Comparative applicability of MCDM‐SWOT based techniques for developing integrated watershed management framework

Environmental planning facilitates decision‐making to achieve sustainable development goals and provides a crucial way to achieve integrated watershed management (IWM). However, such systematic planning has not been adequately conducted worldwide. Therefore, this study was conducted to develop an IWM framework using SWOT (i.e., strengths, weaknesses, opportunities, and threats) model for the Cheshmeh–Kileh Watershed, Mazandaran Province, Iran. The input components were comparatively weighted using different multicriteria decision‐making (MCDM) techniques, including Game Theory Algorithm (GTA), Best–Worst Method (BWM), VIekriterijumsko KOmpromisno Rangiranje (VIKOR), Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), and Simple Additive Weighting (SAW). Semistructured interviews with stakeholders, including watershed residents, executive experts, and policymakers, were used to identify the SWOT factors. The number of interviewees in subgroups of local users, policy‐making institutions and executive organization were 75, 13, and 6 respectively. Five different MCDM techniques were then used to calculate the overall weight of SWOT factors. According to the results, the groups of SWOT factors of abundant water resources, severe floods, promotion of environmental conservation culture, and increasing migration of young age groups with respective weights of 0.298, 0.298, 0.372, and 0.279 scored higher than other factors. Based on the overall weights, it can be said that positive factors scored more points than negative factors. Therefore, according to the opinions of the stakeholders, the Cheshmeh–Kileh Watershed was in a good condition based on strengths and opportunities compared to weaknesses and threats, and for this reason, GTA and SAW, which had included the watershed status in the growth and development strategy, were selected as the best method. BWM and TOPSIS provided relatively acceptable results, and the weakest result was associated with VIKOR, which showed the watershed status in the defensive strategy. The present study results can help managers in optimal decisions for planning and optimal management in the Cheshmeh–Kileh Watershed to create ecological balance and increase the welfare of watershed residents.

. Therefore, the use of comprehensive and multiobjective methods has been developed due to the complexity of natural systems.
Integrated and sustainable watershed management ensures safe water supply, sustainable use of ecosystems, prevention of land degradation, and restoration of degraded resources to achieve the sustainable development goals (SDGs) (Tang & Adesina, 2022).Accordingly, watersheds are not only a topographic boundary (Lee, 2012) but also a hydrological unit (Wang et al., 2016), and on the other hand, they can be used as an economic, social, and political unit for sustainable management of natural resources (Lee, 2012;Ratna Reddy et al., 2017).In recent decades, the concept of watershed management has been proposed as a sustainable process for managing natural resources and improving socioeconomic issues in human life (Lee, 2012;Wang et al., 2016).On the other hand, watersheds reflect temporal and spatial heterogeneities related to the landscape, which indicates necessary to use these units in the planning and development of natural resources (De Steiguer et al., 2003).In this regard, the IWM goals in watersheds are a multifaceted view of phenomena and increasing the participation of various stakeholders (Tang & Adesina, 2022) for better watershed management and optimal sustainability of water resources (Narendra et al., 2021).
Based on the IWM approach, one of the reasons for the failure of management programs in watersheds is the nonacceptance of activities by the watershed residents and lack of attention to economic and social issues (Tahseen & Karney, 2017).Therefore, there are various methods for reviewing management and systematic plans, one of which is strategic planning models (SPMs).One of the most important SPMs is SWOT (strengths, weaknesses, opportunities, and threats) analysis.This analysis includes strengths, weaknesses, opportunities, and threats, and the matrix evaluation system of this model is one of the most famous SPMs (Mousavizadeh et al., 2015;Nilsson & Dalkmann, 2001).This model is a tool for evaluating and formulating strategies and designing them.A matrix of internal factors (strengths and weaknesses) and external factors (opportunities and threats) has been developed to achieve the optimal decision in this model (Goli et al., 2021;Yang et al., 2018).
