**2. Materials and methodology**

Besides for making decisions, some scientists [23–25] have used MCDA methods also as assessment tools. Ferrarini et al. [26] used MCA to assess and compare municipal performance in environmental quality and sustainability in the province of Reggio Emilia in Italy. Gómez-Limón et al. [27] also made use of MCA to analyse input usage in agriculture and the way it affects the environment. Hellstrand [28] found MCA useful to survey the sustainability effects of increasing concentrate intensity in Swedish milk production. Solomon and Hughey [29] proposed a MCA decision support tool for international environmental policy issues on the example of emissions control in the international aviation sector. Crete Tsoutsos et al. [30] showed how sustainable energy planning can be done by MCA. The analysis of air pollution [31] and soil pollution [32] in an urban area in Serbia was done by Nikolić et al. using MCA. For improving strategic environmental assessment of water programmes in Brazil, MCA was also taken by Garfì et al. [58]. In Malaysia, Al-Hadu et al. [33] showed how useful MCA is for environmental management. Payraudeau and Gregoire [34] modelled pesticides transfer to

40 Applications and Theory of Analytic Hierarchy Process - Decision Making for Strategic Decisions

One of the most common methods of MCA is analytical hierarchical process (AHP). AHP has found its use in a few branches of the agricultural field for more than two decades, though not extensively. It has since been very attractive and useful for water management engineers. Pillai and Rasu [35] used this method for ranking irrigation management alternatives in an Indian region in order to increase the effectiveness of the irrigation system, which was underutilised. Tiwari et al. [36] used AHP to develop a framework for environmental-economic decision making that includes the environmental and economic sustainability criteria, and local people's preferences in the context of a lowland irrigated agriculture system in Thailand. The method was also relevant for Ni and Li [37], who used it for the assessment of soil erosion in terms of land use structure changes. Tran et al. [38] used AHP to prioritize future renewals of irrigation and drainage assets in the La Khe irrigation scheme in North Vietnam; Srdjevic and Medeiros [39] also demonstrated the use of AHP for the assessment of water management plans. Braunschweig and Becker [40] showed how AHP could be used in international agriculture to choose research priorities. Pažek et al. [18] used AHP for evaluation of business alternatives on organic farms. Liu et al. [5] made an assessment of how sustainable a high-yield agro ecosystem in Huantai County, China, was. In Iran, Rezaei-Moghaddam and Karami [41] used AHP for the evaluation of sustainable agricultural development models. Ziolkowska [42] used AHP in combination with cost-effectiveness analysis for the evaluation of the AEM and analysis of the economic aspects to support the decision-making process of the Polish gov‐ ernment. In the same year, Ziolkowska [43] also combined AHP and linear programming to estimate the importance of AEM with respect to the environmental objectives and calculate an objective orientated budget allocation for AEM. Ziolowska also used the AHP to investigate/ evaluate the importance of AEM from the regional perspective in Poland. Mortazavi et al. [44] showed how AHP can successfully be used for prioritizing agricultural research projects. Vindis et al. [45] also used AHP to perform a further evaluation of energy crops for biogas

The aim of this paper is therefore a further attempt to show how multicriteria decision models can be successfully applied in the assessment of agricultural problems, the main focus being

surface water with MCA.

production.

The assessment in this paper is based on one of the most used multicriteria decision methods (hereafter MCDM), the AHP. As recommended by Saaty [46], Meixner and Haas [47], the paper is organised as follows:

**Figure 1.** Aggregated weights based on group judgements.

Step 1: The main goal, subgoals (attributes), criteria, people involved and/or affected and their objectives and the means of reaching the goal were identified and formulated.

**Figure 2.** Synthesis with respect to goal (group judgements).

Step 2: The models in AHP were built by decomposing the complex main goal into smaller less complex sub-goals, factors that affect the subgoals and ending with the outcome of the strategies [46,48]. This led to a hierarchical structure (**Figure 1**) with criteria, attributes and alternatives [48,49]. The concrete measures to fulfil the defined objectives to finally reach the main goal, 29 of them in this case compiled by the state, are at the bottom of the hierarchy.

