You are on page 1of 11

Organizacija, Volume 38 Scientific Papers Number 9, November 2005

An Alternative Criteria Research
Methodology for Selecting a New Product
Milan Nikoli}1, Zvonko Sajfert1 and Branka Nikoli}2

1
University of Novi Sad, Mihajlo Pupin Technical Faculty, 23000 Zrenjanin, Djure Djakovi}a bb, Serbia and Montenegro,
bmnikolic@ptt.yu; z.sajfert@eunet.yu
2
Sonja Marinkovi} Primary School, 23000 Zrenjanin, Tepli~ka 7, Serbia and Montenegro, bmnikolic@ptt.yu

Metodologija za analizo kriterijev alternativ ob izbiri novega proizvoda

Prispevek obravnava metodologijo uporabljeno pri raziskavi kompleksnih problemov pri poslovnem odlo~anju: dolo~itev rela-
tivne pomembnosti kriterijev ob izboru novega proizvoda. Metodologija podpira generiranje predlogov za definicijo relativne
pomembnosti kriterijev in podkriterijev za izbiro novega proizvoda glede na trenutno situacijo v podjetju (ki izbira nov proizvod)
in stanje. Z uporabo predlagane metodologije lahko dolo~imo vpliv relativne pomembnosti kriterijev za izbor novega proizvo-
da na stopnjo uspešnosti podjetja. Zaradi omenjenega ima celotna procedura dinami~en zna~aj in je lahko uporabljena v
razli~nih situacijah ter ob razli~nih ~asih opazovanja. Oblikovani predlogi predstavljajo vhodne podatke ter podporo slede~emu
ve~kriterijskemu rangiranju alternativ za nove proizvode.
Metodologija je potrjena v praksi vendar pri~ujo~i prispevek ne podaja podrobnosti v zvezi s tem, saj je cilj prispevka pred-
stavitev raziskovalne procedure na podro~ju dolo~itve kriterijev pri izboru novega proizvoda kakor tudi opredelitev pomemb-
nosti le-te. Poleg tega je uporabljena procedura pomembna zaradi svoje univerzalnosti saj jo lahko uporabimo pri opredelitvi
kriterijev v drugih odlo~itvenih situacijah z ali brez dodatnih sprememb.

Klju~ne besede: metodologija, raziskave, kriteriji za izbiro novega proizvoda

1 Introduction However, it is very difficult, almost impossible, to ma-
ke a universal model (a model which could be suitable for
One of the most significant problems in business decision- a large number of companies) for defining the relative
making is the selection of a new product, which usually importance of criteria when selecting a new product. In
has to be performed on the basis of a larger number of cri- the research carried out in (Nikoli}, 2004), the initial
teria. That is why various multi-criteria analysis methods hypothesis was that every company is a multi-dimension-
are frequently applied (AHP, PROMETHEE, and oth- al system, with its specificities and characteristics which
ers). Selecting a new product is not only important for the have direct influence on assigning certain levels of impor-
survival and future of every company, but it also presents tance to individual criteria. As a consequence of this,
an extremely complex task because of the range of this there is the question of variability of the importance of
problem area and a large number of influential values individual criteria that depend on numerous characteris-
which are of a prevailingly stochastic character. Com- tic features of the company, and above all, the company’s
plexity and subjectivity are particularly prominent in the degree of success. Furthermore, if one takes into account
procedure of assigning relative values to criteria when the fact that a company is a dynamic system in which
selecting a new product. there are constant numerous changes, as well as changes
Existing research of the criteria for selecting a new in its setting, then there is the question of variability of
product (indices of the future success of a product) main- importance of some criteria in time, for one and the same
ly deal with singling out several such indices and deter- company.
mining the intensity of their influence. Then potentially Furthermore, it must not be overlooked that the
new products are compared with these indices, and the existing research, for example (Cooper & Kleinschmidt,
product with the best assessments is adopted. This is the 1987; Henard & Szymanski, 2001; Montoya-Weiss & Ca-
basic idea in much important research which has been lantone, 1987), mainly refers to countries with developed
carried out in this way, for example (Cooper & Klein- economies. Hence, it follows that the results of this resear-
schmidt, 1987; Henard & Szymanski, 2001; Montoya- ch especially favour the indices related to the originality
Weiss & Calantone, 1987; Parry & Song, 1994). and attractiveness of a product, user’s needs and the like,