A review of the research background shows that the weighting of SPMs using MCDM techniques have been used in many sciences (Amiri et al., 2018;Fan et al., 2023;Yavuz & Baycan, 46°38′ N and 50°23′ and 50°59′ E. The elevation ranges from 131.67 m above sea level (m a.s.l.) at the watershed outlet and 4756.06 m a.s.l. at the Alam Kooh Mountain.Dominant land uses include rangeland, forest, agriculture, water body, and residential development.The Sehezar and Dohezar rivers are the most important rivers of the Cheshmeh-Kileh Watershed, originating from the Takht-e-Soliman, Alamut, and Khashchal mountains.The high capacity of the riverside lands and the limitation of suitable lands in the watershed have caused many agricultural activities to be concentrated along the river, which is severely affected by floods (Khiavi et al., 2022).

| Research methodology
In the present study, a semistructural interview (Goli et al., 2021) was used to collect the primary data.Informal semistructured interviews were conducted.Some questions were predetermined and new questions were created during the discussion.The questions were mostly open-ended and the respondent was given an opportunity to give his opinion.The random sampling method was used (Adhami et al., 2018(Adhami et al., , 2019) ) and the duration of the interview with each person was between 30 and 60 min.Semistructural interviews were carried out with three groups (Adhami et al., 2019).The first group was local users who live in 21 Cheshmeh-Kileh Watershed villages.The second group included executive experts in the General Department of Natural Resources and Watershed Management of West Mazandaran.In the third group, policymakers, including the governor of Tonekabon, the deputy governor, the district of Khorram-Abad, and village councils were interviewed.Interviews were randomly conducted (Adhami et al., 2018).The number of interviewees in subgroups of local users, policy-making institutions, and executive organization were 75, 13, and 6, respectively.It is worth mentioning that to eliminate the effect of the number of interviewees in the final result; the final weights were standardized and dimensionless.
During the interview, stakeholders were asked to identify strengths, weaknesses, opportunities, and threats (Yavuz & Baycan, 2013a) related to the ecological, environmental, economic, and social issues of the watershed.After compiling the internal and external factors (Nazari et al., 2018;Solaymani, 2018), the importance coefficient (Sudia et al., 2021), and the score (Goli et al., 2021) were considered for each factor, and the local weight of each factor was calculated.MCDM techniques were then used to weigh and rank the factors of strength, weakness, opportunity, and threat.These techniques included GTA, BWM, VIKOR, TOPSIS, and SAW.The flowchart of the research methodology is presented in Figure 2. The following is a description of each method.

| GTA
The Condorcet algorithm based on GT was used to weight the SWOT factors.First, in each factor, the ranking was determined based on the values of each criterion.The pairwise comparison matrix of the Condorcet algorithm was used to weight the factors.The Condorcet algorithm was one of the few algorithms based on GT and decision-making that selected the candidate who won the majority of votes in each head-to-head election (Adhami et al., 2018;Sheikhmohammady et al., 2010).
The purpose of this method was to reach a level in which the maximum demand of each player was met, provided that the maximum needs of other actors were met (Skardi et al., 2013;Üçler et al., 2015).Regardless of the voting system, the winner was always the same and was determined for a set of candidates by counting the preferences of the voters two by two (Khiavi et al., 2022).One of the main goals of the Condorcet algorithm was to create groups based on all individual priorities.This function selected the winner for each vote accounting for all pairs of options (Elkind et al., 2011 were summed over n alternatives and m individuals as calculated (Equation 1): (1) Rows of voter preferences from top to bottom are: According to this formation, the framework of the Condorcet matrix is (Equation 2): The winner was determined based on pairwise comparisons by the number of times it was present in the matrix (Adhami et al., 2019).In Equation (2), option B is the most numerous, the winner.

| BWM
BWM was used to weigh the SWOT factors.The first stage, SWOT factors, were selected to apply BWM (Rezaei, 2016).The best factor and the worst factor were then selected.The preference of the best criterion over other criteria was introduced with a number between 1 and 9 based on the following equation (3) where a Bj indicates the superiority of the best B criterion over j criteria.The previous step was repeated for the worst criterion, and the corresponding preference of other criteria over the worst criterion W was accordingly determined using Equation ( 4) where a jw expresses the degree of superiority of the j criterion over the worst criterion of w.
