This action might not be possible to undo. Are you sure you want to continue?

BooksAudiobooksComicsSheet Music### Categories

### Categories

### Categories

Editors' Picks Books

Hand-picked favorites from

our editors

our editors

Editors' Picks Audiobooks

Hand-picked favorites from

our editors

our editors

Editors' Picks Comics

Hand-picked favorites from

our editors

our editors

Editors' Picks Sheet Music

Hand-picked favorites from

our editors

our editors

Top Books

What's trending, bestsellers,

award-winners & more

award-winners & more

Top Audiobooks

What's trending, bestsellers,

award-winners & more

award-winners & more

Top Comics

What's trending, bestsellers,

award-winners & more

award-winners & more

Top Sheet Music

What's trending, bestsellers,

award-winners & more

award-winners & more

Welcome to Scribd! Start your free trial and access books, documents and more.Find out more

2, 303–314 303

© 2008 Institute of Mathematics and Informatics, Vilnius

A New Logarithmic Normalization Method in

Games Theory

Edmundas Kazimieras ZAVADSKAS, Zenonas TURSKIS

Vilnius Gediminas Technical University

Sauletekio al. 11, LT-10223 Vilnius, Lithuania

e-mail: edmundas.zavadskas@adm.vgtu.lt

Received: April 2007

Abstract. Multi-criteria decision making is used in many areas of human activities. Each alterna-

tive in multi-criteria decision making problem can be described by a set of criteria. Criteria can be

qualitative and quantitative. They usually have different units of measurement and different opti-

mization direction. The normalization aims at obtaining comparable scales of criteria values. The

normalization of criteria values is not always needed, but it may be essential. In the new program

LEVI 3.1 the following normalization methods are possible: vector, linear scale, non-linear and

new logarithmic techniques. Logarithmic normalization has never been used before. The present

research is focused on introducing a new logarithmic method for decision making matrix normali-

zation.

Keywords: games theory, multiple criteria, decision matrix, logarithmic normalization, decision

making in engineering, case study.

1. Introduction

A review of standard decisions made in engineering, management and economy has

shown that the deﬁciency of information is often ignored. Experts not always make use of

the appropriate initial data. The values applied are often exaggerated. Poor quality models

are used which, if required, are slightly corrected based of practical experience. However,

the actual situation is not properly reﬂected and possible effects of external actions are

not known. A decision is often made by comparing costs and beneﬁts of the available

alternative under various environmental conditions.

The evaluation of all possible actions is not always sufﬁcient. Each action may lead to

several, sometimes conﬂicting results. As the actual result is not known, the criteria tak-

ing into consideration all possible results are needed. Therefore, multi-criteria decision

making becomes extremely important. An alternative in multi-criteria evaluation is usu-

ally described by quantitative and qualitative criteria. These criteria have different units

of measurement. Normalization is aimed at obtaining the comparable scales of the crite-

ria values. Different techniques of criteria value normalization are used. Normalization

of the criteria values is not always necessary.

The impact of the decision matrix normalization methods on the decision results has

been investigated by many authors (Weitendorf, 1976; Hwang and Yoon, 1981; Peld-

304 E.K. Zavadskas, Z. Turskis

schus et al., 1983, 2002; Peldschus, 1986, 2001, 2007; Peldschus and Zavadskas, 1997;

Dejus, 2002; Körth, 1969; Stopp, 1975; Jüttler, 1966; Migilinskas, 2003; Migilinskas

and Ustinovichius, 2007; Kaklauskas et al., 2007; Kalibatas et al., 2007; Turskis et al.,

2006; Zagorskas and Turskis, 2006; Ginevicius and Podvezko, 2007; Noorul Haq and

Kannan, 2007; Antucheviciene et al., 2006; Brauers and Zavadskas, 2006; Brauers et al.,

2007; Brauers, 2007a, 2007b; Viteikien˙ e, 2006; Viteikien˙ e and Zavadskas, 2007; Hov-

anov, 1996; Cloquell and Santamarina, 2001; Zavadskas, 1987; Zavadskas et al., 1994,

2002, 2003, 2004, 2006, 2007a, 2007b; Ustinovichius, 2001, 2004, 2007; Ustinovichius

and Zavadskas, 2004; Ustinovichius et al., 2007; Vaidogas and Zavadskas, 2007; Vaido-

gas et al., 2007). The authors of many well-known programs chose one particular problem

solution method or approach to decision-making matrix normalization. There are still no

rules determining the application of multi-criteria evaluation methods and interpretation

of the results obtained.

Vilnius Gediminas Technical University (VGTU) and Leipzig University of Applied

Sciences (HTKW) have been investigating the application of games theory principles to

civil engineering technology and management problems for more than 20 years (Peld-

schus et al., 1983, 2002; Peldschus, 2001, 2007; Peldschus and Zavadskas, 1997; Zavad-

skas, 1987; Zavadskas and Kaklauskas, 2007). The program LEVI 3.0 was a result of

the co-operation between VGTU and HTKW. All calculations were made with LEVI 3.0

(Zavadskas et al., 2002, 2003, 2004; Peldschus et al., 2002; Peldschus, 2007). The pro-

gram LEVI 3.1 was created for evaluating various processes in economics, engineering

and management.

In the new program LEVI 3.1 (Fig. 1), a new logarithmic normalization method is

implemented. This new software allows us to ﬁnd a solution under the conditions of risk

and uncertainty and to compare the results by applying different methods.

Fig. 1. Block-diagram of choosing the best alternative in LEVI 3.1 program.

A New Logarithmic Normalization Method in Games Theory 305

2. Structure and Methodology of the Program LEVI 3.1

In the program LEVI 3.1, the games theory of the discrete optimization problem solution

is used. Any problem to be solved is represented by a matrix, containing the alternatives

(rows) and the criteria (columns). Usually, the criteria have different dimensions. In order

to avoid the difﬁculties caused by different dimensions of the criteria, the ratio to the

optimal value is used. There are various theories describing the ratio to the optimal value.

However, the values are mapped either on the interval [0;1] or the interval [0;∞] by

applying the normalization of a decision-making matrix. In the program LEVI 3.1, only

the widely known and logarithmic normalization methods are used (Table 1).

When the normalization is completed, it is possible to evaluate the criteria with

weighting factors 0 ≺ q

j

≺ 1. The sum of the weighting factors should be equal to 1.

Only well-founded weighting factors should be used because weighting factors inﬂu-

encing the solution are always subjective. Using the Games Theory (von Neumann and

Morgenstern, 1943), the two-sided test aims at ﬁnding the equilibrium as a result of the

rational behaviour of two parties having the opposite interests or searching for the equi-

librium in a game against nature.

