SELECTION OF A DISTRIBUTION CHANNEL USING THE INTEGRATED FUCOM – MARCOS MODEL

The management of manufacturing companies faces a number of decisions, and one of the most important is the selection of distribution channels. A large number of these companies do not sell their products directly to end consumers. For this reason, there are marketing intermediaries between manufacturers and end consumers whose primary function is to connect manufacturers and consumers. Their task is to provide the goods from manufacturers to consumers with the satisfaction of logistics characteristics: at the right time, at the right place and in a form that is convenient to use, and certainly with minimal costs. Distribution is one of four marketing mix instruments without which the optimal combination of the instruments would not be obtained. Thus, the decision on selecting distribution channels is as important as the decisions regarding products, prices and promotion. Based on the set criteria and the evaluation of certain distribution channels by the criteria, the management of the company will be able to make the best decision. The evaluation of distribution channels based on the set criteria was performed by marketing experts and experts in certain markets using an integrated multi-criteria model. The FUCOM method was applied to determine the significance of the criteria, and then the distribution channels were evaluated by applying the new MARCOS method. Thereafter, a sensitivity analysis was performed using other MCDM methods to verify the results previously obtained.


Introduction
All four instruments have to be considered when deciding on the optimal combination of marketing instruments, i.e., the marketing mix. Therefore, decisions about products, prices, channels. Marketing instruments are interconnected and one without the others would not represent the optimal combination of the marketing mix, therefore the decision on one significantly influences the decision on the other instruments. The business success and competitive advantage of a company significantly depends on decisions about how to combine marketing instruments in the best possible way [5].
No matter what business the company is engaged in, it has to decide on how to make the products available to consumers. It is also confirmed by scientific papers referring to distribution channels in various fields of production (McCabe et al., [6]; Schegg et al., [7]; Camilleri, [8]; Vasiliauskas et al., [9]; Atănăsoaie, [10]; Thakran & Verma, [11].
The factors that influence the selection of distribution channels are most commonly product characteristics, consumer habits, financial situation of the company, price, geographic concentration and width of product assortment. All of these factors determine to some extent which distribution channel a particular business will use for the placement of their products to consumers. Accordingly, certain authors also addressed the factors that influence the selection of distribution channels. Saremi & Zadeh [12] analyzed decision-making regarding distribution on the basis of certain criteria and its impact on the overall marketing system of the company. When selecting a distribution channel, the management of the company have to consider the goals of the company as a whole since the decision on selecting distribution channels has a significant impact on achieving business results as it is directly related to the placement of products and their sallling in the market [13]. With the development of science and technology, there have been changes in all spheres of business, as well as in distribution channels. In their research, Watson et al., [14] observe a period of development of distribution channels from 1980 to 2014, where significant changes are noticed in the functioning of distribution channels. Galkin [15] analyzes the influence of factors on the selection of distribution channels by specific regions, and he has proved that all factors that may influence the selection of distribution channels should be considered for each region individually since their influence varies from region to region. Sabiote et al., [16] state that consumer behavior, consumer habits and culture can significantly influence the selection of distribution channels.
Liu & Cui [17] analyzed the impact of product line length on distribution channels. Stoddard et al., [18] determine a link between consumer habits, distribution channels, and products. The way in which a particular product or service can influence the selection of distribution channels is explained in the research by Kim et al., [19].
In order for a company to be always prepared for a competitive response, it has to consider any changes in the factors affecting distribution channels and, accordingly, change the decision on the selection of distribution channels [20].
The FUCOM method is widely used as proved by numerous scientific papers where the method has been used. The FUCOM method assists managers in prioritizing criteria using simple algorithms, as well as in assessing phenomena according to current requirements of decision-makers [38]. Sofuoğlu [22] notes that the FUCOM method in combination with other methods provided successful results when making decisions in a manufacturing company engaged in the manufacture and processing of newly developed high strength parts.
Durmić [23] uses the FUCOM method to evaluate the criteria when selecting suppliers. Prentkovskis et al., [24] use the FUCOM method to determine the weight coefficients of quality dimensions when measuring service quality. The FUCOM method is used in various decision-making fields and when evaluating certain alternatives Nunić, [25]