543

step eight. account the above-mentioned difficulties. For this reason. Montoya-Weiss & Ca. Step Three: Polling of experts in the field of product a greater influence. 2001. These weights are used in step eight when deter- parameters (within each criterion) does not have to be the mining the similarity value of the companies which same and it is usually not the same: some parameters have belong to the same classes. the selected criteria and sub-criteria uct. above all financial. are largely ignored.. It should be pointed out that the parameter numbers which describe particular criteria (represented by ‘n’) do not have to be in any way equal for all the criteria. but to research the product are identified . 1994). 2. 2. step. As a conse. foresees defining the sub-criteria for each criterion (one 1987.. taking into given point of time. the importance of these rately. . is that the methodolo- gy proposed here highlights the importance of defining Step Two: Selection of parameters which influence precisely at the start the criteria for selecting a new prod. ting. Step One: Selection of criteria and sub-criteria for 2 Alternative research methodology of selecting a new product criteria for selecting a new product In this step. So.. and that they are fixed. p – the number of the observed criterion Xij – j-th parameter which describes the i-th criterion. neously influence a larger number of the selected criteria. These criteria are researched in the further The main difference between this methodology and some procedure. a special research methodology company could define the importance (relative weights) has been developed to research the relative importance of of the criteria according to its own characteristics at a the criteria for selecting a new product. dure. Table 1: Mutual dependence of the criteria for selecting a new product and the parameters that describe these criteria Labels in Table 1 have the following meanings: Ci – criterion for selecting a new product. 544 . The number of these parameters depends on the criterion itself. that the research can be repeated and realized for differ- quence of this. The proposed research methodology of the criteria ogy is essentially based on expert methods and fuzzy sets for selecting a new product foresees an eight-step proce- theory. which will also be discussed in a further presen- same importance for all companies and in all conditions. mine the recommendations on the basis of which every Because of all this.These parameters are in reality the importance of predetermined criteria depending on the characteristics of companies influencing the importance characteristics of a company and its setting. the criteria for selecting a new product are defined .. as well as when applying the results of the nificant problem is that these criteria do not have the research. 3. This methodol. of a particular criterion. Therefore. Certain parameters can simulta- tributed to such formulation is the concept that the crite. . the stress is laid on establishing the dependence of Table 1 was formed with the aim of illustrating mutu- importance of criteria for selecting a new product on the al dependence between the criteria for selecting a new company’s degree of success and the conditions in its set. November 2005 whereas indices related to the very capacities of the com.Organizacija. where: i = 1. ria are generally the same and that they can be defined on The parameters have the greatest importance in the last the grounds of existing knowledge. as when applying the models. (Cooper & Kleinschmidt. where j = 1. Detailed descriptions of each step follow. for example. The ultimate objective is to deter- capacities of both the company and the market. and some less. the objective of the In step two the parameters which have influence on research is not to determine which criteria are important the selected criteria and sub-criteria for selecting a new and to what level they are important. can also work without any sub-criteria). similar research in the countries in transi.. n – the number of the observed parameter. Parry & Song. tation. Volume 38 Scientific Papers Number 9. Henard & Szymanski. ent groups of criteria when it is necessary.. lantone. However. What characterizes the proposed methodology is pany. What has con. it is development (Survey 1) necessary to determine relative weights of the selected Each criterion is described with a certain number of parameters within the framework of each criterion sepa- influential parameters. or in the case of tion must take into account more seriously the financial a different approach. product and the parameters which describe these criteria. and that a more sig. 3. 1987. The methodology that is presented here also existing research.