The optimal weight vector in the form of a vector (w w w *, *, …, * n 1 2 ) was formed as shown in Equation ( 5) such that Equation (6) gets satisfied: The min-max model in Equation ( 5) was written as Equation (7) to fulfill Equation (8):

| VIKOR
The VIKOR is one of the MCDMs based on the agreed planning of problems (Chen & Wang, 2009).If there is an n criterion and m alternative in a multicriteria decision problem, to select NASIRI KHIAVI ET AL.
Natural Resource Modeling | 7 of 27 the best alternative using this model, the steps of the VIKOR implementation algorithm are as follows (Opricovic & Tzeng, 2007): The first step is to define the criteria and create the matrix (Arif et al., 2020) as Equation ( 9): In the next step, after forming the decision matrix, the normalization of this matrix was performed based on Equation (10).Dimensioning was used to accommodate different measurement scales (Shekhovtsov & Sałabun, 2020): The criteria weight vector was then determined.At this stage, according to the importance coefficient of different criteria, the vector was defined as below equation GT was used for initial weighting.After the matrix presented in Equation ( 10) was normalized, the weight of each of the indices obtained from GT was multiplied, and then the normalized matrix of Equation ( 12) was obtained where R ij is the normalized matrix and w j is the weight of the criteria calculated by GT.In the next step, the best and worst criteria were determined.The best value ( f * j ) and the worst value ( f j − ) for the criteria were calculated from Equations ( 13) and ( 14), respectively (Haider et al., 2020;Vakilipour et al., 2021): In the sixth step, the amount of usefulness or maximum desirability (S) and the lack of desirability (R) were quantified by the following equations (Salimi et al., 2020) W i is the weight for criterion i and f ij for each criterion.VIKOR index (Q i ) was then calculated by following equation (Salimi et al., 2020): Finally, the alternatives were arranged according to the values of R, S, and Q in three groups from the smallest to the largest, and finally, an alternative was selected as the best alternative, which was recognized as the superior alternative in all three groups (Opricovic & Tzeng, 2004;Papathanasiou & Ploskas, 2018).

| TOPSIS
This model was first proposed by Hwang and Yoon in 1981 and is one of the best MCDM techniques (Krishnamurthy et al., 1996;Raviraj et al., 2017).First, using Equation ( 18), the existing decision matrix became a dimensionless matrix (Shekhovtsov & Sałabun, 2020) where r ij is the value of each criterion for each alternative and N is the dimensionless matrix, and w is the last option.The dimensionless matrix was then multiplied by the diameter of the weights matrix to create a weighted matrix (Sałabun, 2013;Vakilipour et al., 2021): where V is a weighted matrix and V mn is an array of matrices.The ideal positive and negative solutions were then calculated using Equations ( 20) and ( 21) (Shekhovtsov & Sałabun, 2020): where max and min are the maximum and minimum of each criterion for each alternative.In the next step, using Equations ( 22) and ( 23), the distance of each alternative from the positive and negative ideals was calculated (Chung et al., 2017):  ( ) Finally, the relative proximity of an alternative to the ideal solution was determined using Equation ( 24), and the alternatives were ranked (Li et al., 2018).

| SAW
It is one of the simplest and most widely used MCDM techniques (Salehi & Izadikhah, 2014;Serrai et al., 2017) and is suitable for evaluating the results of other techniques (Wang et al., 2016).To apply this method based on Equation( 25), the decision matrix was normalized: where Ω max and Ω min are a set of positive and negative decision alternatives, respectively.The weight of decision factors was then quantified, and the ranking value was calculated based on Equation (26).Finally, the highest value of S i was selected as the best alternative (Ozkaya et al., 2021;Sotoudeh-Anvari et al., 2018;Vakilipour et al., 2021).
