Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1598-7248 (Print)
ISSN : 2234-6473 (Online)
Industrial Engineering & Management Systems Vol.17 No.2 pp.318-326

Analytical Network Process to Prioritize the Influencing Parameters on Local Participation: The Development of Livestock Drinking Water Resources

Iman Islami, Amir Sadoddin*, Hossein Barani, Ahmadreza Asgharpourmasoule, Masih Akhbari
Ph.D. Graduate of Gorgan University of Agricultural Sciences and Natural Resources, Iran
Gorgan University of Agricultural Sciences and Natural Resources, Iran
Ferdowsi University of Mashahad, Iran
Colorado Water Institute, Colorado State University, Fort Collins, USA
Corresponding Author, E-mail:
December 31, 2016 February 3, 2017 March 21, 2017


Participation of local ranchers for developing livestock drinking water resources has been identified as inadequate in Yazd Province located in the central part of Iran. Thorough understanding of barriers among various factors influencing the public participation is complex. In this study, to evaluate and prioritize the influencing criteria and parameters, ANP was used in which relationships are considered in form of non-linear. The information required in this study were collected using tools of questionnaire and expert knowledge elicitation. The results show a combined effect of various criteria on level of participation with different relative weights. Economic criterion in this study with normal weight of 0.455 represents 1.5 times more importance compared with climatic criterion and 2 times more importance in comparison to social criterion. Analysis indicates that following parameter of “Low income of animal husbandry” with normal weight of 0.319, as the most important influencing economic factor, climatic parameter of “Recurrent droughts and reduced rainfall” with normal weight of 0.153, and social parameter of “Poor role of NGOs” with normal weight of 0.093, respectively represent the most important parameters in the context of this study. ANP model facilitates assessment and provides more detailed information for the decision makers.



    Studies in Yazd province in the central part of Iran show that public participation plans are weak and faced with low acceptance of beneficiaries (Yazdi and Khaneiki, 2007). Due to the growing social conflicts rooted in scarcity of water resources in arid ecosystems such as Yazd Province, the development of public participation has been emphasized as an important tool to confront with environmental threats and to benefit from the capacity of local communities (Piran, 2005). Participation is regarded as a central requirement in many cases; in the sense that it allows the micro actors, to enter the macro decision-making. People can overcome their problems, become self-supporting and solve their basic needs through partnerships and without need to formal programs (Papoly Yazdi and Ibrahimi, 2006). However, promising future, will occur when people are eager to participate in the development and utilize their resources in the way of development of society (Agwu, 2005). This interest occurs when the villagers access financial resources and sufficient powers to conduct their projects. Otherwise, participation will be quickly replaced by indifference and will lead to social alienation (Ankony and Kelley, 1999). The goal of the local community is to motivate people to organize themselves in the institutions and associations. In case of such participatory if people avoid social action, their community participatory will be weakened, social awareness will be decreased and negative individualism and self-interest will be increased in the society. But if trust replace instead of distrust and conflict, people’s collaborate will be increased and the social basis will be strengthened (Putnam, 1995). Factors that increase people’s participation are not yet well understood. Much research has been done in the world on the factors influencing participation, each in turn, have pointed to several factors. Many researchers focused on the impact of demographic and personal characteristics such as age, gender, education, household size, income, and ownership (e.g. Bagdi, 2005; Dolisca et al., 2006; O’Faircheallaigh, 2010, Islami et al., 2013). Some researchers have also referred to the role of knowledge and the factors increasing it such as educational courses and appropriate informative systems in the partnership (e.g. Bond, 2014).

    Some people have noted the propensity of individuals and its impact on participation. They believe that the positive attitude toward an issue or a plan increases participation in the project (e.g. Vicente and Reis, 2008; Kerchner et al., 2010). Some researchers have also studied the role of social concepts and theories as factors involved in participation (e.g. Arnold and Fernandez- Gimenez, 2007). Obviously, a range of factors can be named and to achieve lasting partnership each of them must be eliminated separately and linked together (Islami et al., 2017). The point that should not be neglected is the social factors influencing participation, they often depend on and are affected by different aspects of rural and local life; it makes it difficult to identify them (Islami et al., 2017). Meanwhile, Modeling with the defined methodology simplifies scientific understanding of the reality and improve sociological insights into the reality (Islami et al., 2017). In recent years, the multi-criteria decision-making methods are considered in order to select the effective factors and alignment of executive measures and guidelines (Karimi Sangchini et al., 2017).