through the function y criteria and sub-criteria. Ranking of companies into classes is carried out at ter for the i-th criterion (item 2. ny describing the selected criteria. economists. tial. In That is why stress is laid on the subjective opinion of an the process. Research entrusted to experts from scientific institutions and to must also include the plan of the area in which the experts from the companies who deal with the problems research would be carried out.). A set is and sub-criteria with one single assessment in the formed from these fuzzy assessments.Organizacija. and especially product devel. technologists and si- influence on more criteria. 2004). applied in (Nikoli}. technichi. processing of the data provided in the parameters according to the current state of the com- previous step is carried out. identified parameters (characteristics) of the compa- ans. Those polled will be asked to = x / 10.). These assess- characteristics of polling in this case are: ments are translated into the interval [0. For the scope of getting reli- different criteria. and are by vocation economists. and also when applying the research results rately. In order to focus on the problem. Number of Managers. able and relevant data. The Polled. eter with an assessment in the intervals between 0 and 10. 2000. ments (given by the experts) are translated into fuzzy 2. 1] by means of 1. Questions which provide the data on all the selected assessments.e. In a further text. quantita- This is the step in which polling of company managers tively assess the importance of the criteria with one in certain companies is carried out (Survey 2). In this way relative weights of Step Six: Ranking of companies into classes according parameters within each criterion are obtained separately. All the parameter assess. one of the fol- 545 . parameters in the assessment range between 0 and ed parameters for each criterion 10. Volume 38 Scientific Papers Number 9. It is also very important that performed with this set by determining the average value the subjects assess the criteria and sub-criteria of the set. mechanical engineers. November 2005 Relative weights of the selected parameters which proposed methodology was carried out for the com- influence the criteria and the sub-criteria for selecting a panies in the food processing industry. etc . It is assumed that they are uni- a new product. and processing is range between 0 and 10.200 managers. and the company is in the back- like. versity-educated. its development on the importance of the observed criterion for selecting strategy. and the like. according to the size of the relative importance of the observed criterion in relation to other criteria. relative weights of the parameters for that criterion. Step Five: Polling company managers (Survey 2) The managers. Company Type. The polled must be at high-level leading This assessment should represent the quantitatively positions in companies so that they can have insight expressed strength of influence of a particular parameter in the company’s business results. where: x ∈ [0. 1997. 2. ground serving as the framework in which an individual Podinovski. in addition to The necessary opinions are collected through the survey being the part of economy with the greatest poten- (Survey 1). setting. 1. i. where x ∈ [0. individual (manager). The final fuzzy assessment of the parameters within The primary objective of this survey is to establish the one criterion represent the input data for determining the mutual dependence between parameters and criteria.. 10]. Research Domain in Geographical Terms. Those polled will be required to quantitatively assess the required Step Four: Determining relative weights of the select. 2004). a research which applied the assessment (FA(Ci)). The obtained average values represent the final according to the current state in that company and its fuzzy assessments of all the parameters for each criterion. questions in Survey 2 will be called ‘managers’. which is also presented in (Nikoli} & Sajfert 2004). eter can have a different influence on the importance of 3. tance for a manager is carried out according to the fuzzy ched. such parameters ought to be milar. Noghin. In (Nikoli}. the same param. it is foreseen that it is neces- The research related to determining the relative sary to poll N = 150 . it is the function y = x / 10. They are applied in the level of a single criterion. economy of Serbia and Montenegro. It is very important that those polled assess the In this step. The polled are asked to evaluate each param. 2002) and others. The basic assessment ranging from 0 to 10 (step five). Determining the criterion’s relative impor- form of the type of company which will be resear. to the relative importance of the criteria These are values wij – relative weight of the j-th parame. In view of the fact that some parameters can have managers. 1]. It is essential that the ranking be carried out (item 3.1. Those polled should be organization engineers. 10] (columns FA(Ci) necessary to select the subject of the research in the in Table 2). for example (Leskinen. Questions that can provide the data about all the managers. Fuzzy assessments of each quantitatively assess the importance of the criteria parameter for a particular criterion are observed. of company development. or the interval [0. mechanical engineers. those polled who answer assessed more than once. according to their opinion. An original approach is forms his/her opinion. The research requires asking two groups of questions: opment. In order to do this. one of numerous procedures can be applied. pany. weights of the parameters for each criterion should be 4. for each criterion sepa- step eight. In other words. The rationale new product are determined on the basis of the opinion is that food industry is the most vital part of the given by the experts in the field of product development.

