ð3Þ

2.3. Additive transitivity consistency of fuzzy preference relation Definition 2.3. Additive transitivity property given by

p_ij1 2   þ pjk 1 2   ¼ p_ik1 2   8i; j; k; or equivalently, p_ijþ pjkþ pki¼ 3 2 8 i; j; k: ð4Þ

Proposition 2.3. Let A = (aij) be a consistent multiplicative preference relation, then the corresponding reciprocal

additive fuzzy preference relation P = g(A), verifies additive transitivity property.

Proof. For being A = (aij) consistent we have that aijÆ ajk= aik "i, j, k, or aijÆ ajkÆ aki= 1 "i, j, k. Take an

action of logarithms on both sides, we obtain log₉aijþ log9ajkþ log9aki¼ 0 8i; j; k:

Adding 3 and dividing by 2 on sides to that 1 2 ð1 þ log9aijÞ þ 1 2 ð1 þ log9ajkÞ þ 1 2 ð1 þ log9akiÞ ¼ 3 2 8i; j; k: The fuzzy preference relation P = g(A), being pij ¼12 ð1 þ log9aijÞ verifies

p_ijþ pjkþ pki¼

3

2 8i; j; k:

Undoubtedly, we can conclude that P = g(A) verifies additive transitivity property. h Definition 2.4. A reciprocal additive fuzzy preference relation P = (pij) is consistent if

p_ijþ pjkþ pki¼

3

Proposition 2.4. For a reciprocal additive fuzzy preference relation P = (pij), the following statements are equivalent: ðaÞ p_ijþ pjkþ pki¼ 3 2 8i; j; k; ð6Þ ðbÞ p_ijþ pjkþ pki¼ 3 2 8i < j < k: ð7Þ

Proof. (a)) (b). Obvious (b) ) (a). There are five possible cases to be discussed:

Case 1: i < k < j. Applying reciprocity property and we have that p_ikþ p_kjþ p_ji ¼3

2; pijþ pjkþ pki¼

1 p_jiþ 1  p_kjþ 1  p_ik¼ 3  ðp_jiþ p_kjþ p_ikÞ ¼ 3 3 2¼

3 2.

Case 2: j < i < k. Using reciprocity property and we have that p_jiþ pikþ pkj¼32; pijþ pjkþ pki¼

1 pjiþ 1  pkjþ 1  pik¼ 3  ðpjiþ pkjþ pikÞ ¼ 3 32¼ 3 2.

Case 3: j < k < i, pjkþ pkiþ pij¼ pijþ pjkþ pki¼32.

Case 4: k < i < j, pkiþ pijþ pjk¼ pijþ pjkþ pki¼32.

Case 5: k < j < i, In case 1, we have pikþ pkjþ pji¼32, and use reciprocity then pijþ pjkþ pki¼

1 pjiþ 1  pkjþ 1  pik¼ 3  ðpkjþ pjiþ pikÞ ¼ 3 32¼ 3 2. h

Proposition 2.5. For a reciprocal additive fuzzy preference relation P = (pij), the following statements are

equivalent: p_ijþ pjkþ pki¼ 3 2 8i < j < k; ð8Þ p_iðiþ1Þþ p_{ðiþ1Þðiþ2Þ}þ    þ p_ðj1Þjþ pji ¼ j i þ 1 2 8i < j: ð9Þ

2.4. Construct a consistent fuzzy preference relation

If the preference matrix contains any values that are not in the interval [0, 1], but in an interval [a, 1 + a], a linear solution is required to preserve the reciprocity and additive transitivity, that is F : [a, 1 + a] ! [0, 1]. Therefore, byProposition 2.5, we can construct a consistent fuzzy preference relation P0_{on X = {x}

1, x2, . . . ,

xn, n P 2} from n 1 preference values {p12, p23, . . . , pn1n}, the steps are described in the following:

• Compute the set of preference values B as B¼ fp_ij; i < j^ p_ij 62 fp12; p23; . . . ; pn1ngg; p_ji¼j i þ 1 2  piiþ1 piþ1iþ2    pj1j; ð10Þ a¼ minfB [ fpj 12; p23; . . . ; pn1nggj; ð11Þ P¼ fp12; p23; . . . ; pn1ng [ B [ f1  p12;1 p23; . . . ;1 pn1ng [ :B: ð12Þ

• The consistent fuzzy preference relation P0_{is obtained as P}0_{= F(P)}

f :½a; 1 þ a ! ½0; 1; fðxÞ ¼ xþ a

1þ 2a: ð13Þ

3. Framework for predicting knowledge management implementation

This section builds a hierarchical analysis structure for tackling the problems of predicting KM implemen-tation using the reciprocal additive consistent fuzzy preference relations. The content comprises four subsec-tions: investigating the influential factors on KM initiative, determining the priority weights of influential

factors, determining the synthetic rating of possible outcomes, and obtaining the priority weights for predic-tion.Fig. 1illustrates the framework for predicting KM implementation.