Table 1

Normalization methods in the program LEVI 3.1

Normalization

method (NM)

Preferable max a

ij

Preferable min a

ij

Notes

1 Vector (VE)

normalization

(Van Deft and

Nijkamp, 1977)

b

ij

=

a

ij

¸

m

i=1

a

2

ij

b

ij

= 1 −

a

ij

¸

m

i=1

a

2

ij

The ratio of the

values remains

constant for this type

of normalization in

the interval [0;1].

2 Weitendorf’s

(1976) linear

(WL)

normalization

The calculated values

are dependent on the

size of the interval

max

i

a

ij

; min

i

a

ij

b

ij

=

a

ij

−min

i

a

ij

max

i

a

ij

−min

i

a

ij

b

ij

=

max

i

a

ij

−a

ij

max

i

a

ij

−min

i

a

ij

3 Jüttler’s–Körth’s

(1969)

normalization

The application of

this type of

normalization is

limited to the interval

[0;1].

b

ij

= 1 −

max

i

a

ij

−a

ij

max

i

a

ij

b

ij

= 1 −

min

i

a

ij

−a

ij

min

i

a

ij

4 Peldschus et al.

(1983) non-linear

(NL)

normalization

The values are

diminished more than

when using other

methods

b

ij

=

a

ij

max

i

a

ij

2

b

ij

=

min

i

a

ij

a

ij

3

5 New

Logarithmic

(LN)

normalization

The sum of

normalized criterion

values is always equal

to 1.

b

ij

=

ln(a

ij

)

ln

¸

n

i=1

a

ij

b

ij

=

1−

ln(a

ij

)

ln

¸

n

i=1

a

ij

n−1

306 E.K. Zavadskas, Z. Turskis

Wald’s rule (WA): This method is used to search for the best of the worst solutions

(Wald, 1945). The decision-maker acts according to the worst situation occurring (a pes-

simistic attitude):

S

∗

=

¸

S

i

/S

i

∈ S ∩ max

i

min

j

b

ij

¸

. (1)

Savage criterion (SA): The aim is to minimize the loss of appropriateness, which is

the difference between the greatest and the achieved beneﬁt (Savage, 1951):

S

∗

=

S

i

/S

i

∈ S ∩ min

i

max

j

c

ij

∩ c

ij

=

max

r

a

rs

−a

rs

, (2)

where r = 1, m and s = 1, n. A disadvantage of the method is in the presence of non-

optimal strategies affecting the solution.

Hurwicz’s rule (HU): An optimal strategy is based on the best and the worst results

(Hurwicz, 1951). These values, calculated from the row’s minimum and row’s maximum,

are integrated into a weighted average using optimism parameters:

S

∗

=

S

i

/S

i

∈S ∩ max

i

h

i

∩ h

i

=λmax

j

b

ij

+ (1−λ) min

j

b

ij

∩ 0λ1

¸

. (3)

The value λ = 1 gives the most pessimistic solution (Wald’s rule). For the value of λ = 0

only the maximum (the greatest risk) values are considered.

Laplace’s rule (LA): The solution is calculated under the condition that all probabil-

ities for the strategies of the opponent are equal (Bernoulli, 1738):

S

∗

=

S

i

/S

i

∈ S ∩ max

i

1/n

n

¸

i=1

b

ij

. (4)

Bayes’s rule (BA): Given the probabilities for the strategies of the opponent, the

maximum for the expected value can be used (Arrow et al., 1949):

S

∗

=

S

i

/S

i

∩ max

i

n

¸

j=1

q

j

b

ij

∩

n

¸

j=1

q

j

= 1

. (5)

Hodges-Lehmann rule (HL): According to this rule, the conﬁdence in the knowl-

edge of the probabilities of the opponent’s strategies can be expressed by the parameterλ

(Hodges and Lehmann, 1952):

S

∗

=

S

i

/S

i

∈ S ∩ max

i

λ

n

¸

j=1

q

j

b

ij

+ (1 −λ) min

j

b

ij

∩ 0 λ 1

, (6)

where λ = 0 (no conﬁdence) gives the solution according to Wald’s rule, while λ = 1

(great conﬁdence) gives the solution according to Bayes’s rule.

A New Logarithmic Normalization Method in Games Theory 307

3. Normalization Test in the Case of Various Data Intervals

In order to test the described new normalization method, we will consider the normal-

ization of test matrices. The alternatives of the initial data for normalization are designed

according to various distribution laws:

x

1i

= 0 + i; x

2i

= 10 + i; x

3i

= 100 + i;

x

4i

= 1000 + i; x

5i

= 10

(i−1)

; x

6i

= 0.001 · 10

i

, where i = 1, 10. (7)

The values of the initial data are changed depending on particular intervals. The as-

sumption is made that all criteria are maximized. A comparison of the test results is given

in Table 2.

Based on the comparison of normalization results, we can make the following con-

clusion: the application of a new logarithmic normalization method yields the values of

the normalized matrix elements approaching the average values obtained in other nor-

malization methods. In some cases, the obtained values are approximately equivalent to

the values of linear normalization.

Table 2

Comparison of the test results of matrix logarithmic normalization

308 E.K. Zavadskas, Z. Turskis

Table 2 (continued)

Comparison of the test results of logarithmic normalization presented in the matrix

4. A Case Study of Alternatives’ Evaluation Using Various Solution Methods and

Normalization Techniques

To illustrate the application of the described methods, the problem of the selection of

a rational option of the external ﬁnishing for cast-in-place buildings will be considered

(Zavadskas et al., 1994). To select a rational alternative of external wall ﬁnishing in cast-

in-place buildings, a survey of ﬁve technological alternatives was conducted. A13-storey

block-of ﬂats in Justiniskes, a suburb of Vilnius, served as an example of analysis. The

initial data for evaluation of the alternatives are given in Table 3. The ﬁrst alternative deals

with developing a relief surface of concrete by assembling the structures on formwork

when concrete is placed in metal forms. The second alternative deals with the decoration

of facades by open work structural elements of 1.0 × 0.5 m made of gypsum concrete

slabs. The third alternative presents facing of ceramic tiles. The fourth alternative deals

with painting of the balconies with a long-lasting paint. The ﬁfth alternative deals with

painting of the whole facade with a long-lasting paint. The evaluation of the alternatives

was based on the following four attributes: 1) costs x

1

, thousand $; 2) labour input x

2

,

man-days; 3), 4) the criteria 3 and 4 (durability x

3

and aesthetics x

4

) were evaluated

in points by 28 experts. The criteria weights were determined by the method of pairwise

A New Logarithmic Normalization Method in Games Theory 309

Table 3

The initial data for assessing project alternatives and results of decision matrix normalization

comparison based on the estimates of 28 experts. The obtained weight vector of attributes

was q = (0.10, 0.12, 0.42, 0.36).