the next conclusion is that the companies in one rion separately.. as a result of which the relative Then.. The tive weights of the criteria. which is also pre. . Therefore. and not according to the fuzzy companies are thus divided into three classes according to assessments of the criteria because the managers can use each criterion separately: different parts of the 0–10 scale. This The classification is carried out according to the rela- procedure is carried out for each criterion separately. It would require consid. the observed criterion has an average importance in rela- The first 50 companies make up the first class. since there would be 50 to 60 compa. if 150 managers were sur- to other criteria.. 1]. 2000. N. the observed criterion has a great importance in Noghin.. as a result of which the relative impor. 1997. This is repeated for average fuzzy assessment to the observed criterion in every polled manager separately (columns wCi in Table 2). to have the category “average” and several qualitative values on both sides of the average. 2. 2004). 2. the second 50 companies the second. agers. The total number of companies is divided by the desired Class III – Companies whose managers assigned a number of classes and the number of companies in one low fuzzy assessment to the observed criterion in relation class is obtained. tance in relation to the other ones). approach is applied in (Nikoli}. starting from the highest to the lowest. 3. class must have significant mutual similarities. for drawing certain conclusions. then the number of companies in each class is 50. Seven classes would Step Seven: Statistical processing of the data within be far too many. where k = 1. All the assessments obtained by Survey 2 are translated nies within one class.Organizacija. and the third 50 companies tion to the other ones). classes. it is possible to 3. For groups It is proposed here that the number of classes be S = of similar companies within one criterion.. . FA(Ci) – fuzzy assessments of the i-th criterion for each manager from k = 1. make up the third class for the observed criterion.. on separately. 2. For example. the number of classes S = 3 is company managers (Survey 2) is carried out in this step. more appropriate. agers into classes is shown in Table 2. the procedure of ranking the man- tance of the observed criterion is great (for these man. Class II – Companies whose managers assigned an sented in (Nikoli} & Sajfert 2004)). Podinovski. companies (managers’ opinions) are ranked importance of the observed criterion is average (for these according to the size of the relative importance of the managers. so that the main dilemma was whether each class for each criterion separately the number of classes ought to be three or five. N – ordinal number of the observed manager.. 3. and not only its initial fuzzy assessment.. what is impor- Class I – Companies whose managers assigned a high tant is the position of the observed criterion in relation to fuzzy assessment to the observed criterion in relation to other criteria. that is. Determining the criteria fuzzy assessments 546 . N. Processing is carried out within each class for each criteri- erably more managers to form five classes. and we want to rank them into three tance of the observed criterion is low (for these managers. values wCi are ranked according to the size for each crite. Statistical processing of the data provided by polling Considering the fact that the planned number of polled managers was N = 150–200 . The columns with Table 2: Ranking of managers into classes for all the criteria Labels in Table 2 have the following meaning: Mk – the k-th polled manager. 2002) and others (original relation to the other ones). In a general case. which represent a sufficient number into fuzzy assessments. November 2005 lowing procedures can be applied (Leskinen. the other criteria. the interval [0. Data processing is carried out in two seg- Since managers form their opinions on the basis of ments: the characteristics and potentials of their respective com. the observed criterion has an average impor- observed criterion.. . Volume 38 Scientific Papers Number 9. panies. 1. relation to other criteria. When determining the number of classes it was certain carry out objective statistical processing of the data that the number of classes ought to be an odd one in order obtained. as a result of which the relative impor- veyed in companies. 3. wCi – the relative weight of the Ci criterion for each manager from k = 1.

av. as age relative weight.. 2002) and Table 3: Determining relative weights of the Ci criterion sub-criteria for the s-th class Labels in Table 3 have the following meaning: Msik – the k-th polled manager from the s-th class for the i-th criterion. wci.cl) is found carried out for every average of every class. and in the assessments represent the fuzzy assessments of each class third the lowest. step are translated into fuzzy assessments. It is obvious that in the first essary to carry out fuzzification of 36 values. and it indicates the criterion’s maximum aver- number of classes (it is proposed that S = 3 classes). November 2005 Values from one wci column (Table 2) are ranked in relation to which fuzzification is carried out is labelled according to their size and are divided into the desired as wCi. After that.. then these assessments actually represent a fuzzy set application of the calculated recommendations in a con. H = N / S – the number of managers in one class. For example: . 1997.av. Podinovski.av. others. 1] interval. one of the procedures can be applied.. All the parameter assessments obtained in the fifth which is also presented in (Nikoli} & Sajfert 2004). 2000.cl). In Table 3. If the assessments obtained for the parameters tive for each class which affect a single criterion in one company are obser- Step eight is necessary because of the possibility of ved. then it is nec- class within each criterion. with the function y = x / 10. for each class of companies with- in each criterion separately. for all the criteria average relative with dividing into classes (step six). where k = 1. 2. 10]. in other words.Organizacija.. 3. into the [0. do that. and within each criteri- and the average of respective values. Step Eight: Determining of the company-representa. that is. is found for each on the companies are divided into 3 classes.. 2004).max. To On the graph abscissa on Figure 1 the average crite. Noghin. influence on a certain criterion and its sub-criteria for that 547 . of the model (described in item 3). The assessments given by the managers from the s-th class to the sub-crite- ria of the i-th criterion (columns Msik in Table 3) are ente- 0 wCi. Then follows the fuzzification of these of a single criterion. The sub-criteria relative weights are determined for each cri- terion separately. The sub-criteria relative weights criteria for each class are calculated on the basis of fuzzy assessments of these sub-criteria for each class of that criterion separately. . each criterion separately Managers follow their opinion when they quanti- FA tatively evaluate the importance of the sub-criteria with one of the assessments from the 0–10 range (step five). wci. 10]. where: x ∈ [0. The fuzzification is carried out by applying the 2.cl red first.av. Volume 38 Scientific Papers Number 9. These fuzzy class there will always be the highest average. 2. an example of the s-th class of the Ci criterion is given. The ranking is carried out for each criterion separately. the average value of the fuzzy assess- 0 ment for each sub-criterion (the column with averages in Figure 1: Fuzzification of average relative weights of the Table 3) is determined. Fuzzification is average of values. 1] interval by means of the function y = x / 10. Determining the sub-criteria relative weights for function shown on Figure 1. where: x ∈ [0.. For each class the weights in all the classes (values wCi. An original approach is applied in (Nikoli}. where: r = 1. q – the number of the observed i-th criterion’s sub-criterion. terion. of characteristics of the company observed which have crete case. H – the number of the polled company from the s-th class for the i-th criterion. 1 These assessments are translated into the [0. in that class (values wCi. and they are used in the application average relative weights for every class of every single cri. if there are 12 criteria in a model. . SCir – the r-th sub-criterion of the i-th criterion.max wCi. for exam- ria relative weights for each class are plotted . The value ple: (Leskinen.av.