3.1. Investigating the influential factors on KM implementation

The hierarchical structure for dealing with the problem of forecasting KM implementation is shown in

Fig. 2. The influential factors are derived through widespread investigation and consultation with several experts, including two professors in information management, one professor in information engineering, three

Investigating the influential factors Determining the weights of influential factor Obtaining the ratings of outcome Predicted value Higher chance of success? Implement knowledge management Yes Remedial actions for improvement No No Inhibit adoption

Fig. 1. The framework for predicting knowledge management implementation.

Leadership of CEO Organizational structure Project procedure

Staffs character Audit &

assessment Organizational culture Application of technology Successful KM implementation Failure KM implementation Predict chance of successful or failure KM implementation

professors in business administration and five experienced KM project managers. Synthesizing the literature review from[1–3,7,8,18,19,25,33], the opinions of these experts are utilized to yield the seven key influential factors used in this study. They are described as follows.

• F1-Staff character. This describes staff experience, specialty, ability to create knowledge, recognition of

knowledge management, widespread practice of personnel training, participation, learning aspiration, learning opportunities, and acceptance of information technology. The high intensity of recognition toward knowledge management and high learning incentive represents the high probability of successful KM implementation.

• F2-Project procedure. This task establishes a project group for planning the KM project, and a chief

knowl-edge officer takes charge of the task. The major objective of the KM project is to create unique core com-petency through utilizing the collective knowledge and experience internal and external to the organization whenever and wherever they are required. The organization establishes a team for implementing knowledge management. All team members are responsible for their dominated domains during the implementation process. Furthermore, a system for document management should also be created.

• F3-Organizational structure. The organizational structure is sufficiently flexible to implement KM. An

orga-nizational structure for knowledge management is required to integrate all of the internal even external resources, and to devise implementation strategies. The organization structure should be flat, and should possess the characteristics of being learning-oriented and project-oriented. Open spaces allow staff to con-duct mutual discussions for brainstorming.

• F4-Leadership of superintendent. The CEO is definitely aware of the intended goals and visions of KM

implementation. Superintendents encourage the staff to solve problems and create knowledge via KM, and placing importance on the effectiveness of KM. Senior management should provide critical support for knowledge management initiatives.

• F5-Audit and assessment. This is related to the organization establishing a project flow audit and assessment

management system. The organization provides a clear system or rewards and punishments to assist in KM implementation. Rewards are frequently offered to individuals who perform their duties faithfully, while punishments are administered to those who do not endeavour in knowledge management.

• F6-Organizational culture. This refers to mutual communication, negotiation, cooperation, learning

atti-tude, and trust among organization members. Members are glad to share core knowledge with others, and a common consensus exists regarding the pursuit of knowledge-based core competency. The staffs are empowered to share the interpersonal and intermediate departmental knowledge. Additionally, a ser-vice-oriented culture should be emphasized to improve human and customer relationships.

• F7-Application of technology. This factor includes IT personnel ability, the budget available for establishing

the IT infrastructure, the ability to apply IT management, the use of Internet and Intranet, and human sources of information technology.

3.2. Determining the priority weights of influential factors

The influential factors on predictions relating to KM implementation have different meanings; and not all of them can be assigned equal importance. Since it is easier and more humanistic for evaluators to present ‘‘factor Y is absolutely more important than factor Z’’ than to express ‘‘the relative importance of factor Y and factor Z is nine to one’’. Therefore, this study provides the evaluators simple linguistic terms quantified on a scale of½1

9;9 to express their strength of preference among influential factors.

