Local Action Groups and Rural Sustainable Development. A spatial multiple criteria approach for efficient territorial planning

Land Use Policy 59 (2016) 12–26

Contents lists available at ScienceDirect

Land Use Policy journal homepage: www.elsevier.com/locate/landusepol

Local Action Groups and Rural Sustainable Development. A spatial multiple criteria approach for efficient territorial planning Giovanni Ottomano Palmisano a,∗ , Kannan Govindan b , Antonio Boggia c , Rosa Viviana Loisi a , Annalisa De Boni a , Rocco Roma a a

Department of Agricultural and Environmental Science – University of Bari Aldo Moro, Via Amendola 165/A, 70126 Bari, Italy Department of Technology and Innovation – University of Southern Denmark, Campusvej 55, 5230 Odense M, Denmark c Department of Agricultural, Food and Environmental Sciences – University of Perugia, Borgo XX Giugno 74, 06121 Perugia, Italy b

a r t i c l e

i n f o

Article history: Received 8 November 2015 Accepted 2 August 2016 Keywords: Rural Sustainable Development Local Action Groups Spatial Decision Support Systems Multiple Criteria Decision Aiding TOPSIS DRSA

a b s t r a c t Rural Sustainable Development is a very important topic under the European Union policy, and it is currently promoted through the European Agricultural Fund for Rural Development 2014–2020. This fund is managed at sub-regional level by the Community-Led Local Development approach that involves Local Action Groups in order to promote the objectives of Rural Sustainable Development within rural municipalities. Each Local Action Group applies the Strengths, Weaknesses, Opportunities and Threats analysis in order to identify for its own rural municipalities the strategic elements to which it will allocate the European Agricultural Fund for Rural Development budget. Nevertheless, this analysis has some general shortcomings, including difficulties in managing a large number of Strength and Weakness factors. In addition, the importance of each factor cannot be measured quantitatively, and the same factor may be characterized both as a Strength and a Weakness. Further difficulties may occur in the case of partnerships between different Local Action Groups, such as disagreement about whether a given factor is a Strength or a Weakness, lack of information about the relationships between Strength and a Weakness factors and decision alternatives, as well as impossibility of ranking the decision alternatives. Thus, this research aims to overcome the drawbacks of the Strengths, Weaknesses, Opportunities and Threats analysis and to support Local Action Group partnerships in the sustainability evaluation of their rural municipalities, and therefore to aid the identification of a common Rural Sustainable Development strategy to allocate the European Agricultural Fund for Rural Development budget. This decision problem was tackled by applying a Multiple Criteria Spatial Decision Support System that integrates a Geographic Information System with the Multiple Criteria Decision Aiding methods “Technique for Order Preference by Similarity to Ideal Solution” and “Dominance-based Rough Set Approach”. In order to demonstrate the validity of this methodological approach, this Multiple Criteria Spatial Decision Support System was applied to a study area of thirteen rural municipalities located in Apulia Region (Southern Italy); these municipalities belong to the same landscape unit, but they are managed by five different policy makers that represent the Local Action Groups. The results provided the maps of environmental, economic and social sustainability rankings of rural municipalities as well as their overall sustainability value. Based on these rankings, a specific Rural Sustainable Development strategy was identified for the allocation of the European Agricultural Fund for Rural Development. This methodology provided a common decision making framework that can also be applied to Local Action Group partnerships within the European Union. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction

∗ Corresponding author. E-mail addresses: [email protected] (G. Ottomano Palmisano), [email protected] (K. Govindan), [email protected] (A. Boggia), [email protected] (R.V. Loisi), [email protected] (A. De Boni), [email protected] (R. Roma). http://dx.doi.org/10.1016/j.landusepol.2016.08.002 0264-8377/© 2016 Elsevier Ltd. All rights reserved.

Sustainable Development (SD) was described for the first time by the Brundtland Commission in 1987 as “development that meets the needs of the present without compromising the ability of future generations to meet their own needs” (World Commission on Environment and Development, 1987). Although the Brundtland

G. Ottomano Palmisano et al. / Land Use Policy 59 (2016) 12–26

Commission illustrated SD as a model based on environmental and developmental dimensions, this is currently defined as a Triple Bottom Line (TBL) model, since SD includes and integrates environmental, economic and social dimensions (Pope et al., 2004). In recent years, SD has also become a central topic in rural areas (Paˇsakarnis et al., 2013) due to complex interactions between natural resources, agricultural production and local communities (Mennella, 2006), and therefore policy makers have started to plan and enhance these aspects within the European Union’s (EU) rural development policies. In particular, the EU Rural Development Policy 2014–2020 helps EU rural areas to meet and face the wide range of environmental, economic and social challenges and opportunities of the 21st century. This policy is also known as the second pillar of the Common Agricultural Policy (CAP) (European Commission, 2013). This second Pillar is implemented in each EU Member State through the national and regional Rural Development Programmes (RDPs) (European Commission, 2010), which are supported by the following European Structural and Investment (ESI) funds (European Parliament, 2013): the European Agricultural Fund for Rural Development (EAFRD), the European Regional Development Fund (ERDF), the European Social Fund (ESF), the Cohesion Fund (CF) and the European Maritime and Fisheries Fund (EMFF). Specifically, the EAFRD contributes to improving Rural Sustainable Development (RSD) by taking into account the competitiveness of agriculture and forestry, the natural environment and the countryside, as well as the quality of life and the management of economic activities in rural areas (European Commission, 2005). Thus, the EAFRD provides an integrated management strategy that is related to the following five objectives of RSD (Baldock et al., 2001): 1) Diversification of traditional farming activities (Sharpley and Vaas, 2006; Di Domenico and Miller, 2012); 2) Multifunctionality of agriculture regarding landscape use and enhancement of environmental, historical and cultural heritage (Van Huylenbroeck et al., 2007; Marsden and Sonnino, 2008); 3) Improvement of food security and promotion of local food products (Ayres and McCalla, 1996; De Noronha Vaz et al., 2009); 4) Local community involvement in conservation of social and cultural traditions (MacKinnon, 2002; Daskon, 2010); 5) Employment and income generation in agriculture (Bhakar et al., 2007). The EAFRD is managed at the sub-regional level through the Community-Led Local Development (CLLD) approach, a model that involves the local actors in order to promote the RSD within rural municipalities (Soto and Ramsden, 2014). These local actors consist of public and private socio-economic bodies, and are called Local Action Groups (LAGs) (Lukesch, 2007). Each LAG decides how its EAFRD budget will be used for RSD strategies according to the territorial features of their own rural municipalities. Partnerships among different LAGs are also promoted in order to perform a common RSD strategy within rural municipalities with similar territorial features. The allocation of the EAFRD budget in each LAG is carried out by setting up the Local Development Strategy (LDS) (European Network for Rural Development, 2013) that identifies specific RSD needs and potentials via application of the Strengths, Weaknesses, Opportunities and Threats (SWOT) analysis (Helms and Nixon, 2010). SWOT analysis is an effective strategic development tool that is used in the preliminary stages of decision-making and as a precursor to strategic planning (Srivastava et al., 2005). Specifically, the SWOT framework consists of internal and external assessments. The internal assessment is performed to describe Strength and

13

Weakness factors of a given context (such as a company, territory, etc.), while the external assessment is applied to discover Opportunities and Threats (Matthews, 2004). This framework makes it possible to identify the best strategy that maximizes the Strengths and Opportunities and minimizes the Weaknesses and Threats (Hill and Westbrook, 1997). SWOT analysis is mostly applied in the fields of health care, industry, renewable energy and waste management (Zhang and Chen, 2013; Aslan et al., 2014; Chen et al., 2014; Neagu et al., 2015), but can be used also in rural areas. In this specific context, SWOT analysis is useful as a first approximation when the analysis impetus does not arise from a well-identified problem but from a desire to develop a rural area and to optimize the use of its available resources (Tapiador, 2008). Nevertheless, SWOT analysis has the following general drawbacks: difficulties in managing a large number of Strength and Weakness factors (Coyle, 2004), the impossibility of quantitatively measuring the importance of each Strength and Weakness factor (Schmoldt and Peterson, 2000), and the possibility of characterizing the same factor both as a Strength or as a Weakness (Henry, 2008). Further difficulties may occur when SWOT analysis is applied in larger contexts, like partnerships between several LAGs. These drawbacks include disagreement about whether a given factor is a Strength or a Weakness (Henry, 2008), lack of information about the relationships between Strength and Weakness factors and the decision alternatives (i.e. rural municipalities) (Kurttilaa et al., 2000), as well as the impossibility of ranking the decision alternatives (Pahl and Richter, 2007). This ranking should be performed for every sustainability dimension in order to provide policy makers with a complete overview of the decision problem that can guide them towards global sustainability (Hacking and Guthrie, 2008; Bond and Morrison-Saunders, 2011) and help them to choose the proper sustainable strategies (Devuyst, 2001). Thus, this research aims to overcome the above SWOT drawbacks and to support partnerships between several LAGs in sustainability evaluation of their rural municipalities, and therefore in identifying a common RSD strategy for allocation of the EAFRD budget. Spatial decision problems in agriculture require a large number of alternatives to be evaluated based on multiple criteria (Silva et al., 2014), therefore a possible solution comes from the integration of a Geographic Information System (GIS) and a Multiple Criteria Decision Aiding (MCDA) technique (Malczewski, 2006, 2010); this integration is known as Multiple Criteria Spatial Decision Support Systems (MC-SDSS) (Sugumaran and DeGroote, 2011). MC-SDSS have been used in many research fields over the last twenty years, as pointed out by Malczewski (2006), such as environment and ecology, hydrology, agriculture and forestry, geology, transportation, waste management. Within these research fields we underline the works of Dragan et al. (2003), Gilliams et al. (2005), Rahman et al. (2012), De Luca et al. (2012), Vaskan et al. (2013), Comino et al. (2014), Wanderer and Herle (2015). MC-SDSS link concepts and methods of GIS and MCDA, providing new ways to face decision problems (Malczewski and Rinner, 2015), because a MC-SDSS is a decision support tool that makes it possible to combine geographic data and policy makers’ preferences, so that specific information for a decision is presented (Greene et al., 2010; Bottero et al., 2013). In particular, the application of a MC-SDSS to deal with territorial planning decision problems may provide the following benefits: use of appropriate analytical tools for direct involvement of people in a collaborative spatial planning process (Jelokhani-Niaraki and Malczewski, 2015a); possibility of structuring and evaluating the decision problem according to a variety of evaluation criteria that are prioritized quantitatively according to a specific decision rule (Massei et al., 2014); ranking a set of alternatives according to their relative importance