Volume 38 Scientific Papers Number 9. 2. 1] interval.. where N is the the number of analyzed compa. of that class aim at. tives can occur. a The essence of this step is to determine for each class decrease in the representativeness of these representa- those companies which are ‘most similar’ to the compa. The class represen- fuzzy sets. greater stability of expressed by the manager which was formed on the basis the results is expected than in the case of each class hav- of the situation in his/her company. according to the i- th criterion. ∑ m kji j=1 j≠ k msrk – average similarity measure of the k-th company in relation to the other companies from the s-th class. It is adopted that for each class there Note: What is meant as ‘company’ here is the opinion are three representatives. where k = 1. . Observed at the level nies of its class (within a given criterion). H – the number of the company sur- veyed from the s-th class for the i-th criterion. there are N nies are representatives of their class. increasing the number of the class representatives. Cksi – the kth surveyed company from the s-th class for the i-th criterion. equality mkji = mjki still applies. 3. The similarity measure takes its value from the [0.. .. November 2005 company (Cksi columns on Table 4). tatives have parameter values which all other companies nies. µksji – degree of membership (fuzzy assessment) of the k-th company from the s-th class.Organizacija. These compa- of one single criterion for all the companies. On the other hand. according to the i-th criterion. 2.. H – the number of the company surveyed from the s-th class for the i-th criterion. In this way. which is more convenient ing one representative. for the j-th parameter. where j = 1. .... Table 5: Determining the similarity measure of each fuzzy set with each other. 2. In the process. by further than the term manager for this presentation. H . between the companies belonging to one class Labels in Table 5 have the following meaning: Cksi – the k-th polled company from the s-th class for the i-th criterion. n – the number of the observed parameter.. It is calculated in the following way: H (1) ∑ m kji j=1 j≠ k m srk = H −1 548 .sum total of the measures of similarity of the k-th company in relation to the other companies from the s-th class. mkji – the similarity measure of the k-th company to the j-th company. That is why the term company is used in further text. where k = 1. Table 4: Tabular presentation of fuzzy sets for the companies of the s-th class according to the i-th criterion Labels in Table 4 have the following meaning: Xji – the parameter which describes the i-th criterion..

that is. which is man. The measure. there is H of fuzzy sets. A table showing elements contained in th company according to the i-th criterion. Table 6: Presentation of the final research results in a general case (for the Ci criterion) 549 . Triantaphyllou. where H – the number ber of parameters which describe the i-th criterion. The final results of the research can be presented in table Xjai – fuzzy assessment of the j-th parameter for the a. useful. 1− ni where: 2. and the where: same is then repeated for all other criteria. the ref. 1− ifest from the last column in Table 4. 2000) were particularly the basis of Survey 1 (step three). and the results are entered in Table 5. and with it the mabi = ∑ X jai − X jbi (2) j=1 unreal increase in the degree of the similarity measure. 1998. a fuzzy measure was because in this case there is no unreal increase in the simi- defined which rests on Hamming’s distance: larity measure. Wij values must be used. For each class of companies th company according to the i-th criterion.2 Presentation of the final research results mabi – the similarity measure of the a-th company to the b-th company according to the i-th criterion. Table 6 is formed is created for each criterion. The class representatives are three companies which n have the highest average similarity measures in relation mabi = ∑ X jai − X jbi ⋅ Wij (3) j=1 to the other companies from the same class. on Verdegay. of companies (managers) in one class. form (Table 6). 2. individual parameters within the same criterion. This is shown in The subtrahend in expression (2) represents. specificities of the given research. which takes into account the relative weights of (3) is used to determine the degree of similarity. the average difference (distance) between the a-th and After forming Table 4.Organizacija. where: When defining the corresponding fuzzy measure. the num- important). Table 4. If the wij values were used. The procedure is ni repeated for all the classes within one criterion. fuzzy set to each other within the companies of one class In order to determine more objectively the similarity are determined. Xjbi – fuzzy assessment of the j-th parameter for the b- sented with Tables 4 and 5. (the companies in which the observed criterion is equally ni – the parameter number of pairs. 2000. there would be n unreal decrease in the average difference. On the basis of this. wij – the relative weight of the j-th parameter for the erences which treat the area of multi-criteria decision. The expression duced. the similarity measures of each the b-th company according to the i-th criterion. in fact. and based on the needs and In the expression (3). expansion of the previous expression is intro- companies are compared with themselves. November 2005 The method of processing in step eight can be repre. Royo & within all the criteria are obtained in the step four. These weights for all the parameters making and fuzzy sets (Pedrycz & Gomide.1 Fuzzy measure selection and defining Wij = wij ⋅ ni (4) The right selection of the fuzzy measure that will be applied in the research represents a special problem. i-th criterion. Volume 38 Scientific Papers Number 9. Wij – proportional (expanded) relative weight of the j- th parameter for the i-th criterion.