In the present investigation, the vector, linear, non-linear and new logarithmic meth-

ods of initial decision-making matrix normalization were used. A number of different

problem solution methods, such as Wald’s rule, Savage criterion, Hurwicz’s rule, also

Laplace’s rule, Bayes’s rule, Hodges-Lehmann-rule, were also applied. Tables 4 and 5

provide the solution results and graphical representation of their comparative analysis.

The use of logarithmic normalization improves the quality of decision matrix normaliza-

tion in solving economic and organizational problems.

When the criteria weights are taken into account, the priority order of the alternatives

is presented as v

4

v

3

v

5

v

2

v

1

(implying that the “fourth” alternative is

better than the “third” one, the “third” alternative is better than the “ﬁfth” one, the “ﬁfth”

alternative is better than the “second” one and the “second” alternative is better than the

“ﬁrst” one). A similar set v

4

v

5

v

3

v

2

v

1

is obtained when the criteria weights

are not taken into account in the process of alternative assessment.

310 E.K. Zavadskas, Z. Turskis

Table 4

Ranking of the alternatives

Table 5

Ranks of the project alternatives. Solutions according to Wald, Savage, Laplace and Hurwicz (RF = 0.5), Bayes

and Hodges-Lehmann (λ = 0.5)

Ranks of alternatives

NM q not included q included

WA SA HU LA BA HL

VE 43215 42135 42315 41235 45321 45321

LW 43215 4321=5 4321=5 43215 54213 54213

NL 142=3=5 142=3=5 13=452 45132 45312 45312

LN 43251 45321 45321 45321 45321 45321

A New Logarithmic Normalization Method in Games Theory 311

The results of problem solution are more stable. For example, when the weights of

the criteria are included in calculation, various methods and logarithmic normalization

used in solving the problem determine the alternative 4 as the most effective, while other

normalization methods give different results. If weights of the criteria are not included in

the evaluation process, the most effective option according to the logarithmic normaliza-

tion (e.g., linear normalization) is the alternative 4, while other normalization methods

describe other alternatives as optimal.

According to the results obtained in the analysis, the most effective fourth alternative

was implemented.

5. Conclusions

It is hardly possible to evaluate the effect of various methods of normalization of a

decision-making matrix on the numerical results obtained. These problems can be solved

by applying the program LEVI 3.1.

Some modules of the program LEVI 3.1 can be used for creating decision-making

systems.

Logarithmic normalization of a decision-making matrix has been used for the ﬁrst

time. Logarithmic normalization of a decision-making matrix yields more stable results

in solving multi-criteria decision problems.

The calculations show that logarithmic normalization may be used in the cases when

the values of the criteria differ considerably.

The logarithmic normalization method used in solving the problems segregates nor-

malized values more effectively than other methods.

A comparison of results obtained by different solution methods is needed because it

is not always possible to apply the games theory equilibrium to economics, engineering

and management.

References

Antucheviˇ cien˙ e, J., Z. Turskis, E.K. Zavadskas (2006). Modelling renewal of construction objects applying

methods of the game theory. Technological and Economic Development of Economy, 12(4), 263–268.

Arrow, K.J., D. Blackwell and M.A. Girshick (1949). Bayes and minimax solutions of sequential decision

problems. Econometrica, 17, 213–243.

Bernoulli, D. (1738). Specimen theoriae novas de mesure sortis. Comentarii Academiae Scientarium Inperialis

Petropolitanae, 5, 1750–192. Translated by L. Sommer (1954). Exposition of a new theory on the measure-

ment of risk. Econometrica, 22, 23–36.

Brauers, W.K. (2007a). What is meant by normalization in decision making? International Journal of Manage-

ment and Decision Making, 8(5–6), 445–460.

Brauers, W.K. (2007b). Normalisation in multiobjective optimization: a general overview. International Journal

of Management and Decision Making, 8(5–6), 461–474.

Brauers, W.K., M.R. Ginevicius, E.K. Zavadskas and J. Antucheviciene (2007). The European Union in a

transition economy. Transformations in Business and Economics, 6(2), 21–37.

Brauers, W.K., and E.K. Zavadskas (2006). The MOORA method and its application to privatization in a tran-

sition economy. Control and Cybernetics, 35(2), 443–468.

312 E.K. Zavadskas, Z. Turskis

Cloquell, V.A., and C. Santamarina (2001). A new procedure for the numerical values normalization in multicri-

teria decision techniques. In MCDA 54th Meeting in Durbuy. Universidad Politécnica de Valencia, Valencia.

pp. 1–10.

Dejus, T. (2002). The model of determining the sensitivity of elements of multiple criteria evaluation methods.

Journal of Civil Engineering and Management, 8(4), 263–286.

Ginevicius, R., and V. Podvezko (2007). Some problems of evaluating multicriteria decision methods. Interna-

tional Journal of Management and Decision Making, 8(5–6), 527–539.

Hodges, J.L., and E.L. Lehmann (1952). The use of previous experience in reaching statistical decision. Annals

of Mathematics Studies, 23, 396–407.

Hovanov, N. (1996). Analysis and Synthesis of Parameters under Information Deﬁciency. St. Petersburg Uni-

versity Press, St. Petersburg.

Hwang, C.L., and K.P. Yoon (1981). Multiple Attribute Decision Making – Methods and Applications. Springer-

Verlag, New York.

Hurwicz, L. (1951). Optimality criteria for decision making under ignorance. cowles commission paper. Statis-

tics, 370, 45–52.

Jüttler, H. (1966). Untersuchungen zur Fragen der Operations aforschung und ihrer Anwendungsmöglichkeiten

auf ökonomische Problemstellungen unter besonderer Berücksichtigung der Spieltheorie. Dissertation A an

der Wirtschaftswissenschaftlichen Fakultät der Humboldt–Universität, Berlin.

Kaklauskas, A., E.K. Zavadskas, A. Banaitis and G. Šatkauskas (2007). Deﬁning the utility and market value

of real estate: a multiple criteria approach. International Journal of Strategic Property Management, 11(2),

107–120.

Kalibatas, D., M. Krutinis and M. Viteikien˙ e (2007). Multi-objective evaluation of microclimate in dwelling.

Technological and Economic Development of Economy, 13(1), 24–31.

Körth, H. (1969). Zur Berücksichtigung mehrer Zielfunktionen bei der Optimierung von Produktionsplanen.

Mathematik und Wirtschaft, 6, 184–201.

Migilinskas, D. (2003). The inﬂuence of normalization methods selection in construction including game theory

adaptation. Technological and Economic Development of Economy, 9(2), 73–79 (in Lithuanian).