14

G. Ottomano Palmisano et al. / Land Use Policy 59 (2016) 12–26

Fig. 1. The map of the study area.

in satisfying the analysis objective (Demesouka et al., 2013); GIS maps help parties to reach a consensus in certain types of conflicts (Ozawa, 1999) by using a visual language that is easily explained and understood (Haertsch and Smith, 2012; Jelokhani-Niaraki and Malczewski, 2015b); identification of the RSD potentialities of rural municipalities (Boggia et al., 2014). In this research, a MC-SDSS was applied to a study area of thirteen rural municipalities in Apulia Region (Southern Italy), which are managed by five different policy makers that represent the LAGs. Specifically, we applied a MC-SDSS that integrates a GIS with the MCDA methods “Technique for Order Preference by Similarity to Ideal Solution” (TOPSIS) (Hwang and Yoon, 1981) and “Dominance-based Rough Set Approach” (DRSA) (Greco et al., 2001). The paper is structured as follows: after the description of the case study in Section 2, the GIS-MCDA integration and the chosen MC-SDSS are described in Section 3. The results and discussion are reported in Section 4, while the conclusions are stated in Section 5.

1) LAG “Murgia più”: including Gravina in Puglia, Minervino Murge, Poggiorsini, Ruvo di Puglia and Spinazzola; 2) LAG “Conca barese”: including Cassano delle Murge, Grumo Appula and Toritto; 3) LAG “Terre di Murgia”: including Altamura and Santeramo in Colle; 4) LAG “Le città di Castel del Monte”: including Andria and Corato; 5) LAG “Fiori d’olivi”: including Bitonto. All the above municipalities are considered as rural municipalities because they fit the following principles of rurality (Kaiser, 1990; Boggia et al., 2014):

2. Case study

- Low density of population and buildings, with a prevalence of landscape features; - Prevalence of agricultural areas, woodlands and pastures; - Inhabitants life takes place in small urban centers and they have a very close relationship with the surrounding environment; - Specific identity and self-representation of the people influenced by a rural background.

The study area is in Apulia Region, Southern Italy, and consists of 13 municipalities in the Provinces of Bari and Barletta-AndriaTrani; it covers a total area of 2,639.7 km2 (Istituto Nazionale di Statistica, ISTAT, 2010) (Fig. 1) and includes 428,720 inhabitants (Istituto Nazionale di Statistica, ISTAT, 2011). Although the entire study area belongs to the “Altopiano Murgiano” landscape unit (Regione Puglia, 2015), it is managed by five LAGs, as follows:

The natural environment is very distinctive, so that the “Alta Murgia National Park” was established in 2004 to include all the municipalities listed above (Gazzetta Ufficiale Serie Generale, 2004). This Park has a total area of 680 km2 and belongs to the Natura 2000 “Murgia Alta” site (SCI/SPA IT9120007) (European Commission, 1992; European Commission, 2009). Specifically, the natural environment consists of complex geological and landscape features formed over the centuries, such as rocky

G. Ottomano Palmisano et al. / Land Use Policy 59 (2016) 12–26

ridges, sinkholes, hills, karst caves and ravines. Furthermore, the natural landscape is very diversified, as the woodlands consist of many tree varieties (Quercus pubescens L., Quercus ilex L., Quercus cerris L., Quercus coccifera L., Quercus calliprinos Webb, Quercus frainetto Ten., Quercus trojana Webb, Pinus halepensis Mill., Cupressus sempervirens L.) but there are also Mediterranean scrub, grasslands and pastures (Olea europea L., Amygdalus communis L., Palirus spinachristi Mill., Mespilus germanica L., Prunus spinosa L., Pyrus amygdaliformis, Prunus webbi Spach, Crataegus monogyna Jacq., Rhamnus saxatilis Jacq.). Furthermore, biodiversity is promoted at the farm level with linear landscape elements, such as dry-stone walls, hedgerows and bushes. Continuous human management of the territory, has created a rich and diverse archaeological and architectural heritage, but also many abandoned or decommissioned manufacturing sites. The archaeological heritage includes specific types of cave settlements, and fossils found in these confirm use of these caves during the Paleolithic age. In addition, there are also burial grounds dating from the seventh to fourth century BC. The architectural heritage consists of castles and churches, but the most important buildings are those related to agriculture and grazing. Specifically, these manufactures can be summarized as follows: “poste”, buildings enclosed by dry-stone walls and used by shepherds to protect herds during bad weather; “jazzi”, typical enclosed pens for cattle, usually located on steep slopes; “masserie”, fortified farmhouses that became organizational centers of agricultural life from the fifteenth century; “neviere”, used as snow stores; “specchie”, stone heaps surrounded by dry-stone walls; boreholes for obtaining groundwater and cisterns for rainwater storage; “tratturi”, long unpaved roads used by shepherds during livestock transhumance (Acciani et al., 2010). Finally, the abandoned or decommissioned manufacturing sites consist of quarries that cover an area of about 1,106 ha (Regione Puglia, 2007). Agriculture is the main economic activity, and the Utilised Agricultural Area (UAA) is approximately 198,490 ha, equivalent to 70% of the entire territory (Istituto Nazionale di Statistica, ISTAT, 2010). There is a strong tradition of agricultural production and food quality related to wheat, extra-virgin olive oil, wine and milk as well as to the rearing of sheep, goats and cattle. In particular, the following agricultural products obtained the Protected Designation of Origin (PDO) (European Commission, 2006; European Union, 2012): “Olio extravergine Terra di Bari” (extra-virgin oil); “Castel del Monte Bombino Nero”, “Castel del Monte Nero di Troia riserva” and “Castel del Monte rosso riserva” (red wine); “Castel del Monte” (white, rosé and red wines); “Gravina” (white wine); “Pane di Altamura” (durum wheat bread), and “Canestrato pugliese” (hard cheese from sheep’s milk). In addition, “Murgia” white, rosé and red wines hold the Protected Geographical Indication (PGI) (European Union, 2007). The tourism sector is growing thanks to the above-mentioned territorial resources. In particular, tourist arrivals were 66,858 in 2013 (Osservatorio del Turismo della Regione Puglia, 2013). Nevertheless, there are several issues that need to be managed, and these include: economic problems related to agriculture, such as the limited development of specific activities (farmhouses, organic farms, sale of products by e-commerce, etc.); environmental concerns about the use of natural resources (consumption of groundwater for irrigation, establishment of renewable energy plants); social aspects related to farms and inhabitants (waste production, use of external workers only for specific farming operations, farm management by entrepreneurs over 60 years old, and consumption of agricultural products from farmer’s own resources).

15

Fig. 2. The MC-SDSS classification according to the GIS-MCDA integrations (Chakhar and Mousseau, 2008).

3. Method 3.1. GIS-MCDA integration Geographic Information Systems (GIS) are considered tools to manage spatial decision problems, but some of these systems do not implement the appropriate add-on functionalities that help the policy makers in facing these problems. On the other hand, MCDA methods provide many techniques that can reveal the policy maker’s preferences and model them into GIS based decision making (Laskar, 2003). Therefore, the integration between GIS and MCDA can improve the Spatial Decision Support Systems (SDSS) (Ascough et al., 2002) by creating a common decision making framework (Malczewski, 1999). Basically, a MC-SDSS consists of three components: a geographical database and its related management system, a MCDA model-based management system and a user interface (Massei et al., 2014). In particular, some authors (Chakhar and Martel, 2003; Laskar, 2003) have classified MC-SDSS into three categories according to the GIS-MCDA integration, so they are weakly or tightly coupled and fully integrated. Fig. 2 shows the classification provided by Chakhar and Mousseau (2008). In weak coupling (also defined as loose or indirect integration), GIS and MCDA are totally separated and use their own database and graphical interfaces; in addition, they are linked by an intermediate system that is managed by the analyst. The advantage of weak coupling is the low development cost and that the analyst is