Step 3: To determine the interrelations in the hierarchy, pairwise comparisons of the parame‐ ters at each level of the hierarchy were done with respect to the element immediately above them. Through pairwise comparison of the elements at each level of the hierarchy and within the hierarchical levels, weights were determined (**Figure 2**), which help to show the correla‐ tions within the structure [49]. AHP allows comparisons using real measurements (quantita‐ tive judgement) or a scale created by Saaty which expresses the degree of preference, importance or likelihood (qualitative judgement) [18,46].

Step 1: The main goal, subgoals (attributes), criteria, people involved and/or affected and their

Step 2: The models in AHP were built by decomposing the complex main goal into smaller less complex sub-goals, factors that affect the subgoals and ending with the outcome of the strategies [46,48]. This led to a hierarchical structure (**Figure 1**) with criteria, attributes and alternatives [48,49]. The concrete measures to fulfil the defined objectives to finally reach the main goal, 29 of them in this case compiled by the state, are at the bottom of the hierarchy.

Step 3: To determine the interrelations in the hierarchy, pairwise comparisons of the parame‐ ters at each level of the hierarchy were done with respect to the element immediately above them. Through pairwise comparison of the elements at each level of the hierarchy and within the hierarchical levels, weights were determined (**Figure 2**), which help to show the correla‐

objectives and the means of reaching the goal were identified and formulated.

42 Applications and Theory of Analytic Hierarchy Process - Decision Making for Strategic Decisions

**Figure 2.** Synthesis with respect to goal (group judgements).

Step 4: Aggregation of the priorities to have a ranking of the alternatives was carried out. This was done by determining the ratings of the alternatives with respect to each criterion and then adding up these ratings for all criteria.

Step 5: Control of consistency was done by determining the consistency index, CI that is calculated as follows:

$$CI = \frac{\lambda\_{\text{max}} - n}{n - 1}$$

where ƛmax is the eigenvalue of the matrix and *n* the size of the matrix. A consistency index of up to 10% is tolerable. A slight deviation of the consistency index from 10% is not a problem. A large deviation means that the judgements are not optimal and have to be improved.

The attributes/criteria and alternatives included in the hierarchy (**Figure 1**) were extracted from the agri-environmental measures of the RDP of the Republic of Slovenia. Each EU member state has its own RDP especially compiled to suit its circumstances and special conditions.

The hierarchical structure consists of four levels with the main goal, "Assessment of agrienvironmental measures" as the first level. The second level has three subgoals (•) and the third level a different number of subgoals (-) for each subgoal, as shown in **Figure 1**:

	- **–** Soil quality and fertility will be improved by reducing soil erosion, loss of humus and loss of nutrients through leaching [50].
	- **–** Agri-environmental measures aim at reducing the contamination of groundwater and drinking water sources through chemicals discharged into the environment during agricultural production.
	- **–** According to Latacz-Lohmann and Hodge [51], there has been an exaggerated and uncontrolled use of chemical fertilizers and pesticides in the past. As a result, considerable decline of biodiversity was observed among other negative consequences. The agrienvironmental measures aim at reducing this destructive practice.
	- **–** Conservation of agricultural land implies minimal soil disturbance, permanent soil cover and crop rotation.
	- **–** Costs of measures play a major role as to whether they are successfully implemented until the end of the given period.
	- **–** Successful implementation of the agri-environmental measures also depends on how complex they are for the farmers. Too complex measures will be wrongly put into practice, which leads to the wrong outcome as the intended.
	- **–** To be able to get the produce from the rural areas on the markets, reliable channels for marketing have to be created.
	- **–** If there is economic profitability for the farmers through implementation of the measures, the farmers might not give up farming. They might also not leave rural areas and migrate to urban areas [55].
	- **–** The farmers will probably encounter yield reduction if they changed the method of production to suit the demand for more biologically produced foodstuff. Their products are healthier and of a higher quality. They produce less but will be able to sell their products at higher prices.