Migilinskas, D., and L. Ustinovichius (2007). Normalisation in the selection of construction alternatives. Inter-

national Journal of Management and Decision Making, 8(5–6), 623–639.

Neumann von, J., and O. Morgenstern (1943). Theory of Games and Economic Behavior. Princeton University

Press.

Noorul Haq, A., and G. Kannan (2007). A Hybrid normalised multi criteria decision making for the vendor

selection in a supply chain model. International Journal of Management and Decision Making, 8(5–6),

601–622.

Peldschus, F. (1986). Zur Anwendung der Theorie der Spiele für Aufgaben der Bautechnologie. Diss. B. Tech-

nologie. Diss. B. Technische Hochschule Leipzig.

Peldschus, F. (2001). Sensibilitätsuntersuchungen zu Methoden der merhkriteriellen Entscheidungen. Journal

of Civil Engineering and Management (Statyba), 7(4), 276–281.

Peldschus, F. (2007). The effectiveness of assessment in multiple criteria decisions. International Journal of

Management and Decision Making, 8(5–6), 519–526.

Peldschus, F., D. Messing, E.K. Zavadskas, L. Ustinovichius and Z. Turskis (2002). LEVI 3.0 – multiple criteria

evaluation program under uncertainty. Technological and Economic Development of Economy, 8(1), 3–12.

Peldschus, F., E. Vaigauskas and E.K. Zavadskas (1983). Technologische Entscheidungen bei der Berücksich-

tigung mehrerer Ziehle. Bauplanung–Bautechnik, 37(4), 173–175.

Peldschus, F., and E.K. Zavadskas (1997). Matrix Games in Building Technology and Management. Technika,

Vilnius (in Lithuanian).

Savage, L.J. (1951). The theory of statistical decision. Journal of the American Statistical Association, 46,

55–57.

Stopp, F. (1975). Variantenvergleich durch Matrixspiele. Wissenschaftliche Zeitschrift der Hochschule für

Bauwesen Leipzig, 2, 117.

Turskis, Z., E.K. Zavadskas and J. Zagorskas (2006). Sustainable city compactness evaluation of the basis of

GIS and Bayes rule. International Journal of Strategic Property Management, 10(3), 185–207.

Ustinovichius, L. (2001). Determining integrated Criteria signiﬁcance of attributes. Journal of Civil Engineer-

ing and Management (Statyba), 7(4), 321–326.

A New Logarithmic Normalization Method in Games Theory 313

Ustinovichius, L. (2004). Determination of efﬁciency of investments in construction. International Journal of

Strategic Property Management, 8(1), 25–43.

Ustinovichius, L. (2007). Methods of determining objective, subjective and integrated weights of attributes.

International Journal of Management and Decision Making, 8(5–6), 540–554.

Ustinovichius, L., and E.K. Zavadskas (2004). Assessment of investment proﬁtability in construction from the

technological perspective. Technika, Vilnius (in Lithuanian).

Ustinovichius, L., E.K. Zavadskas and V. Podvezko (2007). Application of a quantitative multiple criteria de-

cision making (MCDM-1) approach to the analysis of investment in construction. Control and Cybernetics,

36(1), 251–268.

Van Delft, A., P. Nijkamp (1977). Multi-Criteria Analysis and Regional Decision-Making. M. Nijhoft, Leiden,

Nl.

Vaidogas E.R., and E.K. Zavadskas (2007). Introducing reliability measures into multi-criteria decision-making.

International Journal of Management and Decision Making, 8(5–6), 475–496.

Vaidogas E.R., E.K. Zavadskas and Z. Turskis (2007). Reliability measures in multicriteria decision making as

applied to engineering projects. International Journal of Management and Decision Making, 8(5–6), 497–

518.

Viteikiene, M. (2006). Sustainable residential areas evaluation. Technological and Economic Development of

Economy, 12(2), 152–160.

Viteikiene, M., and E.K. Zavadskas (2007). Evaluating the sustainability of Vilnius city residential areas. Jour-

nal of Civil Engineering and Management, 13(2), 149–155.

Wald, A. (1945). Statistical decisions functions which minimise the maximum risk. Annals of Mathematics, 46,

265–280.

Weitendorf, D. (1976). Beitrag zur Optimierung der räumlichen Struktur eines Gebäudes. Dissertation A,

Hochschule für Architektur und Bauwesen. Weimar.

Zagorskas, J., and Z. Turskis (2006). Multi-attribute model for estimation of retail centres inﬂuence on the city

structure. Kaunas city case study. Technological and Economic Development of Economy, 12(4), 347–352.

Zavadskas, E.K. (1987). Multiple criteria evaluation of technological decisions of construction. Dissertation of

Dr. Sc. Moscow Civil Engineering Institute, Moscow (in Russian).

Zavadskas, E.K., and A. Kaklauskas (2007). Mehrzielselection für Entscheidungen im Bauwesen. Fraunhofer

IRB Verlag.

Zavadskas, E.K., A. Kaklauskas, F. Peldschus and Z. Turskis (2007a). Multi-attribute assessment of road design

solutions by using the COPRAS method. The Baltic Journal of Road and Bridge Engineering, 2(4), 195–203.

Zavadskas, E.K., F. Peldschus and A. Kaklauskas (1994). Multiple Criteria Evaluation of Projects in Construc-

tion. Technika, Vilnius.

Zavadskas, E.K., F. Peldschus and L. Ustinovichius (2003). Development of software for multiple criteria eval-

uation. Informatica, 14(2), 259–272.

Zavadskas, E.K., F. Peldschus, L. Ustinovichius and Z. Turskis (2004). Game Theory in Building Technology

and Management. Technika, Vilnius (in Lithuanian).

Zavadskas, E.K., Z. Turskis, T. Dejus and M. Viteikiene (2007b). Sensitivity analysis of a simple additive weight

method. International Journal of Management and Decision Making, 8(5–6), 555–574.

Zavadskas, E.K., L. Ustinovichius, Z. Turskis, F. Peldschus and D. Messing (2002). LEVI 3.0 – Multiple criteria

evaluation program for construction solutions. Journal of Civil Engineering and Management, 8(3), 184–

191.

Zavadskas, E.K., A. Zakarevicius and J. Antucheviciene (2006). Evaluation of ranking accuracy in multi-criteria

decisions. Informatica, 17(4), 601–617.