16

G. Ottomano Palmisano et al. / Land Use Policy 59 (2016) 12–26

not obliged to use fixed MCDA methods, but at the same time the separation of the models makes it difficult to understand the spatial dimension of the decision problem and analytical errors may occur during data transfer (Lidouh, 2013). In tight coupling (also called tight integration or built-in integration), GIS and MCDA share a unique graphical interface, but the MCDA method is integrated as a separated module or script and therefore has its own database. Nevertheless, this type of integration is more efficient than weak coupling, because the data transfers and analysis are performed within GIS that works as the main software. Finally, in full integration (also known as complete integration), GIS and MCDA share the same database and graphical interface. The MCDA method can be activated in GIS software like any other analysis function (Chakhar and Martel, 2003). Furthermore, the analyst can access both MCDA and GIS tools during every step of the decision making process; input data can be changed until the research goal is achieved (Massei et al., 2014) and partial results can be viewed in the form of maps. The only disadvantage is about the mandatory use of fixed MCDA methods, nevertheless full integrated MC-SDSS should be preferred to weakly and tightly integrated MCSDSS. Fig. 3 shows the general flowchart of a fully integrated MCSDSS proposed and described by Lidouh in 2013. It consists of three phases that are summarized as follows: 1) Intelligence phase: the analyst performs data collection on the decision problem to identify the most suitable MCDA method to apply and stores the acquired information in the GIS. Specifically, in problem definition, the research goal is identified and a specific decision problem is chosen according to the same goal (for example choice, ranking, etc.). Then, the GIS is used to identify the spatial evaluation criteria and any constraints that make it possible to delineate the spatial alternatives; 2) Design phase: after completing data collection, the analyst establishes the decision problem model, which will be based on the chosen MCDA method. Specifically, once the constraints have been entered, the analyst can create a list of alternatives or identify the set of feasible solutions; after criteria and alternatives have been defined, the analyst can enter the parameters

Fig. 3. The flowchart of the fully integrated MC-SDSS (Lidouh, 2013).

(i.e. weights) within the evaluation matrix according to the policy maker’s point of view. Finally, specific decision rules can be set in relation to the applied MCDA method; 3) Choice phase: the analyst and the policy maker use MC-SDSS to explore feasible solutions. In particular, sensitivity analysis (if implemented) is applied to verify the results robustness or to modify the MC-SDSS in relation to the policy maker’s needs. At the end of the process, the analyst has to provide the policy maker with final recommendations, for example explanation of the problem and/or suggestions about how to implement the decision process.

3.2. GeoUmbriaSUIT MC-SDSS In accordance with the aim of the present research, several MC-SDSS were investigated; of these, “GeoUmbriaSUIT” (Boggia and Cortina, 2008; Massei and Boggia, 2014) was chosen because it is specifically designed to assess rural sustainability and RSD potentialities according to environmental, economic and social dimensions, which are measured through a set of criteria (i.e. indicators) based on the territorial features of the study area. This is a fully integrated MC-SDSS that combines a GIS with the MCDA methods TOPSIS and DRSA. TOPSIS is based on the concept that the alternative to choose is the one at the minimum distance from an ideal best alternative and at the maximum distance from an ideal worst alternative. Therefore, the alternatives are ranked according to their distance from the worst point and their closeness to the ideal point for every set of indicators. TOPSIS is applied separately to each sustainability dimension, so that it provides three sustainability indexes; their linear combination leads to the overall sustainability index for every alternative. Furthermore, the relationship between the performance of indicators and the sustainability ranking of alternatives can be investigated by applying DOMLEM algorithm (Greco et al., 2000) that is based on DRSA. This makes it possible to extract certain decisional rules by identifying the indicators that mostly influence the sustainability ranking. DRSA works starting from exemplary cases from which to extract rules to apply to alternatives to obtain a ranking. In this case, the exemplary cases are the best alternatives already found after TOPSIS application. DRSA is not used here to obtain ranking, but to extract the decisional rules that can explain the positions obtained by the alternatives in the ranking, based on the indicators used. Using DRSA, traceability and back analysis ability are improved. Traceability means that from an alternative’s score it is possible to return to the rules, and from the rules back to the input data. It allows the user to analyse each single step that leads to the final result, revealing which indicators or procedural steps have the greatest impact on the results. Therefore, the EAFRD budget will be allocated according to a specific RSD strategy related to these indicators. “GeoUmbriaSUIT” works within the open source Geographic Information System (GIS) software “QGIS” (QGIS Development Team, 2013), and uses a geographic vector file (i.e. a shapefile), where graphic data represent single alternatives (for example countries, regions or municipalities), while the alphanumeric data (the attribute table) describe the environmental, economic and social indicators of each alternative. The inputs required by TOPSIS are a decision matrix (where rows show the alternatives, while the criteria are put into columns) and a vector of weights that includes information about the policy maker’s preferences. In particular, “GeoUmbriaSUIT” includes the following steps:

G. Ottomano Palmisano et al. / Land Use Policy 59 (2016) 12–26

17

Table 1 Description of the fundamental scale of values (Saaty and Vargas, 2012). Intensity of importance

Definition

Explanation

1

Equal importance

Two activities contribute equally to the objective

2 3

Weak Moderate importance

4 5

Moderate plus Strong importance

6 7

Strong plus Very strong or demonstrated importance

8 9

Very very strong Extreme importance

Reciprocals of above

If activity i has one of the above non-zero numbers assigned to it when compared with activity j, then j has the reciprocal value when compared with i Ratios arising from the scale

Rationals

1) Construction of the normalized decision matrix whose elements are defined as follows: zij =

yij

 n

y2 i=1 ij

Where: i = alternatives j = criteria z ij = normalized decision matrix yij = decision matrix With: i = 1, . . ., n j = 1, . . ., k Since the present problem affects rural municipalities belonging to five LAGs, the official representatives of each LAG were involved in the decision process. They were invited to participate in meetings in order to choose a common set of sustainability indicators to use as criteria in the above normalized decision matrix. A detailed explanation about the indicators is reported in Section 3.3. 2) Definition of the weighted normalized decision matrix, whose elements are obtained from: xij = wj zij Where: xij = weighted normalized decision matrix

Experience and judgement slightly favour one activity over another Experience and judgement strongly favour one activity over another An activity is favoured very strongly over another; its dominance demonstrated in practice The evidence favouring one activity over another is of the highest possible order or affirmation A reasonable assumption

If consistency were to be forced by obtaining n numerical values to span the matrix

j = 1, . . ., k The weighting of indicators can be done either through direct entry of numerical values or using the weights (i.e. priorities) computed through the pairwise comparison matrices implemented in the MCDA method “Analytic Hierarchy Process” (AHP) (Saaty, 1980). The use of weights derived from the AHP in the weighted normalized decision matrix of TOPSIS has been applied in various pieces of research (Goh et al., 2013; Zaidan et al., 2015; Hanine et al., 2016). This was done in order to improve the reliability of the results by identifying potential contradictions in the decision maker’s judgements, and thus to modify them in time (Yu et al., 2013). The meaning of criteria weights in the AHP is that provided by Choo et al. (1999), who defined these weights as the relative contribution of the total or average score of all the alternatives. Considering the difficulty of reaching a rapid agreement among the LAG official representatives about the direct entry of specific values and the need to obtain more reliable results, the pairwise comparison matrices were filled in. Specifically, a questionnaire including three matrices (one for each dimension of sustainability) was distributed to all LAG official representatives, and they assigned a value judgement to every pairwise comparison according to the fundamental scale of values (Saaty and Vargas, 2012, Table 1). In particular, these value judgements were assigned after brainstorming in the management board of each LAG. Thus, the value judgements were aggregated by applying the geometric mean (Saaty and Peniwaty, 2007; Saaty, 2012) and were inserted in the pairwise comparison matrices provided by “GeoUmbriaSUIT”. Then the priorities were calculated applying the eigen value method (Ishizaka and Labib, 2009; Bana e Costa and Vansnick, 2007; Ishizaka and Lusti, 2006) based on the following equation: 



Ap = np Where:

wj = weight of the j-th criterion

A = comparisonmatrix

z ij = normalized decision matrix



p = vector of the priorities

With: i = 1, . . ., n

n = dimension of the matrix

18

G. Ottomano Palmisano et al. / Land Use Policy 59 (2016) 12–26

Table 2 The Random Indices for the computation of the Consistency Ratio (Saaty, 1977).

Where:

n.