Using the five methods described above, the SWOT factors were first weighed.From the multiplication of weights obtained in local weight, the overall weight of the factors was calculated.The SWOT status assessment matrix was then used to assess the status of the Cheshmeh-Kileh Watershed in four strategies.These four strategies included offensive, defensive, conservative, and competitive strategies (Yavuz & Baycan, 2013a).According to each method, the status of strategies was determined by subtracting internal and external factors (Sudia et al., 2021).Finally, a correlation test was used to compare the results of the techniques used.The initial watershed status matrix (based on stakeholder opinions in semistructural interviews) was used to determine the best technique for weighting SWOT factors.

| RESULTS
The results of compiling, and calculating the importance coefficient, score, and local weight of the SWOT factors were presented in Tables 1 and 2. In Tables 3 and 4, pairwise factor comparisons were presented using the Condorcet algorithm based on GT.
The results related to the weighting of the factors using BWM are also listed in Table 5.In this technique, the best criteria for strength, weakness, opportunity, and threat groups were S4, W1, O6, and T4, respectively.The worst criteria were also selected as S9, W6, O4, and T3, respectively.
Tables 6-8 presented the results of ranking the factors based on VIKOR, TOPSIS, and SAW techniques.The results related to the quantification of final weights and total SWOT factors were presented in Table 9.
The evaluation matrix and determination of watershed status based on the total weights of each method were also presented in Figure 3. Based on Figure 3, the status of the studied watershed was examined in four strategies.Finally, the correlation test results related to the comparisons of MCDM in weighting the factors are presented in Table 10.The strategic planning model is one of the best tools used to analyze any system's status and provide strategies to improve and develop its performance (Gharachorloo et al., 2021;Nutley & Reynolds, 2013).Through the evolution of watershed management, the application of IWM, using SPMs, and MCDM techniques in optimal decision-making has become more prominent.
According to Chang and Huang (2006), the use of strategic planning models effectively examined the management situation in the watershed.Based on results 1 and 2, MCDMs were significantly expanded and were very useful in resolving conflicts and optimal decision-making.
IWM is more complete based on economic, social, technical, and environmental aspects (German et al., 2007).Wang et al. (2016) confirmed that this comprehensiveness led to a multidimensional and consistent planning and management process aimed at balancing the economic, social, and environmental conditions of the watershed.Bakker (2012) also stated that IWM is not only based on hydrological phenomena in the watershed, but also has a comprehensive view of the multiple uses of watershed resources, which in addition to assessing human and environmental needs, also support ecosystem services and biodiversity.Environmental planning facilitates decision-making to achieve sustainable development goals by considering political, economic, social, technological, environmental, and legal factors and provides a crucial way to achieve IWM (Khalifipour et al., 2012).Strategic planning and management requires many decisions based on multiple criteria.Applying a single criterion in management discussions is not acceptable, and in many cases, to achieve the best solution, the analysis must be done based on several criteria.The selection of optimal decision criteria is critical in multicriteria analysis (Srdjevic et al., 2012a).Cakir et al. (2020), Christodoulou and Cullinane (2019), Sarsby (2016), and Vorthman (2008) argue that SWOT is a system for identifying internal and external factors.
The main strength of the Cheshmeh-Kileh Watershed was the existence of permanent springs and abundant water resources.On the other hand, the natural and beautiful landscapes of the Sehezar and Dohezar Rivers made them tourist-friendly.However, the severe floods and high slope upstream of the watershed are critical weaknesses of this watershed (Table 1).Also, the high culture of residents to conserve the environment and the high potential for T A B L E 3 Pair matrix of internal factors to quantify the initial weights of SWOT factors using GTA, the Cheshmeh-Kileh Watershed, Iran.