Methods
The methodology for deciding on the selection of a distribution channel used in this paper consists of three phases ( Figure 1). In the first phase, data from a particular company are collected. In this phase, the current state of distribution in the company has been defined.
Based on the current state of distribution and the intention of the company to improve the distribution of its products, the criteria for selecting distribution channels as well as potential alternatives are determined. Subsequently, the second phase is approached, where the FUCOM method is applied. By the application of the FUCOM method, the weight coefficients of the criteria are defined. The third phase consists of the application of the MARCOS method. By applying the steps of this method, the ideal and anti-ideal solution and the values of utility functions of alternatives are obtained. Then the alternatives are ranked from the best to the worst. The best-ranked alternative will be the one with the value of utility function closest to the value of the ideal solution, and the worst will be the alternative with the value of utility function closest to the value of the anti-ideal solution. Thus, the choice of the company should be the alternative that is ranked best. The FUCOM method is based on the principles of pairwise comparison and validation of results through deviation from maximum consistency [21]. Benefits that are determinative for the application of FUCOM are a small number of pairwise comparisons of criteria (only n-1 comparison), the ability to validate the results by defining the deviation from maximum consistency (DMC) of comparison and appreciating transitivity in pairwise comparisons of criteria. The FUCOM model also has a subjective influence of a decision-maker on the final values of the weights of criteria. This particularly refers to the first and second steps of FUCOM in which decision-makers rank the criteria according to their personal preferences and perform pairwise comparisons of ranked criteria.
However, unlike other subjective models, FUCOM has shown minor deviations in the obtained values of the weights of criteria from optimal values [21], [28], [23], [32], [26]. Additionally, the methodological procedure of FUCOM eliminates the problem of redundancy of pairwise comparisons of criteria, which exists in some subjective models for determining the weights of criteria. Figure 2 presents the FUCOM algorithm [33].
Input: Expert pairwise comparison of criteria Step 1: Expert ranking of criteria/sub-criteria Step 2: Determining the vectors of the comparative significance of evaluation criteria

Restriction 1:
The ratio of the weight coefficients of criteria is equal to the comparative significance among the observed criteria

Restriction 2:
The values of weight coefficients should satisfy the condition of mathematical transitivity Step 3: Defining the restrictions of a non-linear optimization model Step 4: Defining a model for determining the final values of the weight coefficients of evaluation criteria Step 5: Calculating the final values of evaluation criteria/sub-criteria Output: Optimal values of the weight coefficients of criteria/subcriteria

Fig. 2. Steps of FUCOM method
In Figure 2, we can see that the algorithm consists of five steps that are clearly explained.
Here we assume that there are n evaluation criteria which are denoted as wj, j = 1,2, ..., n and that their weight coefficients should be determined. The FUCOM method requires decision-makers to determine the impact of criterion i on criterion j.
The MARCOS method is based on defining the relationship between alternatives and reference values (ideal and anti-ideal alternatives). On the basis of the defined relationships, the utility functions of alternatives are determined and compromise ranking is made in relation to ideal and anti-ideal solutions. Decision preferences are defined on the basis of utility functions. Utility functions represent the position of an alternative with regard to an ideal and anti-ideal solution. The best alternative is the one that is closest to the ideal and at the same time furthest from the anti-ideal reference point. The MARCOS method is performed through the following steps [34]: Step 1: Formation of an initial decision-making matrix. Multi-criteria models include the definition of a set of n criteria and m alternatives. In the case of group decision-making, a set of r experts should be formed to evaluate alternatives according to the criteria. In the case of group decision-making, expert evaluation matrices are aggregated into an initial group decision-making matrix.
Step 2: Formation of an extended initial matrix. In this step, the extension of the initial matrix is performed by defining the ideal (AI) and anti-ideal (AAI) solution.
The anti-ideal solution (AAI) is the worst alternative, while the ideal solution (AI) is an alternative with the best characteristic. Depending on the nature of the criteria, AAI and AI are defined by applying Equations (2) and (3): where B represents a benefit group of criteria, while C represents a group of cost criteria.
Step 3: Normalization of the extended initial matrix (X). The elements of the normalized matrix ij mn Nn   =  are obtained by applying Equations (4) and (5): where elements ij x and ai x represent the elements of the matrix X.
Step 4: Determination of the weighted matrix ij mn Vv   =  . The weighted matrix V is obtained by multiplying the normalized matrix N with the weight coefficients of the criterion j w , Equation (6).
Step 5: Calculation of the utility degree of alternatives Ki. By applying Equations (7) and (8), the utility degrees of an alternative in relation to the anti-ideal and ideal solution are calculated.
where i S (i=1, 2, ..., m) represents the sum of the elements of the weighted matrix V, Equation Step 6: Determination of the utility function of alternatives f(Ki). The utility function is the compromise of the observed alternative in relation to the ideal and anti-ideal solution. The utility function of alternatives is defined by Equation (10).  (11) and (12).
Step 7: Ranking the alternatives. Ranking the alternatives is based on the final values of utility functions. It is desirable that an alternative has the highest possible value of the utility function.