3

4

5

6

7

8

9

10

Ci∗ = relative closeness to the ideal point

RI

0.58

0.9

1.12

1.24

1.32

1.41

1.45

1.49

With: i = 1, . . ., n

Furthermore, a consistency check was performed to identify potential contradictions in the entries. The consistency check follows the consistency index (CI) (Saaty, 1977) related to the eigen value method:

Where:

6) Ranking of the alternatives: these are ranked from the best to the worst according to C ∗i . For example, if Ci∗ > Cj∗ , then alternative ai is better than alternative aj . 7) Extraction of the decisional rules: after analysing the results obtained by TOPSIS, the relationship between the performance of indicators and the alternatives’ sustainability ranking was further investigated by applying DRSA.

max = maximal eigen value

3.3. Indicators

CI =

max − n n−1

The consistency ratio (CR) is: CR =

CI RI

Where: RI = random index (average CI of 500 randomly filled matrices) A matrix is consistent when CR is less than 0.1 of 500 randomly filled matrices. RI is computed according to Table 2 (Saaty, 1977). 3) Definition of the ideal point (a∗ ) and the worst point (a− ) as follows: a∗ = {(maxi xij |j ∈ J), (mini xij |j ∈ ˆJ )|i= 1, . . .,n} = a- = {(mini xij |j ∈ J), (maxi xij |j ∈ ˆJ )|i= 1, . . .,n} =



x1∗ , x2∗ , . . ., xk∗



x1 , x2 , . . ., xk





Where: J = set of criteria to be maximized (gains) ˆJ = set of criteria to be minimized (costs) During this step, all the LAG official representatives evaluated together every sustainability indicator as a “Gain” or “Cost”, depending on whether the increase of value leads to an improvement or a worsening of sustainability. In this case, agreement was reached by a consensus vote after some discussion. 4) Calculation of the distance of each alternative from the ideal point (Si∗ ) and the worst point (Si ) as follows:



Si∗

=

k  j=1

 Si =

k  j=1

xij − xj∗

xij − xj

2

2

Where: Si∗ = distance from the ideal point Si = distance from the worst point With: i = 1, . . ., n 5) Calculation of the relative closeness of each alternative to the ideal point: Ci∗ =

Si

Si + Si∗

A set of chosen indicators is the most suitable tool to describe a multidimensional concept like Sustainable Development, and it represents the core of the sustainability assessment process (Moffat et al., 2001; Ness et al., 2007). Indeed, a lot of works have dealt with the sustainability assessment by the implementation and use of indicators (Hezri and Dovers, 2006; Rosenström and Kyllonen, 2007; Cassar et al., 2013; Morrison-Saunders et al., 2014; Bolcárová and Koloˇsta, 2015). The most widely accepted definition of indicator is that provided by the Organization for Economic Co-operation and Development (OECD) (Organization for Economic Co-operation and Development, 2003), according to which the indicator is a parameter associated with an environmental phenomenon, which can provide information on the characteristics of the event in its global form. Therefore, the use of indicators provides targeted details that make it possible to specifically represent the problem under study by maintaining intact the analysis informative content (Schmidt Di Friedberg, 1986). A good indicator should have the following characteristics (Organization for Economic Co-operation and Development, 2003; Cassatella and Peano, 2011): 1) Representativeness: the indicator must be clearly related to a specific phenomenon or feature to measure or control and it should be valid enough in similar contexts; 2) Accessibility: it must be easy to measure and, if possible, monitored automatically through standard and accessible techniques; 3) Collectability: the data for its calculation must be readily accessible both to public and private bodies; 4) Good cost-effectiveness: there must be no excessive consumption of economic resources for the data collection in relation to the final information that the indicator will contain; 5) Communicability: it must be immediately understandable to a technical and non-technical audience, simple to interpret and represent using different tools like tables, charts or maps; 6) Updateability: new values of the same series must be available so that the indicator can be updated. According to the above requirements and the territorial features of the study area, the LAGs’ official representatives chose a common set of six RSD indicators for each dimension of sustainability. Some indicators were related to the total and expressed as percentages in order to enable comparability between larger and smaller municipalities. When an indicator is defined as a cost, it means that the higher its value, the worse the result in terms of sustainability. On the contrary, if the indicator is a gain, it means that the higher its value, the better the result. The following is the complete set of environmental sustainability indicators:

G. Ottomano Palmisano et al. / Land Use Policy 59 (2016) 12–26

- A1: Linear landscape elements (GAIN): farms with maintenance or creation of linear landscape elements as a percentage of total farms; - A2: Quarries (COST): area (ha) of inactive and abandoned quarries; - A3: Woodlands (GAIN): woodland area (ha) as a percentage of the total municipal area; - A4: Waste (COST): production of waste (kg) per inhabitant; - A5: Groundwater (COST): UAA (ha) irrigated with groundwater as a percentage of total UAA; - A6: Renewable energies (GAIN): farms with renewable energy plants as a percentage of total farms. The set of economic sustainability indicators is as follows: - E1: Farmhouses (GAIN): number of farmhouses for agri-tourism; - E2: Organic farms (GAIN): organic farms as a percentage of total farms; - E3: E-commerce (COST): farms not selling products by ecommerce as a percentage of total farms; - E4: PDO crops (GAIN): UAA (ha) for PDO crops as a percentage of total UAA; - E5: Inactive farms (COST): non-working farms as a percentage of total farms; - E6: Certified organic livestocks (GAIN): certified organic livestock as a percentage of total livestock population. Finally, the following set of indicators is related to social sustainability: - S1: Agricultural entrepreneurs (COST): agricultural entrepreneurs older than 60 years as a percentage of total agricultural entrepreneurs; - S2: Cultural heritage (GAIN): number of archaeological and architectural heritage sites; - S3: Educational farms (GAIN): number of farms for didactic activities; - S4: Self-consumption (COST): farms without self-consumption of agricultural products as a percentage of total farms; - S5: Agricultural workers (GAIN): agricultural workforce as a percentage of total workforce; - S6: Tourism (GAIN): number of annual tourist arrivals. The data collection was carried out by referring to the following documents: the Regional Landscape and Territorial Plan (Piano Paesaggistico Territoriale Regionale, Regione Puglia, 2015), the Regional Hydrogeomorphological Chart (Carta Idrogeomorfologica, Regione Puglia, 2007), the Land Use Chart (Carta di Uso del Suolo, Comitato Interministeriale per la Programmazione Economica, 2003a, 2003b), the Regional Annual Report for Tourism Movement (Report del movimento turistico annuale per Comune, Osservatorio del Turismo della Regione Puglia, 2013), the 6th General Census on Agriculture (Istituto Nazionale di Statistica, ISTAT, 2010) and the 15th General Census of Population and Housing (Istituto Nazionale di Statistica, ISTAT, 2011). 4. Results and discussion Data collection made it possible to complete the set of RSD indicators for each rural municipality. The indicators were collected into three separate tables according to the sustainability dimensions (Tables 3–5). The priorities for each set of sustainability indicators are described as follows. In particular, the environmental sustainability indicators (Table 6) Woodlands (A3), Waste (A4) and Groundwa-

19

Table 3 The values of environmental sustainability indicators for each municipality. MUNICIPALITY

INDICATORS

1. Altamura 2. Andria 3. Bitonto 4. Cassano delle Murge 5. Corato 6. Gravina in Puglia 7. Grumo Appula 8. Minervino Murge 9. Poggiorsini 10. Ruvo di Puglia 11. Santeramo in Colle 12. Spinazzola 13. Toritto

A1

A2

4.27 6.93 7.12 15.93 6.60 11.30 20.83 4.46 1.62 6.60 18.72 13.42 21.40

114.08 71.1 98.66 12.98 33.52 136.93 14.94 463.98 0 133.47 6.27 20.37 0.09

A3 2.95 1.89 2.85 15.71 1.80 8.13 2.92 6.36 0 4.49 4.72 4.01 3.81

A4

A5

A6

17.67 69.65 20.27 13.04 31.29 6.24 51.54 11.04 37.94 20.77 16.60 5.86 32.26

0.40 34.61 13.13 4.30 20.05 0.62 12.09 8.59 0.07 6.78 0.38 4.08 4.47

0.24 0.10 0.21 0.78 0.21 0.31 0.07 0.52 0 0.18 1.49 0.92 0.15

Table 4 The values of economic sustainability indicators for each municipality. MUNICIPALITY

INDICATORS

1. Altamura 2. Andria 3. Bitonto 4. Cassano delle Murge 5. Corato 6. Gravina in Puglia 7. Grumo Appula 8. Minervino Murge 9. Poggiorsini 10. Ruvo di Puglia 11. Santeramo in Colle 12. Spinazzola 13. Toritto

E1

E2

E3

E4

E5

E6

9 6 0 9 5 5 1 1 1 4 2 2 2

6.69 1.96 2.12 8.18 1.35 3.75 6.41 5.04 5.83 3.23 6.94 9.47 10.74

99.79 99.71 99.85 99.65 99.84 99.69 99.87 99.74 99.35 99.91 99.94 99.87 99.78

0.17 19.13 16.09 2.98 6.04 0.38 9.83 2.07 2.76 6.49 0.72 0.45 4.06

1.55 2.10 1.73 1.48 1.39 1.60 0.80 2.07 0.65 1.15 1.55 1.18 0.74

10.1 3.1 13.6 16.7 0 5.7 0 5.6 0 8.3 8.6 19.4 33.3

Table 5 The values of social sustainability indicators for each municipality. MUNICIPALITY

INDICATORS

1. Altamura 2. Andria 3. Bitonto 4. Cassano delle Murge 5. Corato 6. Gravina in Puglia 7. Grumo Appula 8. Minervino Murge 9. Poggiorsini 10. Ruvo di Puglia 11. Santeramo in Colle 12. Spinazzola 13. Toritto

S1

S2

S3

S4

S5

S6

13.71 13.12 11.41 13.49 13.25 12.90 11.42 11.15 9.39 12.04 13.56 10.46 11.47

298 308 119 89 160 250 66 157 15 166 143 100 53

3 1 0 3 2 3 0 0 0 1 0 2 0

40.06 22.58 7.49 6.96 13.20 31.06 5.47 29.91 52.10 11.48 16.24 41.45 2.87

4.89 10.46 7.64 7.93 9.65 5.39 8.91 15.95 17.15 11.91 8.79 9.61 11.44

8,349 10,739 4,559 962 24,889 8,021 0 0 0 7,920 1,419 0 0

Table 6 The pairwise comparison matrix for priorities’ computation of the environmental sustainability indicators. Indicators