A Case Study on the Application of the Analytic Hierarchy Process (AHP) to Assess Agri-Environmental Measures of the Rural Development Programme (RDP 2007–2013) in Slovenia http://dx.doi.org/10.5772/63924 45

The fourth and last level at the bottom of the hierarchy has 29 measures that are very specific for Slovenia.

After determining the hierarchy, questionnaires were sent to five experts involved in the preparation and evaluation of agriculture environmental subsidies payment system, on who made pairwise comparisons at each level of the hierarchy. These pairwise comparisons were used to assess the agri-environmental measures. For assessment, the individual judgements obtained from questionnaires were turned into comparison matrices for each expert. Since this decision procedure is considered as a group decision, the decision values from the individual experts have to be aggregated to one matrix. This can be done by aggregating the values of the individual pairwise comparisons at each level of the hierarchy or by first calculating the priority weights for each individual expert at each level of the hierarchy and then aggregating these priority weights. In both cases, the aggregation is done by building the geometric mean as recommended by Saaty and Vargas [49] and Meixner and Haas [47]. The geometric mean (*G*) is obtained by calculating the *n-th* root of the product of the individual expert values:

$$G = \sqrt[n]{x\_1 x\_2 \dots x\_n}$$

where *x* represents each expert and *n* is the total number of experts. The results in this paper were obtained by aggregating the values of the individual pairwise comparisons at each level of the hierarchy. The values of the aggregated matrices were fed into the software programme Expert Choice™, which was programmed to implement AHP. Expert Choice™ calculated the criteria, subcriteria and alternative weights with respect to importance and the consistency index.

### **3. Results and discussion**

**–** Unique traditional and indigenous domestic animal breeds are mostly well known for their toughness and resistance against aggressive animal diseases. So the main aim here is to retain this valuable genetic material. Genetic diversity will help reduce loss in times

44 Applications and Theory of Analytic Hierarchy Process - Decision Making for Strategic Decisions

**–** Climatic change has evoked unreliable weather conditions. The growing seasons are threatened by these unpredictable weather conditions. Traditional and indigenous plant varieties contribute to a greater diversity of crop plants that can be utilised for agricultural food production. They are a valuable genetic source towards food security since many can grow under harsh conditions. Their constituents are usually highly nutritious or medically effective. Preservation of a high agro-biodiversity is one of the important goals

**–** Less favoured areas already have the problem that agricultural land is limited and the conditions for agricultural production are not favourable. The little space that is available has to be used gently/carefully to avoid deterioration. Traditional and indigenous domestic animal breeds and plant varieties could play a role to make these areas usable

**–** The landscape has to preferably be kept in its natural state so that many animals, big and small, have their ideal habitat. This means for example that grasslands have to be

**–** Job creation is vital to make the rural areas an attractive place to live. This might help to

**•** Production and economic consequences: With farmers investing in the rural areas, new jobs will be created. If the rural areas are made attractive enough with the appropriate infra‐

**–** Costs of measures play a major role as to whether they are successfully implemented until

**–** Successful implementation of the agri-environmental measures also depends on how complex they are for the farmers. Too complex measures will be wrongly put into practice,

**–** To be able to get the produce from the rural areas on the markets, reliable channels for

**–** If there is economic profitability for the farmers through implementation of the measures, the farmers might not give up farming. They might also not leave rural areas and migrate

**–** The farmers will probably encounter yield reduction if they changed the method of production to suit the demand for more biologically produced foodstuff. Their products are healthier and of a higher quality. They produce less but will be able to sell their

attract many people out of the industrial/urban areas back to the rural areas.

structure, even young farmers will find it worth settling in the rural areas [54].

of drought and epidemics.

for agricultural purposes [53].

the end of the given period.

marketing have to be created.

to urban areas [55].

products at higher prices.

towards sustainable agricultural production.

maintained to avoid bush encroachment.

which leads to the wrong outcome as the intended.