2.. the fuzzy set for one single criterion is pany (the company which is selecting a new product) in compared with the fuzzy sets of representatives of all the current conditions. which is also pre- data processing in the Seventh step) for the class of sented in (Nikoli} & Sajfert 2004). Fuzzy sets are formed from the given assessments for results’ of the criteria and sub-criteria is the procedure of each criterion separately (elements of these fuzzy determining the fuzzy assessments of all the criteria and sets are parameter values for a corresponding criteri- relative weights of their sub-criteria for a concrete com.. . In this way. 2004). 550 . Volume 38 Scientific Papers Number 9. 1997. original assessment of the observed criterion and relative weights of their sub-criteria (which is obtained by approach is applied in (Nikoli}. we ilar to according to the observed criterion. where: : j = 1. n – the number of the observed parameter. 3.. The procedure is the following: the classes within that criterion through the selected 1. The class within this criterion with which the con- current state in the company and its setting. criteria. The similarity meas- own company according to all the required xij param. Then. With the data on the companies it is most similar to. relative weights of all the criteria and sub-criteria. measure is finally adopted. the struc. Wij – Extended relative weight of the j-th parameter for the i-th criterion. To do this. This procedure is shown of all the criteria and the relative weights of their sub- in Table 7.. The assessments are awarded for the 7).. 2. determining the criteria relative weights. To do this.Organizacija. mksi – the similarity measure of a concrete company to the k-th representative of the s-th Ci criterion class. These are crete company has the highest collective similarity Table 7: Determining the highest collective similarity measure of a concrete company (CC) with the representatives of all the Ci criterion classes The labels in Table 7 have the following meaning: Xij – the j-th parameter which describes the i-th criterion. ␯ji –fuzzy assessment of the k-th representative in the s-th Ci criterion class for the j-th parameter. which are entered in the last row of Table 7. In the last column in Table 7 is required maximum The only thing left is to determine the relative value Sµsi – collective similarity measure of the i-th weights of the basic criteria. 2002) and others. A concrete can use some of the procedures such as: (Leskinen. ures are added up for every three representatives of eters (like the managers in the first part of Survey 2 . November 2005 3 Application of th research results the nji values. 2000. Podinovski. 3. n – the number of the observed parameter. The managers from a concrete company assess their fuzzy measure. it is determined which ments of all the criteria serve as the starting point for class of companies the concrete company is most sim. What is understood under ‘application of the research 2. The values of fuzzy assess- criterion s-th class. In this way. expression (3).. where: j = 1. on). . ␮ksji – fuzzy assessment of the k-th representative in the s-th Ci criterion class for the j-th parameter.The same procedure procedure of multi-criteria selection of a new product is is repeated for all the criteria. company adopts the recommendations on fuzzy Noghin. the tured data from Table 6 are used. 3. each class (the Smsi values in the last column in Table step five did). a concrete entered by applying one of the multi-criteria analysis company determines the values of fuzzy assessments methods.