aquaculture were the most important opportunities for this watershed.Unfortunately, due to economic and livelihood problems, migration of young age groups has increased in recent years, as well as a decrease in rainfall and an increase in drought (Table 2).The ranking of SWOT factors using MCDM techniques had different and similar results (see Tables 3-8).GTA and SAW rankings were the same, but other MCDM techniques, including BWM, VIKOR, and TOPSIS, offered different rankings.It is important to note that the strengths were high in many techniques, indicating that the watershed was in good condition.In GTA and SAW, the factors of opportunity, strength, threat, and weakness are deducted from the first to fourth priorities.In BWM, the strength, threat, weakness, and opportunity factors gained the most and the lowest weights, respectively.VIKOR technique showed different results, according to which the weaknesses and threats were ranked high.However, the TOPSIS technique showed that the points of threat, strength, opportunity and weakness were ranked first to fourth.Based on the overall weights presented in Table 9, in GTA and SAW, factor O6 won first place with a weight of 0.126.In BWM, factor S6 gained the most weight.In the VIKOR, different results of overall weights were presented, in which the negative aspects of the watershed gained high weight, based on which factors W5 and W9 showed the highest weight.
T A B L E 4 Pair matrix of external factors to quantify the initial weights of SWOT factors using GTA, the Cheshmeh-Kileh Watershed, Iran.In the TOPSIS technique, the negative aspects obtained a high score, with factor T1 gaining the highest weight.
The watershed status assessment matrix also showed that based on the techniques of GTA and SAW, and growth and development and based on SWOT factors, this watershed was in good condition.Nevertheless, in VIKOR, the watershed was a defensive strategy.However, based on BWM and TOPSIS, the competitive-offensive strategy in this watershed was the dominant strategy, although the share of the competitive strategy was higher.This strategy showed that strength factors were in a good position, but in external factors, the weight of threats was more significant than the opportunities, which showed the negative effect of threat factors (Figure 3).According to Table 10, the highest correlation coefficient was related to GTA and SAW.However, these techniques showed a very low correlation with VIKOR.Based on the overall weights presented in Tables 1 and 2, the positive factors gained more weight than the negative ones.Based on this, it can be concluded that Cheshmeh-Kileh Watershed was in good condition based on stakeholders' opinions, and the techniques of GTA and SAW were selected as the best techniques.The Condorcet algorithm is among the most often used GTA (Avand et al., 2021;Khiavi et al., 2022).
According to Madani (2010), GTAs best reflect the behavior of actors and stakeholders involved in multiobjective decision-making.The Condorcet algorithm accounts for the majority opinion to determine the degree of importance (Elkind et al., 2011;Mahjouri & Bizhani-Manzar, 2013).The Condorcet algorithm is easy to use, which has enhanced its popularity.Adhami and Sadeghi (2016) and Mahjouri and Bizhani-Manzar (2013) found this technique to be appropriate for studies in which the priorities of the majority of voters are considered.Also, in connection with the technique of SAW, Podvezko (2011) and Podviezko and Podvezko (2014) state that this technique can compensate among variables, and the way of measuring is quite simple and does not require several computer programs or tools.(Kraujalienė, 2019) confirms that the SAW technique integrates the values of variables and weights into a single magnitude and normalized values of the evaluation help visually calculate the differences b Correlation is significant at the 0.05 level (two-tailed).
between the alternatives.The weakest result among MCDM techniques was related to the VIKOR technique (contrary to Fouladgar et al., 2011), which showed the watershed situation in a defensive and hostile strategy.After GTA and SAW techniques, BWM and TOPSIS presented relatively acceptable results.In a study, Ghorbani et al. (2011) also weighted the SWOT factors using the TOPSIS technique and concluded that the TOPSIS technique performed well.