The alternative priority weights with respect to the goal were calculated. The bigger the weight is, the more the measure is considered as important. Ranking the calculated priority weights with respect to the goal gave an insight into which AEM is considered important. Some rankings have more than one AEM, which indicates that the AEM for the particular ranking are considered to have the same importance. Weights of criteria show that production and economic consequences (0.425) are considered to make the most substantial influence on the assessment of AEM, followed by promotion of environmental friendly agricultural practices (0.333); improvement of rural areas to prevent their marginalization (0.241) is on third place. All three criteria are an integral part of the efforts towards improving the rural areas and the environment because they address different aspects of these efforts. The aggregated weights for the criteria and subcriteria of the group model are shown in **Figure 1**.

Further, the control of consistency was done by controlling the consistency index (CI) calcu‐ lated by expert choice. The CI of 0 shows that the assessment results are consistent. The overall results (priorities) for each AE measure are shown in **Figure 2**.

**Figure 3.** Sensitivity analysis with an increased weight of production criteria.

The ranking shows that organic and integrated agricultural production practices make the best contribution to promote sustainable agriculture and protect the environment (**Figure 3**). Permanent green cover on fallow land and greening of arable land are measures that contribute to reducing soil erosion and improving soil quality and fertility. Permanent green cover in water protection areas is vital to prevent the pollution of drinking water and its sources. Also considered as important is bird conservation in humid extensive meadows of Natura 2000 sites, which contributes to stopping the decline of biodiversity. Preserving and maintaining extensive meadows of Natura 2000 sites will not only secure natural habitats for birds, but also for other indigenous wild animals and plants. Almost equally important to promote environ‐ mental friendly agricultural production practices are stopping the decline of biodiversity, reduction in discharging chemicals into the environment and preventing pollution of drinking water and its sources. Improvement of the rural areas is best achieved by creating employment. Conservation of agricultural land is also an important factor to prevent the marginalisation of the countryside.

Stopping the decline of agricultural biodiversity through rearing of indigenous and traditional domestic animal breeds and production of indigenous and traditional agricultural plant varieties is also considered as important. These are key measures towards sustainable agricultural production and food security.

The importance of AEM for the individual criteria and attributes can be shown by synthesizing their priority weights at each node of the hierarchy levels. These values are seen in **Figure 3**. The sensitivity analysis, a test of the reaction of the agri-environmental measures to a change of the priority weight of the objectives, shows no significant changes in the overall ranking of importance of the measures.

By varying the priority weight of each objective, making it most important as shown in **Figure 3**, a slight change in the importance of the agri-environmental measures is only visible for each objective. **Figure 2** shows the highest overall performance in organic production. Integrated crop production, greening of arable land, and preservation of crop rotation are the next important measures.

The criteria weights are 33.3% for promoting environmental friendly agricultural practices, 24.1% for improving the rural areas to prevent marginalisation and 42.5% for production and economic consequences. The organic fruit and vine production and organic horticulture are the most important measures with 6.5% each, followed by organic crop production with 5.2%, integrated crop, fruit, vine production and integrated horticulture with 3.6% each. Bird conservation in humid extensive meadows of Natura 2000 sites, permanent green cover in water protection areas, and on fallow land and greening of arable land got 3.4% each. The preservation of crop rotation was allocated a weight of 3.3% and reduction of soil erosion in fruit and wine growing 3.0%. By changing the priority weight of the criteria: promote envi‐ ronmental friendly agricultural practices to 49.0%, organic fruit and vine production, organic horticulture kept their leading position as most important measures but their weights in‐ creased to 7.5% each, followed by organic crop production which also got a bigger weight of 5.6%; integrated crop, fruit, vine production and integrated horticulture each got 3.7%, a weight bigger than in primary assessment. Bird conservation in humid extensive meadows of Natura 2000 sites and permanent green cover in water protection areas and on fallow land each has a weight of 3.6%. Greening of arable land went down to 3.3%, whereas the preser‐

**Figure 3.** Sensitivity analysis with an increased weight of production criteria.

46 Applications and Theory of Analytic Hierarchy Process - Decision Making for Strategic Decisions

The ranking shows that organic and integrated agricultural production practices make the best contribution to promote sustainable agriculture and protect the environment (**Figure 3**). vation of crop rotation reduced to 3.1% and reduction of soil erosion in fruit and wine growing to 2.6%.