    MCDM involves a series of techniques, including the weighting or convergence analysis; it allows experts and interest groups rate criteria of a topic. These criteria can be quantitative or qualitative or a combination of both (Pohekar and Ramachandran, 2004). The key to the use of decision models is to choose the appropriate method (Pan et al., 2000). The important factors in choosing the right model are the nature of dependence of factors (linear or network), qualitative or quantitative parameters, positive or negative effect of parameters and the need to acquire information from decision maker (Knoeri et al., 2011). The selection of the right model among the different techniques (e.g. AHP, ANP, and TOPSIS) is so important in logical decision-making, solving a problem and a desired outcome because each of these techniques has their own algorithms. The analytic network process (Saaty and Takizawa, 1986) is used in this study, which enables the management of the interdependence between the components and the network (Lee and Wu, 2005). Some of the researches in the field of natural resources that were conducted by using network analysis are: environmental hazards planning and decision-making in situations of crisis or emergency (Levy and Taji, 2007) the application of network analysis in determining the strategies of participation (Cheng and Li, 2007), identification and prioritization of accident-prone spots (Shafabakhsh et al., 2012), the use of network analysis to assess strategies for combating desertification (Sadeghi and Khosravi, 2015). The present study used the multi-criteria decision-making model and the model of analytic network process (ANP). Our purpose is to properly evaluate parameters influencing participation and identify the parameters contributing to low level of participation of local communities in Yazd province in the development of livestock’s drinking water sources.


    2.1. Study Area

    In this research, the study area includes the rangelands of Yazd province, located in central part of Iran (Figure 1), with about 5.4 million hectares area.

    Although in terms of vegetation cover, rangelands of Yazd province are mainly in moderate and weak condition, they are of considerable importance in the livelihoods of rural households. 4,853 households use these lands directly to provide for their families. The low levels of annual rainfall in Yazd province (less than 100 mm) and poor distribution of water resources are con-sidered as the main constraints of farmers and ranchers in this province. Much of the rainfall occurs in the win-ter; in terms of climatic situation, Yazd is situated in dry and cold arid climate.

    2.2. Analytical Network Process

    The Analytical Network Process is the result of the development of the analytic Hierarchy Process by Saaty (Saaty and Takizawa, 1986) where interactions between and among elements of the decision are con-sidered. The linear and AHP relationships are turned into the network model. In this case, the interdependence (two-way feedback) is included. Problem-solving steps by the network analysis methods include 1. Formation of network structure, 2. Pairwise comparisons, adapta-tion control of judgments, 3. Formation Super Matrix and the limit Matrix, 4. Prioritization of criteria and alternatives. Fig-ure 2 shows the difference between network structure and hierarchical structure.

    2.2.1. The Formation of Network Structure of Local Community Participation Issue in the Study Ar-ea

    To solve the problem we should first identify and draw a network of purpose, criteria, alternatives and relationships among them. The purpose here is the same as the problem of the study. The problem being ad-dressed in this study is described as the low participation of local communities and livestock utilization in the development and supply of water resources of cattle in the pastures of Yazd province. The Delphi Technique (Jeffery et al., 1995) was used in this study in order to effectively identify and structure the criteria and alter-natives. Experts consisted of a group of 15 men having the experience and knowledge of livestock farming and were interviewed in this regard. The group was selected in a completely targeted way using the snowball tech-nique; after theoretical saturation, the group consisted of expert population of Yazd province. The expert group also participated in the judging of network analy-sis’ questionnaire. After identifying the criteria and pa-rameters, each criterion was drawn as a network struc-ture that consists of different clusters using Delphi tech-nique. Clusters are the same levels of decision-making (including the issue of the study, criteria, and alterna-tives). Alternatives are clusters that contain the alterna-tives obtained from the Delphi technique. After this step, all communications and dependencies whether external (relationships between two clusters) or internal (intra-cluster elements) should be determined. The links are displayed by flash.