A1

A1 1 3 A2 3 A3 A4 3 3 A5 3 A6 Consistency Ratio (CR)

A2

A3

A4

A5

A6

Priorities

1/3 1 2 2 2 1

1/3 1/2 1 1 1 1

1/3 1/2 1 1 1 1

1/3 1/2 1 1 1 1

1/3 1 1 1 1 1

0.062 0.134 0.207 0.207 0.207 0.185 0.01

ter (A5) received the same priority and also provided the highest relative contribution, as they are preferred 3 times as much as Linear landscape elements (A1) and twice as much as Quarries (A2). Renewable energies (A6) and Quarries (A2) have high and medium

20

G. Ottomano Palmisano et al. / Land Use Policy 59 (2016) 12–26

Table 7 The priorities of the environmental, economic and social sustainability indicators. Sustainability indicators and priorities

Total

Consistency Ratio (CR)

A1 0.062

A2 0.134

A3 0.207

A4 0.207

A5 0.207

A6 0.185

1

0.01

E1 0.146

E2 0.250

E3 0.069

E4 0.226

E5 0.107

E6 0.203

1

0.02

S1 0.125

S2 0.160

S3 0.141

S4 0.100

S5 0.250

S6 0.224

1

0.02

priorities, as Renewable energies is preferred 3 times as much as Linear landscape elements (A1) and received the same preference as A2, A3, A4 and A5; Quarries (A2) is preferred 3 times as much as A1, equally to A6 and 1/2 as much as A3, A4 and A5. Finally, Linear landscape elements (A1) provides the least relative contribution because it is preferred 1/3 as much as the other indicators. Concerning economic sustainability, the priorities of the indicators (Table 7) were obtained using the following pairwise comparisons. Organic farms (E2) received the highest priority because it is equally preferred to PDO crops (E4) and Certified organic livestock (E6), but preferred 3 times as much as Farmhouses (E1) and E-commerce (E3) and preferred twice as much as Inactive farms (E5). So, the second indicator is PDO crops (E4) compared to Organic farms, because it is preferred only twice as much as E1. Certified organic livestock (E6) is the third indicator as it is preferred more than E3 (3 times) and E5 (twice), and equally to the other indicators. Then the LAGs’ official representatives chose Farmhouses (E1), which is preferred twice as much as E3 and E5, equally to E6 and preferred less than E2 (1/3) and E4 (1/2). Inactive farms (E5) provides a low relative contribution as it is preferred only twice as much as E3 and 1/2 as much as the other indicators. The last indicator is E-commerce (E3), because it is preferred 1/2 as much as E1 and E5, and 1/3 as much as the other indicators. Finally, the social sustainability priorities (Table 7) were obtained according to the following pairwise comparisons. Agricultural workers (S5) provides the highest relative contribution because it is preferred twice as much as Agricultural entrepreneurs (S1), Cultural heritage (S2), Educational farms (S3) and Self-

consumption (S4), but is equal to Tourism (S6). Therefore, S6 is the second indicator compared to Agricultural workers (S5), because they have the same preference except when S6 is compared with S2 (equal preference). Cultural heritage (S2) and Educational farms (S3) have high priorities, and differ from each other in the comparison with S6, which is equal to S2 and preferred twice as much as S3. The LAGs’ official representatives chose Agricultural entrepreneurs (S1) and Self-consumption (S4) as indicators with the lowest relative contribution: Agricultural entrepreneurs is equal to S2, S3 and S4 but is preferred 1/2 as much as S5 and S6, while Selfconsumption is also preferred 1/2 as much as S2 and S3. After weighting the indicators, the MC-SDSS elaborated the data set and performed the sustainability ranking of the alternatives by providing cartographic and chart outputs. The cartographic outputs provide a GIS representation of TOPSIS and describe the alternatives ranked according to their sustainability. In particular, the maps show the values of the environmental, economic and social sustainability indexes, and a further map shows the overall sustainability index. The ranking of alternatives is described on the basis of five value classes (very low, low, medium, high, very high), determined by applying the “Equal Interval” algorithm that is implemented in QGIS. Given the highest and lowest values (i.e. the best and the worst municipalities), their range is divided into equal numerical intervals, and therefore the above five value classes are obtained. Furthermore, the values of these five classes are different in each sustainability map, because the alternatives are ranked according to their relative closeness to the ideal point with respect to every set of indicators.

Fig. 4. The map of environmental sustainability ranking.

G. Ottomano Palmisano et al. / Land Use Policy 59 (2016) 12–26

21

Fig. 5. The map of economic sustainability ranking.

The ranking of municipalities according to the environmental sustainability index (EnvIdeal) is shown in Fig. 4. Only two municipalities (Cassano delle Murge and Santeramo in Colle) belong to a very high class, while Gravina in Puglia and Spinazzola are in a high class. In addition, just less than half of the study area (Altamura, Minervino Murge, Poggiorsini, Ruvo di Puglia and Toritto) gets a medium value, and Bitonto, Corato and Grumo Appula belong to the low class. Only Andria is in a very low class of environmental sustainability. Referring to the economic sustainability index (EcoIdeal) (Fig. 5), the ranking is quite different compared to environmental sustainability. In particular, this time Toritto obtains the highest economic sustainability value class, and about half the municipalities are equally distributed between high (Andria, Bitonto and Cassano delle Murge) and medium classes (Altamura, Grumo Appula and

Spinazzola). Finally, Ruvo di Puglia and Santeramo in Colle are in a low sustainability class, while about 30% of the study area (Corato, Gravina in Puglia, Minervino Murge and Poggiorsini) is in the lowest value class. The social sustainability ranking (Fig. 6) provided a less differentiated distribution, since no municipality belongs to the high class. Furthermore, about 60% of the study area (Bitonto, Cassano delle Murge, Grumo Appula, Minervino Murge, Santeramo in Colle, Spinazzola, Toritto and Poggiorsini) is in a very low sustainability class, while Ruvo di Puglia is the only municipality in the low class. Andria, Altamura and Gravina in Puglia are in a medium sustainability class, and Corato is the most socially sustainable municipality. The linear combination of the above indexes provided the overall sustainability index (SustIdeal) shown in Fig. 7. Cassano delle Murge is the most sustainable municipality, and the remaining

Fig. 6. The map of social sustainability ranking.

22

G. Ottomano Palmisano et al. / Land Use Policy 59 (2016) 12–26

Fig. 7. The map of overall sustainability ranking.

study area is equally distributed among the other value classes. Specifically, Altamura, Corato and Toritto are in a high sustainability class, Andria, Gravina in Puglia and Spinazzola are in a medium class, while Bitonto, Ruvo di Puglia and Santeramo in Colle belong to a low class. Finally, the least sustainable municipalities are Grumo Appula, Minervino Murge and Poggiorsini. Unlike the cartographic outputs, the chart outputs do not group the alternatives into classes of values, but illustrate their ranking based on specific values according to each sustainability dimension. “GeoUmbriaSUIT” provides two different interactive graphs. The first is a histogram (Fig. 8), where the horizontal axis shows the alternatives and the vertical axis shows the overall sustainability values. Therefore, the height of each histogram bar is given by the linear combination of environmental, economic and social sustainability indexes. Analysis of this graph showed that the overall sustainability score of the most sustainable municipality (Cassano delle Murge) is equal to 1.58, while the overall sustainability score of the worst sustainable municipality (Poggiorsini) is 0.99. Nevertheless, Cassano delle Murge is also the best in terms of envi-

ronmental sustainability (0.79), but not for economic and social sustainability. Thus, the most sustainable municipality for economic sustainability is Toritto, with a score of 0.57, while the best municipality for social sustainability is Corato (0.74). Although Poggiorsini is the worst municipality in terms of overall sustainability, it is not the worst if the sustainability dimensions are analysed singly. In particular, regarding environmental sustainability, Andria is the worst municipality, with a score of 0.27, while Minervino Murge is the worst in terms of economic sustainability (0.19), and Grumo Appula gets the lowest social sustainability score of 0.21. The second chart (Fig. 9) provided the same information as above, but in a different graphical way. In particular, the horizontal axis shows the environmental values, the vertical axis shows the economic values, and each municipality is represented by a bubble. The colour of bubbles provides social sustainability values according to a chromatic scale from red to green. The municipalities with very low to low social sustainability values have a red chromatic gradation (from the beginning to just before the centre of the bar),

Fig. 8. The histogram of overall sustainability ranking according to each dimension.