Case study
This paper considers making the best decision on selecting a distribution channel for a particular company. The research for the purpose of this paper has been carried out in a company classified as a small business since it employs 26 workers. The company was founded in 1956 and it has not changed the main activities of its business since then. The company carries out production, classified as mixed agricultural production, where the products are divided into three groups. The first group consists of four products in the field of crop production (wheat, corn, soybeans and rapeseed), the second group also consists of four products in the field of seed production (maize seed, wheat seed, barley seed and triticale seed) and the third group consists of one product in the field of animal husbandry (beef cattle). Crop production is carried out on approximately 500 ha, which represents a large land surface and a large annual crop yield. At any moment, the company has about 400 head of beef cattle, which is also a large amount of production. The company also processes 1000 tons of small grain seeds over one production period, which covers a period of one year.
Production is carried out on the territory of the municipality of Derventa, while products are marketed both domestically and abroad. Therefore, the company represents one of the major exporters of these agricultural products. So far, the company has been selling its products exclusively directly to consumers. This type of selling causes distribution costs reaching their maximum, i.e., the company has no additional funds to invest in distribution.
For this reason, the company intends to find a new distribution method to reduce the costs and invest that money in improving the production.

Forming a Multi-Criteria Model
In order for the company to make the best decision on selecting a distribution channel, it was formed a team of experts that consisted of three executives from a company for distribution of agricultural products, all three from BiH, and three foreign professors in the field of marketing specifically engaged in distribution as a narrow field of their interest. They evaluated alternatives by all criteria based on the following scale of values shown in Table 1. Poor -P 3 Medium -M 4 Good -G 5 Very Good -VG All members of the expert team are familiar with the current situation in the company, as well as its needs and requirements, and based on that, they evaluated the criteria and alternatives presented in Tables 2 and 3. Table 2 defines the criteria by which the experts evaluated the alternatives. Characteristics and description of the criteria C1 Product characteristics The basic features and characteristics of the products; durable productslonger distribution channels, less durable productsshorter distribution channels. C2 Financial situation of the company The number of financial resources that the company is ready to allocate to the distribution channels and the funds determine the channels; more moneyshorter distribution channels, less moneylonger distribution channels.

C3 Consumer habits
The practice of purchasing certain products; consumer goodslonger distribution channels; special productsshorter distribution channels.

C4 Costs
The price per unit of distributed product; high priceshorter distribution channels, low pricelonger distribution channels.

C5 Geographic concentration
The geographic locations where the company sells its products; higher concentrationshorter distribution channels, lower concentrationlonger distribution channels. C6 Width of the product assortment The number of various product lines offered by the manufacturer for sale to consumers; wide assortmentshorter distribution channels, narrow assortmentlonger distribution channels. Table 3 provides a description of the alternatives that were evaluated by the experts according to the criteria set out in Table 2. Each alternative is defined individually in the table for the purpose of their easier review. The manufacturer sells the product directly to the consumer.

A2 Manufacturer-Retailer-Consumer
The manufacturer sells the goods to a retailer and the retailer sells the goods to end consumers. A3 Manufacturer-Wholesaler-Retailer-Consumer The manufacturer sells the goods to a wholesaler, the wholesaler to a retailer and the retailer eventually sells the goods to end consumers.