Market portance of the criteria for selecting a new product and Needs (Nikoli}. The example the territory of Serbia and Montenegro.Strength of Competition. parameters one can determine which class is a CC most similar to according to the observed criterion Ci.Continuity in information inflow from the mar. importance of criteria and sub-criteria for selecting a new product. and 4 Examples I parameter values for a CC (Concrete Company) (to be assigned by the decision-maker in the company 4. That is the reason why an original methodology of on the local market. I relative weights of the Ci criterion sub-criteria for each class (determined in Step 7. The criteria relative weights are obtained on the ery. 2004): the company’s degree of success and the situation in its SC91 . terion. recom- Example of criteria (Nikoli}. The company which is selecting a new prod- X94 . process.Personnel Potentials. C7 .Evaluation of the demand for the new product developed.Company Business Policy. dure has a dynamic character and the possibility of appli- ket. value for the user and the like). In the territory of Vojvodina and in the Belgrade metropoli- Table 8 are shown: tan area. and that within each each Ci criterion class (determined in Step 8). I values of extended relative weights of all the criterion 2004). but depending on the current characteristics of a SC93 . the whole proce- X92 . tools. attractiveness. (determined in Step 7.Degree of acquaintance and awareness of con. criterion there are 3 classes of companies.Time period of return of invested funds. and the relative weights of the Ci criterion sub- C2 . more precisely on refers to a single criterion (criterion Ci is viewed here).Management of production capacities (machin. Volume 38 Scientific Papers Number 9.1 Example of criteria. settings.Consumers' satisfaction with company's existing obtained in the form of fuzzy assessments of the impor- products. There is a mutual dependence between the im- Example of sub-criteria of criterion C9 . 2004): mendations are obtained on fuzzy assessments of the Ci C1 .Supply of Raw Materials. for one of 312 = 531441 possibilities of defining the relative I Ci criterion fuzzy assessments (FA) for all classes weights of the basic criteria. The recommendations are X95 .Product Potentials (originality. then a concrete company can get recommendations Ci parameters (expression 4).Market Growth and Prospects. In this way.Market Size. importance (relative weights) for all companies and all C12 . The most significant characteristic and advantage of Example of some parameters of criterion C9 . criterion. and which is selecting a new product). Criteria for selecting a new product do not have equal C11 .Evaluation of the demand for the new product company and its setting. November 2005 If it is assumed that there are 12 criteria for selecting I parameter values for all the three representatives of a new product which are researched. The research was carried out on obtained by the research is shown in Table 8. The how). By applying the procedure from item 3 of the paper. sub-criteria. model. state and situation in its setting. results The proposed methodology also has its practical value.Management of spatial capacities.2 An example of application of the research alternatives for new products. etc. Some of the quoted procedures can be used in the C6 . that the CC is most similar to the criterion Ci I class.Evaluation of the demand for the new product setting. criteria.Degree of change in consumers' needs.Market the proposed methodology is that with a change of certain Needs (Nikoli}. basis of the recommended fuzzy assessments for each cri- C5 . segment 2). segment 1). observation.Market Needs. 2004) research of the criteria in food A shortened example of application of the results industry was carried out.Organizacija.Possession of technological knowledge (know. C8 . C10 . 2004): conditions inside the company and its setting. In the example given in Table 8. equipment). It enabled obtaining recommendations on the on the national market.Timeliness and reliability of information from corresponding changes of exiting the model (recommen- the market. cation in different situations and at different moments of X93 . uct obtains the recommendations according to its current sumers' needs. it can be seen C3 . same procedure is repeated for all the criteria in the C4 . (like in Nikoli}. 5 Conclusion C9 . tance of the criteria and their sub-criteria relative weights. The obtained results have significance primarily 551 . on the international market. The obtained recommendations represent important input data and support for multi-criteria ranking of the 4. research of the criteria for selecting a new product was SC92 . there are X91 . dations for each criterion) For all this. In (Nikoli}.