The positive point of this research, in addition to using the new and optimal MCDM techniques, was to use the opinions of different watershed stakeholders with expert, technical, and policy studies, which in previous research related to various issues of natural resource management were referred to as a deep gap (Mousavizadeh et al., 2015;Mutekanga et al., 2013).According to the problems and rankings presented in the present study, to IWM, proper and practical planning to reduce weaknesses and threats and strengthen the positive aspects in the management of the Cheshmeh-Kileh Watershed is essential.Regarding the final validation of the current study with field surveys, it is worth mentioning that all data used and collected in this study were based on field surveys and the opinions of various stakeholders of the watershed.All analyzes related to the strategic planning model (SWOT) were also based on the strengths, weaknesses, opportunities, and threats related to the opinions of the watershed stakeholders.In connection with the validation of the study, it can be said that based on the investigations, there was more than 73% agreement between the results of the best MCDM techniques (GTA, SAW, and BWM) and the opinions of technical experts in the watershed.This is despite the fact that it had the lowest compatibility with the VIKOR model.Also, based on optimal MCDM technique in this study, the watershed was in the growth and development strategy, and it was in good agreement with the reports of the General Directorate of Natural Resources and Watershed Management of Nowshahr-West Mazandaran.One of the most important sources of uncertainty and limitations in the current study was the lack of quantitative information measured on various aspects in the subwatersheds to determine the best prioritization and weighting approach.Also, due to the fact that semistructured interviews were conducted during the outbreak of the COVID-19 disease, it was not possible to hold training workshops and interviews with different stakeholders at the same time, and the interviews were conducted separately.

| CONCLUSION
This study aimed to implement an IWM framework by weighting SWOT strategic planning model with MCDM techniques, including GTA, BWM, VIKOR, TOPSIS, and SAW, using the views of key stakeholders (watershed residents, executive experts, and policymakers) in the Cheshmeh-Kileh Watershed.Therefore, semistructured interviews with key stakeholders in the watershed were used to gather information on strengths, weaknesses, opportunities, and threats.During the interview, stakeholders raised ecological, environmental, economic, and social issues in the Cheshmeh-Kileh Watershed.36 factors were then formulated for the SWOT model.For each factor, importance coefficient (1-9) and score (1-4) were considered, and the local weight of each factor was calculated.New and familiar MCDM techniques were then used to weigh the factors.In GTA, the Condorcet algorithm was used.In BWM, the best and worst criteria were identified.Also, in VIKOR, TOPSIS, and SAW techniques, the factors were first ranked, and then by multiplying the rank of each factor by the local weight, the overall weight was calculated.
Based on the local weights of SWOT factors, strengths, and opportunities scored higher than weaknesses and threats, which showed that GTA and SAW techniques provided the best results, and the watershed status assessment matrix based on these methods in the growth and development strategy was analyzed.After these techniques, the BWM and TOPSIS presented relatively good results, and the watershed status assessment matrix was placed in a competitive strategy.The VIKOR technique, which showed the watershed status in the defensive strategy, had the weakest performance among MCDM techniques for weighting the SWOT model.This study showed that the Cheshmeh-Kileh Watershed has a good potential for development by using strengths and opportunities and should use these optimal factors to reduce weaknesses in the watershed.The present study results can help managers in optimal decisions for planning and optimal management in the Cheshmeh-Kileh Watershed to create ecological balance and increase the welfare of watershed residents.It is suggested that other SPMs and MCDM techniques (Borda scoring and Fallback bargaining algorithms) be used to help managers make optimal decisions with more complete and comprehensive data in future studies.
Weighting and ranking of SWOT factors using BWM, the Cheshmeh-Kileh Watershed, Iran.
T A B L E 1 Quantification of local weight of internal SWOT factors for the Cheshmeh-Kileh Watershed, Iran.
Quantification of local weight of external SWOT factors, the Cheshmeh-Kileh Watershed, Iran.
T A B L E 2 Weighting and ranking of SWOT factors using VIKOR, the Cheshmeh-Kileh Watershed, Iran.Weighting and ranking of SWOT factors using TOPSIS, the Cheshmeh-Kileh Watershed, Iran.Weighting and ranking of SWOT factors using SAW, the Cheshmeh-Kileh Watershed, Iran.Quantification of local and overall weights using MCDM, the Cheshmeh-Kileh Watershed, Iran.
T A B L E 10 Results of correlation test of MCDM based on SWOT overall weights, the Cheshmeh-Kileh Watershed, Iran.
a Correlation is significant at the 0.01 level (two-tailed).