314 E.K. Zavadskas, Z. Turskis

E.K. Zavadskas is a principal vice-rector of Vilnius Gediminas Technical University and

a head of the Dept. of Construction Technology and Management at Vilnius Gediminas

Technical University, Vilnius, Lithuania. He has a PhD in building structures (1973) and

Dr Sc (1987) in building technology and management. He is a member of the Lithuanian

and several foreign Academies of Sciences. He is doctore honoris causa at Poznan, Saint-

Petersburg, Kiev. He is a member of international organisations and has been a member of

steering and programme committees at many international conferences. E.K. Zavadskas

is a member of editorial boards of several research journals. He is author and co-author of

more than 300 papers and a number of monographs in Lithuanian, English, German and

Russian. Research interests are: building technology and management, decision-making

theory, automation in design and decision support systems.

Z. Turskis (PhD) is a senior research fellow of Construction Technology and Manage-

ment Laboratory of Vilnius Gediminas Technical University, Lithuania. His research in-

terests include building technology and management, decision-making theory, computer-

aided automation in design and expert systems. He is the author of 20 scientiﬁc papers.

Naujasis logaritminis normalizavimo metodas, naudojamas lošim

u

teorijoje

Edmundas Kazimieras ZAVADSKAS, Zenonas TURSKIS

Daugiakriterinis sprendim

u pri˙ emimas yra naudojamas daugelyje visuomen˙ es veiklos sriˇ ci

u.

Alternatyva daugiakriteriniame vertinime dažniausiai yra apibr˙ ežiama kiekybiniais ir kokybiniais

kriterijais, kurie turi skirtingus matavimo vienetus. Normalizuojant šiuos kriterijus siekiama j

u su-

vienodinimo, kad b¯ ut

u galima palyginti kriterij

**u reikšmes. Nors normalizavimas ne visada reikalin-
**

gas, jis gali b¯ uti labai svarbus kai kuriais atvejais, pavyzdžiui, lošim

**u teorijoje. Naujai sukurtoje
**

programoje LEVI 3.1 yra naudojami tokie normalizavimo metodai: vektorinis, linijinis, nelinijinis

ir naujasis logaritminis. Logaritminis normalizavimas niekada nebuvo panaudotas. Šiame tyrime

pagrindinis d˙ emesys yra skiriamas logaritminio metodo, naudojamo sprendim

**u pri˙ emimo matricos
**

normalizavimui, aprašymui.

1986. 2007. Ginevicius and Podvezko.. Zagorskas and Turskis. 1983. 1975.0 was a result of the co-operation between VGTU and HTKW. 2007. a new logarithmic normalization method is implemented. 2006. Körth.. Kalibatas et al. Ustinovichius et al.304 E. Migilinskas and Ustinovichius. 2006. The program LEVI 3. 1969. Fig. 2007.. Zavadskas. Hove e anov. 1). Antucheviciene et al. 2002. Peldschus. 2004. 2007. 1987. Turskis et al. 2007. Jüttler. The program LEVI 3. 2002. 1997. 2001. Peldschus and Zavadskas. 2007. 2003. 2007). Peldschus et al. 2006. 2007a. Zavadskas and Kaklauskas. 1996. 2007. 2001. Ustinovichius. 2004. There are still no rules determining the application of multi-criteria evaluation methods and interpretation of the results obtained. 1983. 2004. 2007.K.1 program. 2006. 1987..0 (Zavadskas et al. Vaidogas and Zavadskas.. Cloquell and Santamarina.. Peldschus. 2001.1 was created for evaluating various processes in economics. Z.. 2006. Zavadskas et al. Noorul Haq and Kannan.. 2003. Kaklauskas et al. 1997. 2007a. 2007. Brauers and Zavadskas. Stopp. 2007). 2002. Viteikien˙ and Zavadskas. 2002. 2007b. Vaidogas et al. Zavadskas. Ustinovichius and Zavadskas. 1. 2001.1 (Fig. The authors of many well-known programs chose one particular problem solution method or approach to decision-making matrix normalization. Brauers. 2007). All calculations were made with LEVI 3. Dejus. engineering and management. 2006. 2007.. Block-diagram of choosing the best alternative in LEVI 3. 2002. 2007. 1966. Peldschus. Brauers et al.. Migilinskas. Viteikien˙ . Zavadskas. .. Peldschus and Zavadskas. 2002. 2007.. 1994. 2007b. 2004. Vilnius Gediminas Technical University (VGTU) and Leipzig University of Applied Sciences (HTKW) have been investigating the application of games theory principles to civil engineering technology and management problems for more than 20 years (Peldschus et al. In the new program LEVI 3. Turskis schus et al. This new software allows us to ﬁnd a solution under the conditions of risk and uncertainty and to compare the results by applying different methods. 2003.

Using the Games Theory (von Neumann and Morgenstern. the two-sided test aims at ﬁnding the equilibrium as a result of the rational behaviour of two parties having the opposite interests or searching for the equilibrium in a game against nature. only the widely known and logarithmic normalization methods are used (Table 1). However. containing the alternatives (rows) and the criteria (columns). In order to avoid the difﬁculties caused by different dimensions of the criteria.1.∞] by applying the normalization of a decision-making matrix. In the program LEVI 3.1 Normalization method (NM) 1 Vector (VE) normalization (Van Deft and Nijkamp. Any problem to be solved is represented by a matrix. The sum of the weighting factors should be equal to 1.1].1. When the normalization is completed.1]. There are various theories describing the ratio to the optimal value. it is possible to evaluate the criteria with weighting factors 0 ≺ qj ≺ 1. Usually. the values are mapped either on the interval [0. the ratio to the optimal value is used. 1943).1 In the program LEVI 3.1] or the interval [0. (1983) non-linear bij = (NL) normalization 5 New Logarithmic (LN) normalization aij maxi aij 2 bij = mini aij aij 3 1− bij = ln ln(aij ) n i=1 ln bij = ln(aij ) n aij i=1 aij n−1 . the games theory of the discrete optimization problem solution is used. a2 ij a2 ij bij = aij −mini aij maxi aij −mini aij bij = maxi aij −aij maxi aij −mini aij bij = 1 − maxi aij −aij maxi aij bij = 1 − mini aij −aij mini aij 4 Peldschus et al. Only well-founded weighting factors should be used because weighting factors inﬂuencing the solution are always subjective. mini aij The application of this type of normalization is limited to the interval [0. The calculated values are dependent on the size of the interval maxi aij . Structure and Methodology of the Program LEVI 3.A New Logarithmic Normalization Method in Games Theory 305 2. The values are diminished more than when using other methods The sum of normalized criterion values is always equal to 1. 1977) 2 Weitendorf’s (1976) linear (WL) normalization 3 Jüttler’s–Körth’s (1969) normalization Preferable max aij bij = aij m i=1 Preferable min aij bij = 1 − aij m i=1 Notes The ratio of the values remains constant for this type of normalization in the interval [0. Table 1 Normalization methods in the program LEVI 3. the criteria have different dimensions.