    2.2.2. Paired Comparisons of Criteria and Alternatives

    After network formation, in the second step, the pair pairwise comparisons between the decision factors (in-cluding the criteria and the relevant or interaction crite-ria) were obtained in the form of a questionnaire - using the scale of preference from 1 to 9 (Table 1). To carry out the needed judgments and obtain pairwise compari-sons, the opinion of experts in the field of economic, social, environmental and livestock farming in Yazd province were received. Then to combine judgments of the experts, the geometric means were used in this method to maintain the vice versa property of compari-sons (Ghodsi, 2002). An important point that should be considered in judging and comparing is compatibility control. This possibility is allowed by the Super Decision software in which the network analysis is done with pre-cision and ease- regarding the large volumes of data. In this case, the inconsistency rate for any given pair-wise comparison is determined. If people judge in a contra-dictory way, their inconsistency rate will exceed 0.1 showing inconsistent judgment (Ghodsi, 2002).

    2.2.3. The Formation of the Super Matrix and Turning it Into the Limit Matrix

    At this point, we need to insert values obtained from the pairwise comparison in the form of a ques-tionnaire into the network analysis software. The appli-cation provides the possibility to analyze qualitative or quantitative data.

    Therefore, at this stage, weight values obtained through pairwise comparisons in the previous step are included in large matrix called Super matrix. Super ma-trix which is obtained in this step is called initial matrix or unweighted Super matrix. In the next stage, weighted super matrix is obtained through multiplying the corre-sponding weights of clusters (criteria) by unweighted super matrix. In the next stage, the weighted super ma-trix is multiplied by itself until its lines are inclined to a constant and equal number (Saaty, 1996).

    In this case, limit super matrix is obtained. Using probability matrix and based on Markov chain, Saaty proved that the final weight (the limit matrix) is ob-tained from equation (1). In this equation, W is the final weight of elements, w is standardized weighted matrix and k is an odd number.

    W = Lim w 2k+1

    2.2.4. Prioritization of Criteria and Alternatives

    If the composed limit super matrix contains the weight of criteria and alternatives, the model will cover a whole network. Therefore, this limit matrix provides the probability of prioritizing alternatives and modeler is able to identify the alternatives that have the greatest weight. Generally the solving of every problem, has the complexity of its own.

    3. Results

    3.1. Modeling the Factors Affecting the Weakness of Social Partnership and the Formation of a Network

    Primarily based on the opinions of experts who had the necessary experience on the issue of the studied area, vital parameters were selected from a range of parameters. The parameters in this study are elements constituting the criteria

    The parameters that were used for structuring the problem of poor public participation were identified by consensus of experts in a three-stage Delphi technique and through interviews. Then, using the identification of criteria and alternatives, a network structure was formed in the Superdecisions software (version 2.2). Arrow lines in this structure (Figure 3) indicate the net-work of interacinteractions between the criteria and between the criteria and alternatives. The first level of this structure is the problem of the study that is replaced with the purpose to lead to greater understanding of the subject. This level can be eliminated in this structure because it does not influence or affect the network. Cluster (level) II is the criteria cluster.

    As mentioned earlier, because of the complexity of social problems, they are not regarded as the only social criteria i.e. the economic and geographical criteria (envi-ronmental) can also be effective, therefore, they should be identified and taken into account as well. Hence, the level of external dependence and interaction between the criteria were conducted and shown by Bidirectional flash. In the third level, the network structure of formed parameters was included in the cluster of alternatives. The interdependence between parameters was observed in this sector shown by a ring flash. The observance of the interdependence provides the possibility of their preferences in a pairwise manner.

    3.2. Pairwise Comparisons, Compatibility Control and Priori-tization Criteria

    By extracting the judgments of experts through pairwise comparison questionnaires and integrating their views based on geometric mean, the relative weight (most important criteria and alternatives) were estimat-ed. Figure 4 shows an example of a pairwise compari-son in the Super decision environment.