G. Ottomano Palmisano et al. / Land Use Policy 59 (2016) 12–26

23

Fig. 9. The bubble chart of overall sustainability ranking according to each dimension.

while the municipalities with medium to very high social sustainability values have a green chromatic gradation (from the centre to the end of the bar). Furthermore, the bubble sizes provide the overall sustainability values, so the larger the bubble, the greater is the municipality’s overall sustainability. After the above results had been discussed with the LAGs’ official representatives, DRSA was applied to identify the indicators that most influence the sustainability ranking, and this made it possible to extract 16 decisional rules that are shown in Table 8. Specifically, four rules regard environmental sustainability, and the indicators are Quarries (A2), Woodlands (A3) and Renewable energies (A6). The first rule means that if the area of inactive and abandoned quarries is ≥ 463.98 ha, then environmental sustainability is at most very low within Minervino Murge. Therefore, it is necessary to carry out recovery and requalification of inactive and abandoned quarries to reduce these areas in this municipality. The second and the third rules are related to woodlands, and establish that if the woodland area as a percentage of the total municipal area is ≤8.13%, then environmental sustainability is at the most high in all municipalities, except in Cassano delle Murge. In addition, if in this municipality the value is ≥15.71%, then it belongs at least to the very high class. Thus, in both cases the LAG policy makers may not perform interventions related to improv-

ing woodland areas (like forestation of agricultural or abandoned land), because sustainability is already high. The last environmental sustainability rule considers the farms with renewable energy plants as a percentage of the total farms. If it is ≤0.07%, then sustainability in Grumo Appula and Poggiorsini is at the most very low; thus it is advisable to strongly encourage the implementation of renewable energy plants through specific projects. Another approach could be related to the implementation of awareness campaigns targeted at farmers to inform them about the importance of these forms of green energy and their environmental impact on crop production. The six decisional rules of economic sustainability are related to Farmhouses (E1), E-commerce (E3), and Inactive farms (E5). In particular, if the number of farmhouses is ≤1.0, economic sustainability is at the most low in Bitonto, Grumo Appula, Minervino Murge and Poggiorsini. Furthermore, if farmhouses are ≥2.0, then sustainability is at least low in the other municipalities, and if the value is ≥9.0 sustainability becomes at least high only in Altamura and Cassano delle Murge. Consequently, EAFRD budget allocation for implementing this form of agricultural business is important to improve sustainability of the municipalities with a number of farmhouses ≤1.0 and ≥2.0, while the LAG policy makers may not intervene in Altamura and Cassano delle Murge because their sustainability level is already high.

Table 8 The decisional rules provided by the application of DRSA. No.

RULE

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

If [(A QUARRIES ≥ 463.98)] then at most class “very low” [8] If [(A WOODLANDS ≤ 8.13)] then at most class “high” [1, 2, 3, 5, 6, 7, 8, 9, 10, 11, 12, 13] If [(A WOODLANDS ≥ 15.71)] then at least class “very high” [4] If [(A RENEWABLE ENERGIES ≤ 0.07)] then at most class “very low” [7, 9] If [(E FARMHOUSES ≤ 1.0)] then at most class “low” [3, 7, 8, 9] If [(E FARMHOUSES ≥ 2.0)] then at least class “low” [1, 2, 4, 5, 6, 10, 11, 12, 13] If [(E FARMHOUSES ≥ 9.0)] then at least class “high” [1, 4] If [(E ECOMMERCE ≥ 99.91)] then at most class “low” [10, 11] If [(E ECOMMERCE ≥ 99.85)] then at most class “medium” [3, 7, 10, 11, 12] If [(E INACTIVE FARMS ≥ 1.6)] then at most class “medium” [2, 3, 6, 8] If [(S EDUCATIONAL FARMS ≥ 2.0)] then at least class “medium” [1, 4, 5, 6, 12] If [(S SELF CONSUMPTION ≥ 52.1)] then at most class “medium” [9] If [(S SELF CONSUMPTION ≤ 2.87)] then at least class “high” [13] If [(S TOURISM ≥ 4,559.0)] then at least class “low” [1, 2, 3, 5, 6, 10] If [(S TOURISM ≥ 10,739.0)] then at least class “medium” [2, 5] If [(S TOURISM ≥ 24,889.0)] then at least class “high” [5]

24

G. Ottomano Palmisano et al. / Land Use Policy 59 (2016) 12–26

The fourth and fifth rules concern the farms without ecommerce sales of their products as a percentage of total farms. If this value is ≥99.91%, sustainability is at most low in Ruvo di Puglia and Santeramo in Colle, while if the percentage is ≥99.85%, then sustainability is at most medium in Bitonto, Grumo Appula, Ruvo di Puglia, Santeramo in Colle and Spinazzola. Therefore, the LAG policy makers should equip local farmers with technological tools for this type of innovative marketing, and also promote training courses to improve farmers’ skills. The last extracted decision rule for economic sustainability regards inactive farms, and it establishes that if the non-working farms as a percentage of the total farms is ≥1.6%, then the sustainability value is at most medium in Andria, Bitonto, Gravina in Puglia and Minervino Murge. Thus, maybe a good choice could be to allocate EAFRD budget for the upturn of farming activity in order to reduce this percentage, but a medium level of sustainability would be guaranteed even without an intervention. Finally, the six decisional rules regarding social sustainability concern Educational farms (S3), Self-consumption (S4) and Tourism (S6). The first rule states that if the number of educational farms is ≥2.0, then sustainability is at least medium in Altamura, Cassano delle Murge, Corato, Gravina in Puglia and Spinazzola. Therefore, an appropriate intervention aimed at increasing the number of farms involved in this activity (including, for example, nutrition and botany lessons for children and students) will certainly improve sustainability. Furthermore, two rules are related to the farms without selfconsumption of agricultural products as a percentage of the total farms. If this value is ≥52.1%, then sustainability is at most medium in Poggiorsini, while if the percentage is ≤2.87%, then sustainability is at least high in Toritto. This means that the increase in farms with self-consumption of agricultural products in Poggiorsini can lead to this municipality’s inclusion in the high and very high classes, while EAFRD allocation to Toritto is not mandatory, because sustainability of this municipality is already high. The last three rules refer to the number of tourist arrivals, and if these are ≥4,559.0, then sustainability in Altamura, Andria, Bitonto, Corato, Gravina in Puglia and Ruvo di Puglia is at least low. Furthermore, if the value is ≥10,739.0, then the class is at least medium at Andria and Corato, and if tourist arrivals are ≥24,889.0, sustainability is at least high at Corato. These rules suggest that the increase in tourist arrivals (usually via specific interventions aimed at differentiating the tourism offer) will improve the sustainability of every municipality.

- The MC-SDSS does not allow more than one policy maker to directly assign the weights to the indicators or fill in the pairwise comparison matrices. This was overcome by distributing a questionnaire to each LAGs official representative, and then their own value judgements were aggregated through the geometric mean; - The MC-SDSS does not allow the indicators to be set as “Gain” or “Cost” by more than one policy maker, thus the LAGs official representatives had to reach an agreement via discussion. On the other hand, the results were easily understood by the LAGs’ official representatives, because the full integration between the GIS and TOPSIS made it possible to represent the sustainability ranking of municipalities with comprehensible maps and by interactive charts. “GeoUmbriaSUIT” does not implement sensitivity analysis, but the application of DRSA is useful to know the relationships between the indicators and the sustainability rankings. In particular, DRSA was fundamental for identifying a suitable RSD strategy for allocation of the EAFRD budget according to the indicators that most influence the sustainability ranking of rural municipalities. It is important to underline that this MC-SDSS provides a support for identifying the strategic elements (i.e. the decisional rules) for EAFRD allocation, but the specific budget allocation will be decided by the LAGs’ official representatives. The LAGs’ official representatives were very interested in tackling this decision problem, so they attended the meetings and shared the final results. Despite the above drawbacks, “GeoUmbriaSUIT” provided a valid decision making framework that can be applied to LAG partnerships in the EU, but the structure of the problem in terms of type and number of indicators should vary according to the spatial features of the study area (Jankowski, 1995). Thus, the application of a MC-SDSS to EAFRD budget allocation may provide a quantitative interpretation of this decision problem by taking into account the sustainability dimensions, and this represents a value-added that can strengthen SWOT analysis outcomes. Future works will study the integration of SWOT analysis with this MC-SDSS, and the obtained results will be used to monitor the allocation of EAFRD 2014–2020 to specific LAG partnership in a mid-term programming period. In addition, this MC-SDSS will be applied to decision problems concerning the allocation of the ESI funds involving the urban and the urban-rural CLLD (Soto and Ramsden, 2014). Acknowledgements

5. Conclusions In this research, a MC-SDSS was applied to help a partnership of five LAGs in Southern Italy to evaluate rural sustainability of their thirteen rural municipalities, and therefore to identify a common RSD strategy for allocation of the EAFRD budget. A common set of 18 indicators was needed to describe the characteristics of the rural municipalities and to rank them according to environmental, economic and social dimensions. The indicators were chosen by the LAGs’ official representatives without particular difficulties, since the entire study area belongs to the same landscape unit. The user-friendly graphical interface of “GeoUmbriaSUIT” provided a support to easily manage the data entry, and the LAGs’ official representatives did not require additional explanations about how the MC-SDSS worked, or about the meaning of the pairwise comparison matrices and the Saaty’s fundamental scale of values. However, the following drawbacks occurred during TOPSIS:

The authors thank Ph.D. Gianluca Massei (University of Perugia, Italy, Department of Agricultural, Food and Environmental Sciences) for his support in data elaboration and interpretation of the obtained results. References Acciani, C., Bianchi, R., Fucilli, V., Panisco, N., 2010. Paesaggio E Territorio Della Murgia Barese: Tipologie Di Manufatti Rurali Mario. Adda Editore, Bari. Ascough II, J.S., Rector, H.D., Hoag, D.L., McMaster, G.S., Vanderberg, B.C., Shaffer, M.J., Weltz, M.A., Ahjua, L.R., 2002. Multicriteria Spatial Decision Support Systems: overview, applications, and future research directions. In: Proceedings of the International Environmental Modelling and Software Society, Grass Development Team, 27 June, 2002, Lugano, Switzerland. The International Environmental Modelling and Software Society, Manno, Switzerland, pp. 175–180. Aslan, I., C¸ınar, O., Özen, Ü., 2014. Developing strategies for the future of healthcare in Turkey by benchmarking and SWOT analysis. Procedia—Soc. Behav. Sci. 150, 230–240. Ayres, W.S., McCalla, A.F., 1996. Rural development, agriculture and food security. Finance Dev. 33, 8–11.