1738): S = ∗ n Si /Si ∈ S ∩ max 1/n i i=1 bij . m and s = 1. (6) where λ = 0 (no conﬁdence) gives the solution according to Wald’s rule. the maximum for the expected value can be used (Arrow et al. are integrated into a weighted average using optimism parameters: S ∗ = Si /Si ∈ S ∩ max hi ∩ hi = λ max bij + (1−λ) min bij ∩ 0 λ 1 . These values. 1949): n n S∗ = Si /Si ∩ max i j=1 qj bij ∩ j=1 qj = 1 . i j (1) Savage criterion (SA): The aim is to minimize the loss of appropriateness. while λ = 1 (great conﬁdence) gives the solution according to Bayes’s rule. The decision-maker acts according to the worst situation occurring (a pessimistic attitude): S ∗ = Si /Si ∈ S ∩ max min bij . Zavadskas.K. Laplace’s rule (LA): The solution is calculated under the condition that all probabilities for the strategies of the opponent are equal (Bernoulli. 1951): S∗ = Si /Si ∈ S ∩ min max cij ∩ cij = i j max ars − ars . calculated from the row’s minimum and row’s maximum. which is the difference between the greatest and the achieved beneﬁt (Savage. the conﬁdence in the knowledge of the probabilities of the opponent’s strategies can be expressed by the parameterλ (Hodges and Lehmann. 1952): n S∗ = Si /Si ∈ S ∩ max λ i j=1 qj bij + (1 − λ) min bij ∩ 0 j λ 1 . (5) Hodges-Lehmann rule (HL): According to this rule. For the value of λ = 0 only the maximum (the greatest risk) values are considered. (4) Bayes’s rule (BA): Given the probabilities for the strategies of the opponent. Hurwicz’s rule (HU): An optimal strategy is based on the best and the worst results (Hurwicz. Z. i j j (3) The value λ = 1 gives the most pessimistic solution (Wald’s rule).. 1951).306 E. n. Turskis Wald’s rule (WA): This method is used to search for the best of the worst solutions (Wald. A disadvantage of the method is in the presence of nonoptimal strategies affecting the solution. . 1945). r (2) where r = 1.

001 · 10i . the obtained values are approximately equivalent to the values of linear normalization. The assumption is made that all criteria are maximized. we can make the following conclusion: the application of a new logarithmic normalization method yields the values of the normalized matrix elements approaching the average values obtained in other normalization methods. (7) x4i = 1000 + i. Based on the comparison of normalization results. The alternatives of the initial data for normalization are designed according to various distribution laws: x1i = 0 + i. A comparison of the test results is given in Table 2. . Table 2 Comparison of the test results of matrix logarithmic normalization . x5i = 10 x3i = 100 + i. (i−1) The values of the initial data are changed depending on particular intervals.A New Logarithmic Normalization Method in Games Theory 307 3. where i = 1. In some cases. Normalization Test in the Case of Various Data Intervals In order to test the described new normalization method. x2i = 10 + i. we will consider the normalization of test matrices. x6i = 0. 10.

The evaluation of the alternatives was based on the following four attributes: 1) costs x1 . Turskis Table 2 (continued) Comparison of the test results of logarithmic normalization presented in the matrix 4.308 E. The criteria weights were determined by the method of pairwise . The fourth alternative deals with painting of the balconies with a long-lasting paint..K. 2) labour input x2 . a suburb of Vilnius. 1994). 3).0 × 0. a survey of ﬁve technological alternatives was conducted. Z. The ﬁrst alternative deals with developing a relief surface of concrete by assembling the structures on formwork when concrete is placed in metal forms.5 m made of gypsum concrete slabs. Zavadskas. 4) the criteria 3 and 4 (durability x3 and aesthetics x4 ) were evaluated in points by 28 experts. The second alternative deals with the decoration of facades by open work structural elements of 1. To select a rational alternative of external wall ﬁnishing in castin-place buildings. thousand $. A Case Study of Alternatives’ Evaluation Using Various Solution Methods and Normalization Techniques To illustrate the application of the described methods. The initial data for evaluation of the alternatives are given in Table 3. served as an example of analysis. The ﬁfth alternative deals with painting of the whole facade with a long-lasting paint. A13-storey block-of ﬂats in Justiniskes. the problem of the selection of a rational option of the external ﬁnishing for cast-in-place buildings will be considered (Zavadskas et al. The third alternative presents facing of ceramic tiles. man-days.

non-linear and new logarithmic methods of initial decision-making matrix normalization were used. A similar set v4 v5 v3 v2 v1 is obtained when the criteria weights are not taken into account in the process of alternative assessment. Hodges-Lehmann-rule. linear.10. The use of logarithmic normalization improves the quality of decision matrix normalization in solving economic and organizational problems.42. the “ﬁfth” alternative is better than the “second” one and the “second” alternative is better than the “ﬁrst” one). such as Wald’s rule. also Laplace’s rule. Savage criterion.12.A New Logarithmic Normalization Method in Games Theory Table 3 The initial data for assessing project alternatives and results of decision matrix normalization 309 comparison based on the estimates of 28 experts. Hurwicz’s rule. the “third” alternative is better than the “ﬁfth” one. 0. the vector. When the criteria weights are taken into account. The obtained weight vector of attributes was q = (0. were also applied. A number of different problem solution methods. Tables 4 and 5 provide the solution results and graphical representation of their comparative analysis. In the present investigation. 0. 0. Bayes’s rule.36). . the priority order of the alternatives is presented as v4 v3 v5 v2 v1 (implying that the “fourth” alternative is better than the “third” one.