    Therefore, through the formation of matrices, the importance of the issues was compared with the defined criteria. Table 2 compares the priority criteria based on their gained weight. According to the analyses, the im-portance of economic criteria was higher regarding the issue of weak participation of local communities in the management of drinking water of livestock. According to network analysis, economic, social climatic criteria have had their impacts on the issue with the following weights 0.455, 0.345, 0.198 respectively. Inconsistency rate of this judgment was less than 0.1 showing a com-patible judgment of experts. According to the judgments of experts, the economic social and climatic criteria indicated an importance of 2 and 1.5 times in the sub-ject of study.

    3.3. Formation of Super Matrix and the Limit Matrix

    To prioritize the constituent parameters (located on the third level of the decision) being identified as alter-natives in the software, the limit super matrix should be obtained through pairwise comparison matrix. As men-tioned in the previous section, the limit super matrix is obtained by being multiplied by the weighted matrix and the weighted matrix is obtained by the normal un-weighted matrix. As shown in Figure 5, the numbers are equal in the limit super matrixes’ lines. Table 3

    3.4. Prioritization OF Alternatives

    After determining the limit matrix, we can obtain the final synthesis of network analysis model. Prioritiza-tion of the parameters affecting the weak participation of local communities in the management of drinking water of cattle in Yazd is the result of network model built in this study. According to the ratings, the economic parameter “Low income of ranchers” with the normal weight of 0.318 is the most important parameter in poor participation of ranchers. After the income parameter, parameters of “frequent droughts and reduced rainfall” “with a normal weight of 0.153 and the “weak role of local NGOs” with a normal weight of 0.094 are the most important parameters in the subject of study. The results show that “inappropri-ate distribution of water” Farmers’ financial depend-ence on the government “and the lack of appropriate legislation” are the most important reasons for the low participation of ranchers. In this table, the role of fac-tors such as “conflicts,” disputes “and implementation of sanctions and penalties” have had a negligible im-pact on the low participation, from the perspective of respondents. Bar graph (Figure 6) shows the compare weight of parameters affecting the participation of local communities in Yazd Province.


    The important result of study is the combined ef-fects of different criteria with different relative weights in the partnership being compatible with the results of previous studies, including Islami et al. (2017) and Mohseni Tabrizi et al. (2006). Therefore, total of three social, economic and climatic criteria with three differ-ent weights have shared their effects in the phenome-non of participation. Double importance of economic criteria compared to social criteria in this study, shows that participatory social process is influenced by pre-vailing economic conditions and livelihood of livestock farmers in the province. So if we want the success of participatory management plans based on broad partic-ipation of local community, we need to consider the economic situation of ranchers of Yazd province which consist their livelihood. The adequate income provides the possibility for the rancher to enter the decision-making and implementation sector as one of the most important types of participation process without having to rely on other financial sources. This result confirms previous studies (Papoly Yazdi and Ibrahimi, 2006 and Islami et al., 2017; Karimi Sangchini et al., 2017) that pointed out that the importance of economic issues and effect of this criterion in comparison to other criteria. The results of prioritization also make clear that a favorable outcome in the partnership is achieved when the government provides effective support for ranchers through overcoming drought conditions (impact of cli-matic criteria). The government also is responsible for promotional activities in the field of participation. From the perspective of experts, the supportive role of gov-ernment in the development of water resources is in lower priority confirming the government’s good per-formance in this area. Despite the emphasis on partner-ship in recent years, enough organizations and local NGOs are not formed in Yazd province. In addition, the formed organizations do not have appropriate efficien-cy. This problem is a serious obstacle in the way of achieving the participatory of local communities caus-ing the high weight in the rating system of network analysis. The lack of a strong association with a mediat-ing role between people and government has caused the customary and legal rights of ranchers be one of the major reasons for their low participation. The lack of this important organ as a mean of efficient partnerships (Gezon, 1997) has prevented the formation of neces-sary social pressure on the government to pass support-ive- deterrent laws. However, as Putnam (1995) has stated the factor preventing the formation of associa-tions and NGOs is the low and weakened trust of people toward each other. This parameter is placed in the 8th row of the top effective parameters. In these circum-stances, the weak bonds of trust are destroyed and have caused conflicts be-tween farmers and mine owners within the rangeland of Yazd; this is largely the result of the lack of crime deter-rent punishments in dealing with protests, especially in the field of law. Strengthening the bonds of trust be-tween the community and the government paves the way for wider participation in the development of pas-tures. The power of local community gradually is in-creased under the listed conditions and their social in-fluence is also multiplied by putting pressure on gov-ernment agencies to meet the customary and statutory needs and demands. However, as it is emphasized in previous studies, including Mohseni Tabrizi et al. (2006) the sustainable social participation will not be fulfilled until these barriers are removed in conjunction with each other. This study is based on the methodology de-fined by ANP and helped to identify a set of parameters and prioritize them. So as a final conclusion it can be regarded that challenges of Public Participation in this study, like many complex social issues is affected by a variety of social, economic and climatic criteria. These measures are not independent so a network of influ-ences and impressions occurs. Network analysis model in this study was used to rate the parameters reasonably and without the assumption of independence of criteria and parameters (Saaty and Takizawa, 1986). In addi-tion, the interdependence between parameters was em-phasized in various studies, such as (Lee and Wu, 2005). This interdependence was also emphasized in this study to make the prioritization more accurately and more reliably. The study involved effectiveness factor of a set of parameters in participation and used a modeling ap-proach based on network analysis to prioritize challeng-es. The findings of this research can help decision-makers to remove barriers of the participation of the local livestock community in Yazd province.