G. Ottomano Palmisano et al. / Land Use Policy 59 (2016) 12–26 Baldock, D., Dwyer, J., Lowe, P., Petersen, J., Ward, N., 2001. The Nature of Rural Development: Towards a Sustainable Integrated Rural Policy in Europe. Synthesis Report. IEEP, London. Bana e Costa, C.A., Vansnick, J.C., 2007. A critical analysis of the eigenvalue method used to derive priorities in AHP. Eur. J. Oper. Res. 187, 1422–1428. Bhakar, R., Banafar, K.N.S., Singh, N.P., Gauraha, A.K., 2007. Income and employment pattern in rural area of chhattisgarh: a micro view. Agric. Econ. Res. Rev. 20, 395–406. Boggia, A., Cortina, C., 2008. Un modello per la valutazione della sostenibilità dello sviluppo a livello territoriale. AESTIMUM 52, 31–52. Boggia, A., Rocchi, L., Paolotti, L., Musotti, F., Greco, S., 2014. Assessing rural sustainable development potentialities using a dominance-based rough set approach. J. Environ. Manage. 144, 160–167. Bolcárová, P., Koloˇsta, S., 2015. Assessment of sustainable development in the EU 27 using aggregated SD index. Ecol. Indic. 48, 699–705. Bond, A.J., Morrison-Saunders, A., 2011. Re-evaluating sustainability assessment: aligning the vision and the practice. Environ. Impact Assess. Rev. 31 (1), 1–7. Bottero, M., Comino, E., Duriavig, M., Ferretti, V., Pomarico, S., 2013. The application of a multicriteria spatial decision support system (MCSDSS) for the assessment of biodiversity conservation in the Province of Varese (Italy). Land Use Policy 30, 730–738. Cassar, L.F., Conrad, E., Bell, S., Morse, S., 2013. Assessing the use and influence of sustainability indicators at the European periphery. Ecol. Indic. 35, 52–61. Cassatella, C., Peano, A. (Eds.), 2011. Assessing and Monitoring Landscape Quality. Springer Science + Business Media B.V. Chakhar, S., Martel, J.M., 2003. Enhancing geographical information systems capabilities with multi-criteria evaluation functions. J. Geogr. Inf. Decis. Anal. 2 (7), 47–71. Chakhar, S., Mousseau, V., 2008. Multicriteria decision making, spatial. In: Shekhar, S., Xiong, H. (Eds.), Encyclopedia of GIS. Springer, pp. 747–758. Chen, W.M., Kim, H., Yamaguchi, H., 2014. Renewable energy in eastern Asia: renewable energy policy review and comparative SWOT analysis for promoting renewable energy in Japan, South Korea and Taiwan. Energy Policy 74, 319–329. Choo, E.U., Schoner, B., Wedley, W.C., 1999. Interpretation of criteria weights in multicriteria decision making. Comput. Ind. Eng. 37, 527–541. Comino, E., Bottero, M., Pomarico, S., Rosso, M., 2014. Exploring the environmental value of ecosystem services for a river basin through a spatial multicriteria analysis. Land Use Policy 36, 381–395. Comitato Interministeriale per la Programmazione Economica. 2003a. Delibera n. 17/2003. Ripartizione delle risorse per interventi nelle aree sottoutilizzate −rifinanziamento Legge 208/1998, Triennio 2003–2005 (Legge finanziaria 2003, art. 61). Comitato Interministeriale per la Programmazione Economica. 2003b. Delibera n. 83/2003. Ripartizione accantonamento di 900 milioni di Euro per interventi nelle aree sottoutilizzate (punto 1.1, delibera 17/2003). Coyle, G., 2004. Practical Strategy. Structured Tools and Techniques. Pearson Education Limited. Daskon, C.D., 2010. Cultural resilience – the roles of cultural traditions in sustaining rural livelihoods: a case study from rural kandyan villages in central Sri Lanka. Sutainability 2, 1080–1100. De Luca, M., Dell’Acqua, G., Lamberti, R., 2012. High-speed rail track design using GIS and multi-criteria analysis. Procedia—Soc. Behav. Sci. 54, 608–617. De Noronha Vaz, T., Nijkamp, P., Rastoin, J.L., 2009. Traditional Food Production and Rural Sustainable Development. Ashgate Economic Geography Series, London. Demesouka, O.E., Vavatsikos, A.P., Anagnostopoulos, K.P., 2013. Suitability analysis for siting MSW landfills and its multicriteria spatial decision support system: method, implementation and case study. Waste Manage. 33, 1190–1206. Development Team, Q.G.I.S., 2013. QGIS geographic information system. Open Source Geospatial Found. Proj. http://qgis.osgeo.org. Devuyst, D., 2001. Introduction to sustainability assessment at the local level. In: Devuyst, D. (Ed.), How Green Is the City? Sustainability Assessment and the Management of Urban Environments. Columbia University Press, New York, pp. 1–41. Di Domenico, M., Miller, G., 2012. Farming and tourism enterprise: experiential authenticity in the diversification of independent small-scale family farming. Tour. Manage. 33, 285–294. Dragan, M., Feoli, E., Fernetti, M., Zerihun, W., 2003. Application of a spatial decision support system (SDSS) to reduce soil erosion in northern Ethiopia. Environ. Modell. Softw. 18 (10), 861–868. European Commission. 1992. Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora. European Commission, 2005. Council Regulation no. 1698/2005 of 20 September 2005 on support for rural development by the European Agricultural Fund for Rural Development (EAFRD). European Commission. 2006. Council Regulation no. 510/2006 of 20 March 2006 on the protection of geographical indications and designations of origin for agricultural products and foodstuffs. European Commission. 2009. European Parliament and Council Directive 2009/147/EC of 30 November 2009 on the conservation of wild birds. European Commission. 2010. Communication from the Commission to the European Parliament, the Council, the European economic and social Committee and the Committee of the Regions. The CAP towards 2020: Meeting the food, natural resources and territorial challenges of the future. European Commission. 2013. Overview of CAP reform 2014–2020. Agricultural Policy Perspectives. Brief, no. 5, December 2013.

25

European Network for Rural Development. 2013. The European Agricultural Fund for Rural Development. Examples of Leader Projects. European Parliament, 2013. Regulation (EU) No 1303/2013 of the European Parliament and of the Council of 17 December 2013 laying down common provisions on the European Regional Development Fund, the European Social Fund, the Cohesion Fund, the European Agricultural Fund for Rural Development and the European Maritime and Fisheries Fund and laying down general provisions on the European Regional Development Fund, the European Social Fund, the Cohesion Fund and the European Maritime and Fisheries Fund and repealing Council Regulation (EC) No 1083/2006. European Union, 2007. Council Regulation no. 1234/2007 of 22 October 2007 establishing a common organisation of agricultural markets and on specific provisions for certain agricultural products (Single CMO Regulation). European Union. 2012. Regulation no. 1151/2012 of the European Parliament and of the Council of 21 November 2012 on quality schemes for agricultural products and foodstuffs. Gazzetta Ufficiale Serie Generale, no. 152, 01 July 2004. Decreto del Presidente della Repubblica, 10 marzo 2004. Istituzione del Parco Nazionale dell’Alta Murgia. http://www.gazzettaufficiale.it/eli/id/2004/07/01/04A06570/sg. Gilliams, S., Raymaekers, D., Muys, B., Orshoven, J.V., 2005. Comparing multiple criteria decision methods to extend a geographical information system on afforestation. Comput. Electron. Agric. 49, 142–158. Goh, H.H., Kok, B.C., Yeo, H.T., Lee, S.W., Mohd Zin, A.A., 2013. Combination of TOPSIS and AHP in load shedding scheme for large pulp mill electrical system. Electr. Power Energy Syst. 47, 198–204. Greco, S., Matarazzo, B., Slowinski, R., Stefanowski, J., 2000. An algorithm for induction of decision rules consistent with the dominance principle. In: Revised Papers from the Second International Conference on Rough Sets and Current Trends in Computing, October 16 −19, 2000, pp. 304–313. ´ Greco, S., Matarazzo, B., Slowinski, R., 2001. Rough sets theory for multi-criteria decision analysis. Eur. J. Oper. Res. 129 (1), 1–47. Greene, R., Luther, J.E., Devillers, R., Eddy, B., 2010. An approach to GIS-based multiple criteria decision analysis that integrates exploration and evaluation phases: case study in a forest-dominated landscape. For. Ecol. Manage. 260 (12), 2102–2114. Hacking, T., Guthrie, P., 2008. A framework for clarifying the meaning of triple bottom-line, integrated, and sustainability assessment. Environ. Impact Assess. Rev. 28, 73–89. Haertsch, S., Smith, A., 2012. Land use conflict mapping. In: Paper Presented at the Planning Institute of Australia, 2012, National Congress, Adelaide, South Australia. Hanine, M., Boutkhoum, O., Tikniouine, A., Agouti, T., 2016. Application of an integrated multi-criteria decision making AHP-TOPSIS methodology for ETL software selection. SpringerPlus 5 (263), 1–17. Helms, M.M., Nixon, J., 2010. Exploring SWOT Analysis −where are we now? A review of academic research from the last decade. J. Strategy and Management 3 (3), 215–251. Henry, A., 2008. Understanding Strategic Management. University Press, Oxford. Hezri, A., Dovers, S., 2006. Sustainability indicators, policy and governance: issues for ecological economics. Ecol. Econ. 60, 86–99. Hill, T., Westbrook, R., 1997. SWOT analysis: it’s time for a product recall. Long Range Plann. 30 (1), 46–52. Hwang, C.L., Yoon, K., 1981. Multiple Objective Decision Making Methods and Applications, A State of the Art Survey. Springer-Verlag. Ishizaka, A., Labib, A., 2009. Analytic hierarchy process and expert choice: benefits and limitations. OR Insight 22 (4), 201–220. Ishizaka, A., Lusti, M., 2006. How to derive priorities in AHP: a comparative study. Cent. Eur. J. Oper. Res. 14, 387–400. Istituto Nazionale di Statistica, ISTAT, 2010 6◦ Censimento Generale dell’Agricoltura. http://www.censimentoagricoltura.istat.it/. Istituto Nazionale di Statistica, ISTAT, 2011. 15◦ Censimento Generale della Popolazione e delle Abitazioni. http://www.istat.it/it/censimento-popolazione/ censimento-popolazione-2011. Jankowski, P., 1995. Integrating geographical information systems and multiple criteria decision-making methods. Int. J. Geogr. Inf. Syst. 9, 251–273. Jelokhani-Niaraki, M., Malczewski, J., 2015a. A group multicriteria spatial decision support system for parking site selection problem: a case study. Land Use Policy 42, 492–508. Jelokhani-Niaraki, M., Malczewski, J., 2015b. Decision complexity and consensus in web-based spatial decision making: a case study of site selection problem using GIS and multicriteria analysis. Cities 45, 60–70. Kaiser, B., 1990. La Renaissance Rurale. Armand Colin, Paris. Kurttilaa, M., Mauno Pesonena, U., Kangasb, J., Kajanusa, M., 2000. Utilizing the analytic hierarchy process AHP in SWOT analysis. A hybrid method and its application to a forest-certification case. For. Policy Econ. 1, 41–52. Laskar, A., 2003. Integrating GIS and Multi Criteria Decision Making Techniques for Land Resource Planning. International Institute for Aerospace survey and Earth Sciences (ITC), Enschede (Netherlands). Lidouh, K., 2013. On the motivation behind MCDA and GIS integration. Int. J. Multicriteria Decis. Mak. 3, 101–113. Lukesch, R., 2007. The LAG-Handbook. A Guide Through the Stunning World of Local Action Groups. Leader+ Observatory, Bruxelles, Belgium. MacKinnon, D., 2002. Rural governance and local involvement: assessing state-community relations in the Scottish Highlands. J. Rural Stud. 18, 307–324. Malczewski, J., Rinner, C., 2015. Multicriteria Decision Analysis in Geographic Information Science. Springer.