J.184. An Introduction to Fuzzy Sets - but the stress is laid on the validity of the procedure itself. although their application in small Research. or completely over. innovated existing products. Measurement. (1997). F. Analysis and Design. come if one uses those criteria which have been identified 11(1).174. With this respect. management. sub-criteria ing a new product proposed here could be accepted as a and parameter defining stage. products that are new for the company obser. 15-30. companies should not be ruled out. Journal of Marketing and large companies. 163 . November 2005 Table 8: An example of application of the research results of criterion Ci for the food industry in Serbia and Montenegro. The results obtained by applying the proposed 16(3). 362 . Nikoli}. above all at the criteria. Identifying New Product criteria and sub-criteria for selecting a new product. (2004). Product Innovation. Industrial Marketing Management. Success Factors in work. ved. Podinovski. 4(3). In this way. (2001). selected in this way achieve very good results on the mar. PhD thesis. In also be significant for other countries in transition. The products that have been Analysis. Turkey. &Calantone. (1994). Leskinen.H. for MCDA. (2002).V. V. as Successes in China. can be used for researching other problems Sakarya. M. The observed draw. 38(3). tive approach. Journal of Multi-Criteria Decision Analysis. & Sajfert. and phenomena in industrial engineering. and above all because of the desire to point to the very procedure of Cooper.M. &Gomide. 215 . ket and positively influence the company’s business. W. well as the influential parameters. Volume 38 Scientific Papers Number 9. The results of the Montoya-Weiss. and its universality. the research method of the criteria for select- certain alterations. 169 . can be applied to research the criteria for selecting a new product in other economies and economic conditions as well.D. Z. with adequate posium On Intelligent Manufacturing Systems IMS’2004. (1998). E. Why some new prod- methodology are primarily significant for medium-sized ucts are more successful than others. up to now. Cambridge and which can be repeated and realized for different sets of Massachusetts Institute of Technology. R. tions analyzed. The 4th International Sym- in this paper is in that its basic principles. The quantitative importance of criteria back can be significantly reduced. Henard. D. D. These claims have been confirmed in 12. (1987). M. Criteria Decision Analysis. Faculty of Mechanical Engineering. The concrete results of the research are not present. Journal of Multi-Criteria Decision Analysis. the 6(6). 9(4). scales and scale independ- gories of new products can be ranked: completely new ence in the analytic hierarchy process. most prominent is the fact that it is essential to define the Parry.363. P. (1987). M. situa- University of Belgrade. Pedrycz. R. Nikoli}.375. Regarding drawbacks of the given methodology. 6-8 September 2004. but could as the key ones in one of the known research projects. Journal of Multi- products.M. Journal of Product Innovation Management. (2000). all cate. This subjectivity is real. &Song. A Bradford Book. criteria and parameters if necessary.Organizacija. M. its significance. 6 References ed here because of the size of the paper. (2004). Noghin.223.G. V. 1-15. Relative importance of criteria: a quantita- but also in other areas. Quantitative model for selecting a new prod- uct with research into relevant criteria. modification. Journal of Product Innovation Management. &Kleinschmidt.E. 552 . etc. X. Widening of Saati’s scale for Additional importance of the methodology presented comparison of criteria in pairs. Determinants of research are already applied in the concrete situations of New Products Performannce: A Review and Meta- selecting a new product. 355 . 11(1). the proposed methodology continuation and addition to similar research.M. &Szymanski.

His his master's degree at the same institution in 2001. Department became an assistant professor at the same faculty. management”. E. International Journal of Uncertanly. November 2005 Royo. nizational Sciences. In 2002 she obtained her Master's degree from the ware tools in management with a particular stress on the University of Novi Sad. He has Mihajlo Pupin Technical Faculty in Zrenjanin. She has published approximately 15 papers in the field of fuzzy sets. He has published more than 30 papers in these Master’s thesis was “Comparison of norms in fuzzy reason- fields. Boston. J. of Mathematics.Based Systems. the same faculty. He is the author of eleven university level text- Milan Nikoli} graduated from the University of Novi Sad. (2000). Branka Nikoli} graduated in 1998 from the University of neering. Department of Industrial Engineering. ics. Multi-Criteria Decision Making and his doctorate in 1994 with the thesis “Exploring the in- Methods: A Comparative Study. Faculty of Science and Mathematics. 553 . Coherence Measures on Zvonko Sajfert graduated from the Faculty of Orga- Finite Fuzzy sets. A. she works as a mathematics teacher at the Sonja Marinkovi} Primary School in Zrenjanin. He is currently associate professor at Uni- versity of Novi Sad. Mihajlo Pupin Technical Faculty in processes of winning new products and business decision Zrenjanin. His ba. In 2004 specific field of interest is human resource management. Triantaphyllou. From Fuzziness and Knowledge .Organizacija. 8(6). University of Belgrade in 1989. & Verdegay.L. Faculty of Mechanical Engi. Department of Computer Science. He received published approximately 60 scientific and expert works. In 2005 he Novi Sad. ing”. Mihajlo Pupin Technical Faculty in Zrenjanin. he defended his PhD thesis entitled: “Quantitative model for selecting a new product with research into relevant criteria” at the University of Belgrade. Kluwer Academic Pub. The title of her making. Currently. 641-663. Volume 38 Scientific Papers Number 9. books and several secondary school text-books. he received his master's degree in 1992. (2000). with a specialization in teaching mathemat- sic fields of interest are using quantitative methods and soft.S. fluence of the structure of the managers' life aims on the lishers.