Turskis Table 4 Ranking of the alternatives Table 5 Ranks of the project alternatives.310 E.K. Solutions according to Wald. Savage. Laplace and Hurwicz (RF = 0. Bayes and Hodges-Lehmann (λ = 0.5) Ranks of alternatives NM WA VE LW NL LN 4 3 2 1 5 4 3 2 1 5 1 4 2=3=5 4 3 2 5 1 q not included SA 4 2 1 3 5 4 3 2 1=5 1 4 2=3=5 4 5 3 2 1 q included LA 4 1 2 3 5 4 3 2 1 5 4 5 1 3 2 4 5 3 2 1 HU 4 2 3 1 5 4 3 2 1=5 1 3=4 5 2 4 5 3 2 1 BA 4 5 3 2 1 5 4 2 1 3 4 5 3 1 2 4 5 3 2 1 HL 4 5 3 2 1 5 4 2 1 3 4 5 3 1 2 4 5 3 2 1 . Z.5). Zavadskas.

when the weights of the criteria are included in calculation. Arrow.g.K.1. E. Bayes and minimax solutions of sequential decision problems. Modelling renewal of construction objects applying c e methods of the game theory. For example. Brauers. Normalisation in multiobjective optimization: a general overview. 21–37.. The logarithmic normalization method used in solving the problems segregates normalized values more effectively than other methods. Zavadskas (2006). Control and Cybernetics. Translated by L. The European Union in a transition economy. various methods and logarithmic normalization used in solving the problem determine the alternative 4 as the most effective. Logarithmic normalization of a decision-making matrix yields more stable results in solving multi-criteria decision problems. Brauers. 461–474. D. Some modules of the program LEVI 3. 263–268. 5.K. 445–460. (2007b). Comentarii Academiae Scientarium Inperialis Petropolitanae. These problems can be solved by applying the program LEVI 3. the most effective option according to the logarithmic normalization (e. W. M. Econometrica. Conclusions It is hardly possible to evaluate the effect of various methods of normalization of a decision-making matrix on the numerical results obtained. J. 5.K. Transformations in Business and Economics. If weights of the criteria are not included in the evaluation process. Econometrica. and E. 6(2). Specimen theoriae novas de mesure sortis. Logarithmic normalization of a decision-making matrix has been used for the ﬁrst time. (1738). Antucheviciene (2007). W. Bernoulli. 8(5–6). A comparison of results obtained by different solution methods is needed because it is not always possible to apply the games theory equilibrium to economics. Zavadskas and J. Brauers. 35(2).. 17. Technological and Economic Development of Economy. Turskis.A New Logarithmic Normalization Method in Games Theory 311 The results of problem solution are more stable. Brauers. 12(4).. E.J. Sommer (1954). Girshick (1949). The MOORA method and its application to privatization in a transition economy. 22.K. W. 1750–192. while other normalization methods describe other alternatives as optimal. linear normalization) is the alternative 4. while other normalization methods give different results. 443–468. 23–36. the most effective fourth alternative was implemented.K.. Ginevicius. Blackwell and M.1 can be used for creating decision-making systems. W. Exposition of a new theory on the measurement of risk. References Antucheviˇ ien˙ . 213–243. What is meant by normalization in decision making? International Journal of Management and Decision Making. Zavadskas (2006). K. .. 8(5–6). engineering and management. Z.A. According to the results obtained in the analysis.R. International Journal of Management and Decision Making.K. D. The calculations show that logarithmic normalization may be used in the cases when the values of the criteria differ considerably.K. (2007a).

8(5–6). and V. N. Zavadskas (1983).K. (1975). Berlin. F. A. F. 370. Messing. Zavadskas (1997). 107–120. 46. Santamarina (2001). International Journal of Management and Decision Making. 185–207.. Determining integrated Criteria signiﬁcance of attributes. C. pp. Vaigauskas and E. Savage. Turskis Cloquell. Dissertation A an der Wirtschaftswissenschaftlichen Fakultät der Humboldt–Universität. International Journal of Strategic Property Management. Turskis (2002). Noorul Haq.0 – multiple criteria evaluation program under uncertainty.L. Variantenvergleich durch Matrixspiele. E. Krutinis and M. F. D. Annals of Mathematics Studies. Vilnius (in Lithuanian). Peldschus. V. and E. 276–281.L. 9(2). A. and E. Journal of Civil Engineering and Management (Statyba). (1986). Some problems of evaluating multicriteria decision methods. and K. 10(3).. (2003). Zur Berücksichtigung mehrer Zielfunktionen bei der Optimierung von Produktionsplanen. H. Sensibilitätsuntersuchungen zu Methoden der merhkriteriellen Entscheidungen. 519–526. Neumann von. Peldschus. and G. e Technological and Economic Development of Economy. Migilinskas. 321–326. Technologische Entscheidungen bei der Berücksichtigung mehrerer Ziehle. Ustinovichius. (2001). St. (1951).L. Zavadskas and J.. Zur Anwendung der Theorie der Spiele für Aufgaben der Bautechnologie. 184–201. 263–286. Peldschus. St. F. Analysis and Synthesis of Parameters under Information Deﬁciency.. Zavadskas. Technological and Economic Development of Economy. Hurwicz. Bauplanung–Bautechnik. Viteikien˙ (2007). E. 73–79 (in Lithuanian). 13(1). Journal of Civil Engineering and Management. (1996). M. 45–52.. The use of previous experience in reaching statistical decision. In MCDA 54th Meeting in Durbuy. Peldschus. Statistics. D.P. 8(1). T. Morgenstern (1943). 55–57.J. International Journal of Management and Decision Making. 8(5–6). 7(4). Körth. Zavadskas. L. Jüttler. 8(5–6). A new procedure for the numerical values normalization in multicriteria decision techniques. Normalisation in the selection of construction alternatives. R. Zavadskas.K.. Technological and Economic Development of Economy. Hodges. Yoon (1981). Mathematik und Wirtschaft. 623–639. Untersuchungen zur Fragen der Operations aforschung und ihrer Anwendungsmöglichkeiten auf ökonomische Problemstellungen unter besonderer Berücksichtigung der Spieltheorie. 7(4). Migilinskas. (2007). Kalibatas. Multi-objective evaluation of microclimate in dwelling. Ustinovichius (2007). (1966). L. Šatkauskas (2007). cowles commission paper. . 8(4). LEVI 3.K. D.. Zagorskas (2006). and L. (1951). Technologie.. International Journal of Management and Decision Making. Peldschus. E. Banaitis and G. 8(5–6). J. 24–31.K. Technische Hochschule Leipzig. B. Turskis. 1–10. 527–539. (1969). Diss. Stopp.312 E. 6. The effectiveness of assessment in multiple criteria decisions. Universidad Politécnica de Valencia.. 601–622. Deﬁning the utility and market value of real estate: a multiple criteria approach. 11(2). Petersburg University Press. H. The theory of statistical decision. F. International Journal of Management and Decision Making. B. L. Valencia. Journal of the American Statistical Association. The model of determining the sensitivity of elements of multiple criteria evaluation methods. Princeton University Press. Z. Peldschus. F. Lehmann (1952). E. Kannan (2007). (2001). L. Wissenschaftliche Zeitschrift der Hochschule für Bauwesen Leipzig. 2.. Hovanov. 396–407. Diss. F.K. Multiple Attribute Decision Making – Methods and Applications. Journal of Civil Engineering and Management (Statyba). 3–12. The inﬂuence of normalization methods selection in construction including game theory adaptation... 117. D. and C. Petersburg. Hwang. Optimality criteria for decision making under ignorance.. Ginevicius. 23. Dejus.K. Sustainable city compactness evaluation of the basis of GIS and Bayes rule. New York. and O. A Hybrid normalised multi criteria decision making for the vendor selection in a supply chain model. Podvezko (2007). SpringerVerlag. International Journal of Strategic Property Management. A. Theory of Games and Economic Behavior. (2002). Technika. Ustinovichius and Z. Z. Kaklauskas. Matrix Games in Building Technology and Management. 173–175.A. J. 37(4).