3.2.1. Linguistic variables

Five linguistic terms, namely ‘‘equally important’’, ‘‘weakly more important’’, ‘‘strongly more important’’, ‘‘very strongly more important’’, and ‘‘absolutely more important’’, are provided for comparing neighboring factors corresponding to a real number (seeTable 1). Additionally, linguistic variables such as ‘‘very high’’, ‘‘high’’, and ‘‘fair’’ are simultaneously used to measure the likelihood of success/failure with respect to each influential factor (seeTable 2).

KM initiative strategies, and enhancing the organizational structure to meet the requirements of the KM implementation process, such as flattening the organizational structure or making the structure learning or project oriented. The predicted values reveal that the chance of successful KM implementation (0.655) is roughly twice than that of failure KM implementation (0.345). This study provides a compromised suggestion that the company should decide to initiate knowledge management and simultaneously implement some reme-dial improvement policies to enhance the poor performance perspectives of this incorporation for increasing the possibility of successful KM implementation.

6. Conclusions

For enterprises and organizations, growing revenues, increasing profits, improving customer service, short-ening product-manufacturing cycle and enhancing competitive competency are cited as objectives for motivat-ing knowledge management initiatives. Since KM implementation is a long-time continuum and its impact is not immediate, a predicted value of successful implementation is required for making decision on whether to initiate KM. To cope with qualitative influential factors in subjective environments, linguistic variables trans-formed into an interval [0, 1] are employed to derive the priority weights of key influential factors and the pre-dicted weights of successful/failure KM implementations.

The proposed approach is based on the reciprocal additive consistent fuzzy preference relation, rather than using conventional multiplicative preference relation. Namely, this method considers only n 1 judgments, whereas the traditional analytic hierarchy approach (that is AHP or FAHP) usesnðn1Þ₂ judgments in a prefer-ence matrix with n attributes or alternatives, it is clear that the proposed approach is faster to execute and more efficient than the conventional analytic hierarchy methodologies. This advantage can be considered as one of the contributions of this paper. Furthermore, an empirical KM implementation case involving a LCD manufacturing corporation located in Taiwan is used to demonstrate the implementation of this approach. Besides fulfilling an examining role in helping organizations to gain awareness of their weaknesses in prediction processes, this study also provides decision makers with useful information to make decision regarding whether to initiate KM, inhibit adoption or undertake some remedial improvement actions to increase the possibility of successful KM implementation. The empirical results not only demonstrate that organizational culture, application of technology and leadership of superintendent are the three most impor-tant influential factors in the KM initiative process, but also reveal the applicability and feasibility of recipro-cal additive consistent fuzzy preference relation for solving complicated hierarchirecipro-cal multiple attribute prediction problems. Subsequently, organizations or enterprises can apply the proposed prediction model to enhance their decision making process and take proper actions to avoid pitfalls (waste of time and money) before KM initiatives.

References

[1] K. Ayas, Professional project management: A shift towards learning and a knowledge creating structure, International Journal of Project Management 14 (3) (1996) 131–136.

[2] K.A. Bates, S.D. Amundson, R.C. Schroeder, W. Morris, The crucial interrelationship between manufacturing strategy and organizational culture, Management Science 41 (10) (1995) 1565–1581.

[3] J.M. Bloodgood, W.D. Salisbury, Understanding the influence of organizational change strategies on information technology and knowledge management strategies, Decision Support System 31 (2001) 55–69.

[4] S.H. Bong, J. Lee, Y. Gil, Effective team processes for technology internalisation with special emphasis on knowledge management: Successful late starter, Samsung case, International Journal of Technology Management 27 (1) (2004) 16–39.

[5] F. Chiclana, F. Herrera, E. Herrera-Viedma, Integrating three representation models in fuzzy multipurpose decision making based on fuzzy preference relations, Fuzzy Sets and Systems 97 (1998) 33–48.

[6] F. Chiclana, F. Herrera, E. Herrera-Viedma, Integrating multiplicative preference relations in a multipurpose decision-making model based on fuzzy preference relations, Fuzzy Sets and Systems 122 (2001) 277–291.

[7] T.H. Davenport, S. Jarvenpaa, M.C. Beers, Improving knowledge work processes, Sloan Management Review 37 (1996) 53–65. [8] T.H. Davenport, G.H. Jeanne, D.W. De Long, A.L. Long Jacobson, Data to knowledge to results: Building an analytic capability,

California Management Review 43 (2) (2001) 117–138.

[9] F. Herrera, E. Herrera-Viedma, Choice functions and mechanisms for linguistic preference relations, European Journal of Operational Research 120 (2000) 144–161.