    The location of Yazd Province in Iran.


    Comparison of the structure of analytical hierarchy process (AHP) and analytical net-work process (ANP) (Ghodsi, 2002).


    Network structure of effective parameters in the design of farmer participation of local communities in Yazd province by using superdecisions software.


    An example of a pairwise comparison in the Superdecisions software.


    The limit super matrix being formed in the ANP model of public participation.


    Compare weight of parameters affecting the participation of local communities in Yazd Province with ANP model.


    Standard scoring system in network analysis

    Comparison of economic, social and climatic criteria influencing the weakness of public participation

    Prioritization of parameters affecting the participation of local communities in Yazd Province


    1. Agwu, E. (2005), Rural peoples attitude towards participation in development project: A study of NGWO community in Enugu State, Nigeria , Journal of Rural Development, 24(2), 271-286.
    2. Ankony, R. C. and Kelley, T. M. (1999), The impact of perceived alienation on police officers’ sense of mastery and subsequent motivation for proactive enforcement, Policing : An International Journal of Police Strategies and Management, 22(2), 120-132.
    3. Arnold, J. S. and Fernandez-Gimenez, M. (2007), Building social capital through participatory research: An analysis of collaboration on Tohono O’odham tribal rangelands in Arizona , Society and Natural Resources, 20(6), 481-495.
    4. Bagdi, G. (2005), People’s participation in soil and water conservation through watershed approach, Indian international book distributing (IBDs), Indian, 311.
    5. Bond, J. (2014), A holistic approach to natural resource conflict: The case of Laikipia county, Kenya , Journal of Rural Studies, 34, 117-127.
    6. Cheng, E. W. L. and Li, H. (2007), Application of ANP in process models: An example of strategic partnering , Building and Environment, 42(1), 278-287.
    7. Dolisca, F. , Carter, D. R. , McDaniel, J. M. , Shannon, D. A. , and Jolly, C. M. (2006), Factors influencing farmers’ participation in forestry management programs: A case study from Haiti , Forest Ecology and Management, 236(2-3), 324-331.
    8. Gezon, L. (1997), Institutional structure and the effectiveness of integrated conservation and development projects: Case study from Madagascar , Human Organization, 56(4), 462- 470.
    9. Ghodsi, P. H. (2002), Analytic Hierarchy Process (1st Ed.), Amirkabir University of Technology, Tehran.
    10. Islami, I. , Sadoddin, A. , Barani, H. , Asgharpourmasouleh, A. , and AKhbari, M. (2017), Modeling socio-ecological structure of local communities participation for managing livestock drinking water using the agentbased approach , Applied Ecology and Environmental Research15(3), 1173-1192.
    11. Islami, I. , Ghorbani, M. , Jafari Shalamzari, M. (2013), Identification and prioritization of factors influencing agricultural water price index from farmers’ viewpoint in Charkhab Village, Yazd Province , Desert, 17, 161-167.
    12. Jeffery, G. , Hache, G. , and Lehr, R. (1995), A groupbased Delphi application: Defining rural career counseling needs , Measurement Evaluation Counseling Development, 28(1), 45-60.