International Journal of Management and Decision Making. Turskis (2006). Zavadskas. Zavadskas. Wald. Evaluation of ranking accuracy in multi-criteria decisions.R. Ustinovichius (2003). International Journal of Strategic Property Management. 601–617. 25–43. Peldschus..0 – Multiple criteria evaluation program for construction solutions. Multi-attribute assessment of road design solutions by using the COPRAS method.. M. 2(4).. Kaunas city case study.K.K. Informatica. Dejus and M. Z. Vaidogas E. T. Nijkamp (1977). Turskis. Ustinovichius. and A. Vilnius (in Lithuanian). A.. E. Technika. 8(3). Statistical decisions functions which minimise the maximum risk. 13(2). Assessment of investment proﬁtability in construction from the technological perspective. A. Van Delft.K. Leiden. Development of software for multiple criteria evaluation. E. and E.K. A. (1976). L. International Journal of Management and Decision Making. 8(5–6).. Zavadskas. L. Mehrzielselection für Entscheidungen im Bauwesen. subjective and integrated weights of attributes. Zavadskas. Moscow Civil Engineering Institute.A New Logarithmic Normalization Method in Games Theory 313 Ustinovichius. Dissertation of Dr. F. 540–554.K. and E. Determination of efﬁciency of investments in construction. 184– 191.K. Peldschus and A. Zavadskas. Zavadskas. Viteikiene (2007b). Kaklauskas.K. 149–155.K. Kaklauskas (1994). Sensitivity analysis of a simple additive weight method. Zavadskas. E. Game Theory in Building Technology and Management. F. Turskis (2007). E. (1945). Fraunhofer IRB Verlag. (2004).K.K. Control and Cybernetics. (2006).. 36(1). Ustinovichius and Z. Vaidogas E. Zakarevicius and J. Informatica. D. Zavadskas. 497– 518. (2007). J. 259–272.. International Journal of Management and Decision Making. 8(5–6). and Z. Viteikiene. E.K.. E.K. International Journal of Management and Decision Making. Sustainable residential areas evaluation. Journal of Civil Engineering and Management. Multi-Criteria Analysis and Regional Decision-Making. Peldschus and L. 12(2). 46. 347–352. Reliability measures in multicriteria decision making as applied to engineering projects. Peldschus and Z. P. Podvezko (2007). M. Sc.. 475–496. F. 17(4). Vilnius (in Lithuanian).. Antucheviciene (2006). 265–280.K. Kaklauskas (2007). Zavadskas (2007). 8(1). Zagorskas. E. Weimar. L. Methods of determining objective. Moscow (in Russian). F. L. Ustinovichius. F. LEVI 3. Ustinovichius. E. Technika. A. M. Zavadskas (2007). E. Nijhoft. . Multiple criteria evaluation of technological decisions of construction. Z. Beitrag zur Optimierung der räumlichen Struktur eines Gebäudes. Turskis (2007a). Multi-attribute model for estimation of retail centres inﬂuence on the city structure. Zavadskas. Viteikiene.. Technological and Economic Development of Economy. Multiple Criteria Evaluation of Projects in Construction. Zavadskas and V.K. Zavadskas and Z. Technika. Weitendorf. 8(5–6). Hochschule für Architektur und Bauwesen. L. 12(4).. Evaluating the sustainability of Vilnius city residential areas. Nl. 195–203. and E. The Baltic Journal of Road and Bridge Engineering. 555–574. Annals of Mathematics. Dissertation A.. 251–268.. E. Technological and Economic Development of Economy. E. (1987). Turskis. Zavadskas (2004). Ustinovichius. Application of a quantitative multiple criteria decision making (MCDM-1) approach to the analysis of investment in construction. 14(2). Messing (2002). Journal of Civil Engineering and Management. Peldschus and D. Turskis (2004). 152–160. Introducing reliability measures into multi-criteria decision-making. Vilnius. L..R. 8(5–6).

Z. Normalizuojant šiuos kriterijus siekiama ju suvienodinimo. German and Russian. Vilnius. He is a member of the Lithuanian and several foreign Academies of Sciences. . He is a member of international organisations and has been a member of steering and programme committees at many international conferences.K. Naujai sukurtoje u programoje LEVI 3. Zavadskas is a member of editorial boards of several research journals. His research interests include building technology and management. Zenonas TURSKIS Daugiakriterinis sprendimu pri˙ mimas yra naudojamas daugelyje visuomen˙ s veiklos sriˇ iu. naudojamas lošimu teorijoje Edmundas Kazimieras ZAVADSKAS. Turskis (PhD) is a senior research fellow of Construction Technology and Management Laboratory of Vilnius Gediminas Technical University.1 yra naudojami tokie normalizavimo metodai: vektorinis. Z. English. Research interests are: building technology and management. E. lošimu teorijoje. jis gali b¯ ti labai svarbus kai kuriais atvejais. naudojamo sprendimu pri˙ mimo matricos e normalizavimui. He is the author of 20 scientiﬁc papers. Šiame tyrime e pagrindinis d˙ mesys yra skiriamas logaritminio metodo. nelinijinis ir naujasis logaritminis. He has a PhD in building structures (1973) and Dr Sc (1987) in building technology and management. He is doctore honoris causa at Poznan. decision-making theory. Naujasis logaritminis normalizavimo metodas. automation in design and decision support systems.K. Turskis E. Logaritminis normalizavimas niekada nebuvo panaudotas. Nors normalizavimas ne visada reikalinu gas. Kiev. Lithuania. kurie turi skirtingus matavimo vienetus.314 E. Zavadskas. decision-making theory. aprašymui. Lithuania. SaintPetersburg. pavyzdžiui.K. of Construction Technology and Management at Vilnius Gediminas Technical University. He is author and co-author of more than 300 papers and a number of monographs in Lithuanian. computeraided automation in design and expert systems. Zavadskas is a principal vice-rector of Vilnius Gediminas Technical University and a head of the Dept. kad b¯ tu galima palyginti kriteriju reikšmes. e e c Alternatyva daugiakriteriniame vertinime dažniausiai yra apibr˙ žiama kiekybiniais ir kokybiniais e kriterijais. linijinis.

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue listening from where you left off, or restart the preview.

scribd