By changing the weight and priority of the objective: improve the rural areas to prevent marginalisation to 49.0%, the ranking of the measures still stayed the same. Organic fruit and vine production, organic horticulture kept their leading position as most important measures, followed by organic crop production with 4.8%; integrated crop, fruit, vine production and integrated horticulture kept their weights of 3.7%. Bird conservation in humid extensive meadows of Natura 2000 sites, permanent green cover in water protection areas and on fallow land, greening of arable land and preservation of crop rotation each got a weight of 3.4%, reduction of soil erosion in fruit and wine growing 3.1%.

By changing the weight and priority of the objective: production and economic consequences to 49.0%, the there was no significant change in the weights of the measures compared to Figure 6. Organic fruit and vine production, organic horticulture kept their leading position as most important measures with weights of 6.4% each, followed by organic crop production with 5.2%; integrated crop, fruit, vine production and integrated horticulture kept their weights of 3.6% each. Bird conservation in humid extensive meadows of Natura 2000 sites, permanent green cover in water protection areas and on fallow land show weights of 3.4% each, greening of arable land 3.5%, the preservation of crop rotation each 3.4% and reduction of soil erosion in fruit and wine growing 3.1%.

However, the organic fruit production is seen as the most important agri-environmental measure is specific for Slovenia because of its geographical features. Vrišer [56] states that in the census of the agricultural sector made in 2000, the proportion of Slovenia's total surface area of plains and low hills amounts to 36.4%, on which 54.5% of the utilised agricultural area is found, whereas on the karst regions that occupy about 25.3% of the total surface area, there is only 17.5% of utilised agricultural area, and in the high mountains (10.8%), only 3.5%. Vrišer also noted that 2.6% of the agricultural area was used for fruit production. In 2006, Sušnik et al. [57] still noted that fruit is grown on 2–3% of all agricultural land in Slovenia. This shows no increase in the fruit-growing area.

Despite limitations (particularly in the field of data acquisition), we found that the approach full field our expectations in the field of AEM assessment. The methodology results in precise AEM ranking with respect to defined criteria priorities. Furthermore, the group approach and priorities aggregation enable inclusion of the large number of experts relevant for the analysis.

### **4. Conclusion**

The attempt in this paper was to show how AHP can successfully be employed in agriculture by assessing the role of agri-environmental measures to improve agriculture and the coun‐ tryside. The goal, criteria and their attributes were compiled by the government of Slovenia. Arranging them in a hierarchy helped to analyse their interactions within the hierarchy and with respect to the main goal. The correct implementation of the necessary steps in AHP results A Case Study on the Application of the Analytic Hierarchy Process (AHP) to Assess Agri-Environmental Measures of the Rural Development Programme (RDP 2007–2013) in Slovenia http://dx.doi.org/10.5772/63924 49

in very comprehensive data, which can be used to verify the goal. Very decisive for the data to be representative is its collection either by questionnaires, brain storming or discussions. Since the AHP process mostly involves a group of experts, stake holders or other persons affected, the collected data have to be compressed by building the geometric mean. In this example, it was successfully done. Though it is time consuming, it is the best way to take the different opinions involved in the assessment procedure into consideration. Compressing the data is necessary to be able to feed the information into the computer software programme Expert Choice™.

Instead of just using SWOT, a form of MCDA could have therefore been successfully used to evaluate the AEM. AHP could have combined different interests, expertise and opinions of all the government institutions involved. The employment of Expert Choice™ could have delivered reliable information. The results obtained in the assessment clearly show that organic and integrated production methods are seen to contribute most to achieving the set environ‐ mental goals and enhancing sustainable agricultural production. At the same time, measures that contribute to stopping the decline of biodiversity and preventing contamination of drinking water and its sources are also seen as an integral part of agricultural activities. However, the results generated by AHP do not end debates on further action regarding environmental friendly agricultural practices and policy. They are a good basis for further discussion.