    13. Karimi Sangchini, E. , Islami, I. , Farajollahi, A. , Arami, S. A. , and Jafari., M. (2017), Development and prioritization of socio-economic strategies to elevate public participation in natural resource management using TOPSIS approach; Case Study: Chaharmahal and Bakhtiari Province (Iran) , Journal of Applied Sciences and Environmental Management, 21 (3), 476-485.
    14. Kerchner, C. , Honzak, M. , Honzák, R. , Richadson, A. , Townsend, J. , Rimmer, C. C. (2010), Designing spatially explicit incentive programs for habitat conservation: A case study of the Bicknell’s thrush wintering grounds , Ecological Economics, 69(11), 2108-2115.
    15. Knoeri, C. , Binder, C. B. , and Althaus, H. J. (2011), An agent operationalization approach for context specific agent-based modeling , Journal of Artificial Societies and Social Simulation, 14(2), 4.
    16. Lee, Y. T. and Wu, G. H. (2005), Development strategies for competency models , Proceedings of the International Trade Department, Ta Hwa Institute of Technology, Taiwan.
    17. Levy, J. and Taji, K. (2007), Group decision support for hazards planning and emergency management: A group analytical network process (GANP) approach , Mathematical and Computer Modeling, 46(7-8), 906-917.
    18. Mohseni Tabrizi, A. , Tabatabai, M. , and Mrjaei, H. (2006), Conceptualized psychological theories and approaches to participation , Domestic Policy, 1(1), 147-175.
    19. O’Faircheallaigh, C. (2010), Public Participation and environmental impact assessment: Purposes, implications, and lessons for public policy making , Environmental Impact Assessment Review, 30, 19-27.
    20. Pan, J. , Teklu, Y. , Rahman, S. , and Castro, A. (2000), An interval-based MADM approach to the identification of candidate alternatives in strategic resource planning , IEEE Transaction on Power System, 15(4), 1441-1446.
    21. Papoly Yazdi, M. H. and Ibrahimi, M. A. (2006), Theories of rural development, Samt Publications, Tehran.
    22. Piran, P. (2005), Water conflicts and conflict management mechanisms in the Middle East and North Africa region, The case of Iran, World bank, 256.
    23. Pohekar, S. and Ramachandran, D. (2004), Application of multi-criteria decision making to sustainable energy planning: A review , Renewable and Sustainable Energy Reviews, 8(4), 365-381.
    24. Putnam, R. (1995), Bowling alone: America’s declining social capital , Journal of Democracy, 6(1), 65-78.
    25. Saaty, T. L. and Takizawa, M. (1986), Dependence and independence: From linear hierarchy to nonlinear networks , European Journal of Operational Research, 26(2), 229-237.
    26. Saaty, T. L. (1996), Decision making with dependence and feedback: The analytical network process, RWS Publications, Pittsburgh.
    27. Sadeghi, R. M. H. and Khosravi, H. (2015), Application of network analysis process (ANP) in assessment of combating desertification alternatives , Desert Ecosystem Engineering Journal(DEEJ), 4(8), 11-24.
    28. Shafabakhsh, G. H. , Alizadeh, H. , and Akbari, M. (2012), Identify and prioritize the Sensitive areas in terms of accident using network analysis process (ANP) , Journal of Traffic Management Studies, 14(24), 1-18.
    29. Vicente, P. and Reis, E. (2008), Factors influencing households’ participation in recycling , Waste Management and Research, 26(2), 140-146.
    30. Yazdi, A. A. S. and Khaneiki, M. L. (2007), The Drought of 2001 and the measures taken by Yazd regional water authority , Proceedings of the International Workshop, Groundwater for Emergency Situations, Tehran, 58-70.