A4 Manufacturer-Agent-Consumer
Agents do not take ownership of the goods but sell the goods to the buyers on behalf of the manufacturer, so they work in their own name and on behalf of others. A5 Manufacturer-Broker-Consumer Brokers work for a fixed period of time until an agreement is reached between the manufacturer and the consumer, they do not take any risks, work on the commission from the sales price. They work in the name and on behalf of others. A6 Manufacturer-Auction House-Consumer Auction houses have the task of bringing together the manufacturer and the buyer in one place and arranging public bidding. They work on the commission from the sales price.

A7 Manufacturer-Commissioner-Consumer
The commissioners take the goods from manufacturers and sell the goods at the best price to consumers; they work on commission and settle their costs. The evaluation of the alternatives by the criteria was performed based on the scale previously presented in Table 1. Ratings of the alternatives are shown in Tables 4 and 5. Table 4 shows the ratings of the alternatives by the criteria using the scale from 1 to 5 from Table 1.  Table 5 shows the ratings of the alternatives according to the criteria set by the decisionmakers using the linguistic scale from Table 1. Table 5, therefore, represents the ratings from Table 4, but in this case, the ratings are presented descriptively.

Evaluation of the Criteria by the Steps of the Fucom Method
The members of the expert team compared the criteria by significance. Then, they rated how much the significance of the best criterion was greater compared to the others by applying a scale of 1-9. The ratings of the criteria by the expert team members are shown in Table 6.  Table 6 shows that the first decision-maker identified Criterion 2, i.e., the financial situation of the company, as the most significant criterion, and C5, i.e., geographic concentration, as the least significant. Further, the table shows that the second decision-maker also marked the financial situation of the company as the most significant criterion, and the geographic concentration as the least significant, as well as the first decision-maker. The third decision-maker indicated that product characteristics as the criterion for selecting a distribution channel are the most significant criterion, and that consumer habits are the least significant criterion. The fourth and fifth decision-makers, as well as the first and second decision-makers, considered that the most significant criterion for selecting a distribution channel was the financial situation of the company, but the fifth decision-maker also indicated that the costs were as significant as the financial situation of the company. According to them, the least significant criterion was geographic concentration. The sixth decision-maker identified costs as the most significant criterion and geographic concentration as the least significant criterion for selecting a distribution channel.
After all the steps of the FUCOM method, the results of the evaluation of the criteria are obtained, as shown in Table 7. The table is accompanied by a graphic representation of the results where it is easier to see which criterion is the most significant and the least significant for the decision-makers. © Filodiritto Editore  Table 7, as well as Figure 3, shows that the first decision-maker estimates that the financial situation of the company is the most important in selecting a distribution channel (0.358). The first decision-maker considers geographic concentration to be the least significant criterion when selecting a distribution channel (0.83). The second decision-maker also rates the financial situation of the company as the most significant criterion when selecting a distribution channel (0.295). The second decision-maker rates consumer habits and geographic concentration equally, and considers the two criteria to be the least significant when selecting a distribution channel (0.98). The third decision-maker considers product characteristics to be the most significant criterion when selecting a distribution channel (0.338). The third decision-maker considers consumer habits to be the least significant criterion (0.050). The fourth decision-maker rates that the financial situation of the company is the most significant criterion when selecting a distribution channel (0.337). The least significant criterion for the fourth decision-maker when selecting a distribution channel is geographic concentration (0.073). The fifth decision-maker considers the financial situation of the company and the costs to be the most significant criteria compared to other criteria (0.322), while the width of product assortment is the least significant criterion when selecting a distribution channel (0.080). According to the sixth decision-maker, the most significant criterion when selecting a distribution channel is the costs (0.428). The least significant criterion when selecting a distribution channel according to the sixth decision-maker is geographic concentration (0.071). The table and the graph show that, according to the decision-makers, the financial situation of the company is the most significant criterion when selecting a distribution channel; four decision makers (DM1, DM2, DM4 and DM5) rated this criterion as the most important (0.280). The costs as a criterion are in the second position with a slightly lower score (0.257). The least significant criterion, when considering the overall Faculty of Business Economics and Entrepreneurship International Review (2020 No. 3-4) © Filodiritto Editore score, is geographic concentration, and even five members of the expert team (DM1, DM2, DM4, DM5 and DM6) rated it as the least significant (0.079). Applying Equation (4), the normalized values of the cost criterion are obtained, for example:

Fig. 3. A graphic representation of the evaluation results of the criteria by the FUCOM method
The normalized values of the benefit criterion are obtained by applying Equation (5): and a complete normalized matrix is shown in Table 9. The next step is to extend the normalized matrix using Equation (6) by multiplying all the values of the normalized matrix by the values of the criteria. The extended normalized matrix is shown in Table 10. Using Equations (7), (12), the final results in Table are obtained by the MARCOS method. Finally, the utility function of alternative A1 is obtained by applying Equation (11): ( 1 ) = The final results obtained by the MARCOS method are shown in Table 11. In Table 11, the alternatives are ranked using all seven steps of the MARCOS method. Ideal and anti-ideal solutions are obtained, the values of 1.000 and 0.619, respectively. The alternative with the utility function value closest to the ideal solution value is the best alternative and is ranked first. In this research, it is alternative A3, i.e., Manufacturer-Wholesaler-Retailer-Consumer, with the utility function value of 0.751. The worst-ranked alternative is the alternative with the utility function value closest to the value of the anti-ideal solution, and here it is alternative A1, i.e., Manufacturer-Consumer, with the utility function value of 0.561.

Sensitivity analysis
In order to verify the results, we compared the results obtained by the MARCOS method with the results obtained by other multi-criteria decision-making methods. Therefore, in this section of the paper, a sensitivity analysis of the obtained results is performed by the MARCOS method. The results of the analysis are shown in Table 12 and Figure 4.  [37]. The first column of Table 12 lists the ranking results of the alternatives obtained by the MARCOS method. The second column contains the ranking results obtained by the SAW method. The third column contains the results obtained by the ARAS method. The fourth column contains the results obtained by the WASPAS method.
Comparing the results obtained by the last three methods it can be noticed that the results are generally the same, which means that we can with high probability claim that the results obtained are accurate. The table and figure show that there is a slight deviation in ranking by © Filodiritto Editore the ARAS method compared to the other three methods. By this method, alternatives A2 and A7 has changed their positions, i.e., alternative A2 is ranked fourth by this method and alternative A7 as fifth. It is the only difference we can see using all four methods. All other methods show identical ranking results. Thus, by applying all four methods, A1 alternative, i.e., direct sales to consumers, is ranked as the worst alternative, i.e., it is in the seventh place.
Alternative A3, i.e., sales via a wholesaler and then retailer, is the best alternative according to the results obtained by all four methods. That is, alternative A3 is ranked first according to all the methods. Therefore, the results obtained by the MARCOS method represent accurate and reliable results. If the company wants to make the best decision about selecting a distribution channel, it can rely on the results obtained by this method. According to the results of this research, the best decision of the company on selecting a distribution channel would be the decision to use alternative A3 as the distribution channel.

Conclusion
The research in this paper was carried out in a company engaged in mixed agricultural production, i.e., crop production, seed production and animal husbandry. The company performs its production at one place, while marketing the products domestically and abroad.
So far, the company has used only one distribution channel and it is direct sales to consumers. However, this method of selling creates huge costs. At present, the company has no additional funds to invest in this distribution channel, and the current budget for this activity has reached their maximum. The company expressed the need and request to conduct the research that would contribute to decide on selecting the best distribution channel. In order to reach the best decision, experts in this field were hired. The expert team consisted of six experts who, on the basis of a questionnaire sent to each of them, evaluated the criteria for the selection of a distribution channel and ranked distribution channels according to each criterion.
Based on the application of the FUCOM method, the criteria were evaluated, and using the MARCOS method, the alternatives, i.e., distribution channels, were ranked according to experts' rates. Applying the method, the worst-ranked alternative was alternative A1, i.e., direct sales to consumers. The method showed that alternative A3, i.e., the sale of products via a wholesaler and a retailer, is the best-ranked method. This channel of distribution would be a much better alternative than the current way of selling products directly to consumers. Therefore, the company should replace direct sales by selling through intermediaries, in this case wholesale and retail trade. In this way, the company would sell its products to end consumers with some financial savings. This method of selling would also provide other benefits that this channel of distribution brings in relation to direct sales.
In regard to this paper, it remains for future researchers to address the savings that the company would make if it decides on selecting the recommended distribution channel. Certainly, future research may also refer to investing the savings in the development of production and further growth of the company.