[10] F. Herrera, E. Herrera-Viedma, Linguistic decision analysis: Steps for solving decision problems under linguistic information, Fuzzy Sets and Systems 115 (2000) 67–82.

[11] F. Herrera, L. Martinez, A model based on linguistic 2-tuples for dealing with multigranular hierarchical linguistic contexts in multi-expert decision-making, IEEE Transactions on Systems, Man, and Cybernetics 31 (2001) 227–234.

[12] F. Herrera, E. Herrera-Viedma, F. Chiclana, Multiperson decision-making based on multiplicative preference relations, European Journal of Operational Research 129 (2001) 372–385.

[13] F. Herrera, E. Herrera-Viedma, J.L. Verdegay, A sequential selection process in group decision making with linguistic assessment, Information Sciences 85 (1995) 223–239.

[14] F. Herrera, E. Herrera-Viedma, J.L. Verdegay, A model of consensus in group decision making under linguistic assessments, Fuzzy Sets and Systems 78 (1996) 73–87.

[15] F. Herrera, E. Herrera-Viedma, J.L. Verdegay, Direct approach process in group decision making using linguistic OWA operators, Fuzzy Sets and Systems 79 (1996) 175–190.

[16] F. Herrera, E. Herrera-Viedma, J.L. Verdegay, A rational consensus model in group decision making using linguistic assessments, Fuzzy Sets and Systems 88 (1997) 31–49.

[17] E. Herrera-Viedma, E. Herrera, F. Chiclana, M. Luque, Some issues on consistency of fuzzy preference relations, European Journal of Operational Research 154 (2004) 98–109.

[18] R. Hull, R. Coombs, M. Peltu, Knowledge management practices for innovation: An audit tool for improvement, International Journal of Technology Management 20 (5-8) (2000) 633–656.

[19] K.C. Lee, S. Lee, I.W. Kang, KPMI: Measuring knowledge management performance, Information and Management 42 (2005) 469– 482.

[20] D. Leonard, S. Sensiper, The role of tacit knowledge in group innovation, California Management Review 40 (3) (1998) 112–132. [21] S.H. Liao, Knowledge management technologies and applications-literature review from 1995 to 2002, Expert Systems with

Applications 25 (2002) 155–164.

[22] J. Liebowitz, K. Wright, Does measuring knowledge make cent? Expert Systems with Applications 17 (1999) 99–103.

[23] A.P. Massey, M. Montoya-Weiss, A performance environment perspective of knowledge management, Proceedings of the 36th Hawaii International Conference on System Sciences (2003).

[24] A.T. Mitra, M. Brown, R. Hackney, Evolutionary knowledge management: A case of system development within the manufacturing industry, International Journal of Technology Management 31 (1–2) (2005) 98–115.

[25] L.T. Ndlela, A.S.A. du Toit, Establishing a knowledge management programme for competitive advantage in an enterprise, International Journal of Information Management 21 (2001) 151–165.

[26] I. Nonaka, A Dynamic Theory of organizational knowledge creation, Organization Science 5 (1) (1994) 14–37.

[27] J.C. Spender, Organizational knowledge, learning and memory: Three concepts in search of a theory, Journal of Organizational Change Management 9 (1996) 63–78.

[28] T. Stewart, The Wealth of Knowledge: Intellectual Capital and the Twenty-first Century Organization, Doubleday, New York, NY, 2001.

[29] J. Swan, S. Newell, M. Robertson, Limits of it driven knowledge management for interactive innovation processes: Towards a community-based approach, in: Proceedings of the 33rd Hawaii International Conference on System Sciences, 2000.

[30] T. Tanino, Fuzzy preference orderings in group decision making, Fuzzy Sets and Systems 12 (1984) 117–131.

[31] T. Tanino, Fuzzy preference relations in group decision making, in: J. Kacprzyk, M. Roubens (Eds.), Non-Conventional Preference Relations in Decision Making, Springer-Verlag, Berlin, 1998, pp. 54–71.

[32] D. Teece, Capturing value from knowledge assets: The new economy, markets for know-how, and intangible assets, California Management Review 40 (3) (1998) 55–79.

[33] Y. Weber, O. Shenkar, A. Raveh, National and corporate cultural fit in mergers/acquisitions: An exploratory study, Management Science 42 (8) (1996) 1215–1228.