26

G. Ottomano Palmisano et al. / Land Use Policy 59 (2016) 12–26

Malczewski, J., 1999. GIS and Multicriteria Decision Analysis. John Wiley & Sons Inc, New York. Malczewski, J., 2006. GIS-based multicriteria decision analysis: a survey of the literature. Int. J. Geogr. Inf. Sci. 7 (20), 703–726. Malczewski, J., 2010. Multiple criteria decision analysis and geographic information systems. In: Ehrgott, M., Figueira, J., Greco, S. (Eds.), International Series in Operations Research & Management Science, vol 142. Springer, New York, pp. 369–395. Marsden, T., Sonnino, R., 2008. Rural development and the regional state: denying multifunctional agriculture in the UK. J. Rural Stud. 24, 422–431. Massei, G., Boggia, A., 2014. GeoUmbriaSUIT. Manuale d’uso E Tutorial. http:// maplab.alwaysdata.net/geoUmbriaSUIT.html. Massei, G., Rocchi, L., Paolotti, L., Greco, S., Boggia, A., 2014. Decision support systems for environmental management: a case study on wastewater from agriculture. J. Environ. Manage. 146, 491–504. Matthews, J.R., 2004. Technology Planning: Preparing and Updating a Library Technology Plan. Libraries Unlimited, London. Mennella, V.G.G. (Ed.), 2006. Greenway Per Lo Sviluppo Sostenibile Del Territorio. Il Verde Editoriale. Moffat, I., Hanley, N., Wilson, M.D., 2001. Measuring and Modelling Sustainable Development. The Parthenon Publishing Group, Bristol, UK. Morrison-Saunders, A., Pope, J., Bond, A., Retief, F., 2014. Towards sustainability assessment follow-up. Environ. Impact Assess. Rev. 45, 38–45. Neagu, C., Bulearc˘a, M., Sima, C., M˘argus¸, T., 2015. A SWOT analysis of romanian extractive industry and re-industrialization requirements of this industry. Procedia Econ. Finance 22, 287–295. Ness, B., Urbel-Piirsalu, E., Anderberg, S., Olsson, L., 2007. Categorising tools for sustainability assessment. Ecol. Econ. 60, 498–508. Organization for Economic Co-operation and Development (OECD), 2003. OECD Environmental Indicators. Development, Measurement and Use. http://www. oecd.org/environment/indicators-modelling-outlooks/24993546.pdf. Osservatorio del Turismo della Regione Puglia, 2013. Report Del Movimento Turistico Annuale Per Comune. http://www.agenziapugliapromozione.it/ portal/osservatorio-del-turismo. Ozawa, C., 1999. Making the best use of technology. In: Susskind, L., McKearnan, S., Larmer, J.T. (Eds.), The Consensus Building Handbook: A Comprehensive Guide to Reaching Agreement. SAGE Publications Inc, Thousand Oaks, CA, pp. 401–435. ˙ V., 2013. Rural development and challenges Paˇsakarnis, G., Morley, D., Maliene, establishing sustainable land Eastern European countries use in Eastern European Countries. Land Use Policy 30, 703–710. Pahl, N., Richter, A., 2007. SWOT Analysis—Idea, Methodology and a Practical Approach. GRIN Verlag. Pope, J., Annandale, D., Morrison-Saunders, A., 2004. Conceptualising sustainability assessment. Environ. Impact Assess. Rev. 24, 595–616. Rahman, M.A., Rusteberg, B., Gogu, R.C., Lobo Ferreira, J.P., Sauter, M., 2012. A new spatial multi-criteria decision support tool for site selection for implementation of managed aquifer recharge. J. Environ. Manage. 99, 61–75. Regione Puglia, 2007. Delibera Giunta Regionale n. 1792, 31 Ottobre 2007. Redazione della Carta Idrogeomorfologica della Regione Puglia. Bollettino Ufficiale della Regione Puglia, n. 165, 20 Novembre 2007. Regione Puglia, 2015. Deliberazione Giunta Regionale n. 176, 16 Febbraio 2015. Approvazione del Piano Paesaggistico Territoriale della Regione Puglia (PPTR). Bollettino Ufficiale della Regione Puglia, n. 40, 23 Marzo 2015. http:// paesaggio.regione.puglia.it/PPTR 2015/DGR 176 2015 burp.pdf.

Rosenström, U., Kyllonen, S., 2007. Impacts of a participatory approach to developing national level sustainable development indicators in Finland. J. Environ. Manage. 84, 282–298. Saaty, T.L., Peniwaty, K., 2007. Group Decision Making: Drawing Out and Reconciling Differences. RWS Publications, Pittsburgh. Saaty, T.L., Vargas, L.G., 2012. Models, Methods, Concepts & Applications of the Analytic Hierarchy Process. Springer Science + Business Media, New York. Saaty, T., 1977. A scaling method for priorities in hierarchical structures. J. Math. Psychol. 15 (3), 234–281. Saaty, T.L., 1980. The Analytic Hierarchy Process. McGraw-Hill, New York. Saaty, T.L., 2012. Decision Making for Leaders. RWS Publications, Pittsburgh. Schmidt Di Friedberg, P., 1986. Gli Indicatori Ambientali. Franco Angeli, Milano. Schmoldt, D.L., Peterson, D., 2000. Analytical group decision making in natural resources: methodology and application. For. Sci. 46, 62–75. Sharpley, R., Vaas, A., 2006. Tourism, farming and diversification: an attitudinal study. Tour. Manage. 27, 1040–1052. Silva, S., Alc¸ada-Almeida, L., Dias, L.C., 2014. Development of a web-based multi-criteria spatial decision support system for the assessment of environmental sustainability of dairy farms. Comput. Electron. Agric. 108, 46–57. Soto, P., Ramsden, P., 2014. Guidance on community-led local development for local actors. European Commission http://ec.europa.eu/regional policy/ sources/docgener/informat/2014/guidance clld local actors.pdf. Srivastava, P.K., Kulshreshtha, K., Monhanty, C.S., Pushpangadan, P., Singh, A., 2005. Stakeholder-based SWOT analysis for successful municipal solid waste management in Lucknow, India. Waste Manage. 25 (5), 531–537. Sugumaran, M., DeGroote, J., 2011. Spatial Decision Support Systems Principle and Practices. CRC Press, Boca Raton, London, New York. Tapiador, F.J., 2008. Rural Analysis and Management. An Earth Science Approach to Rural Science. Springer. Van Huylenbroeck, G., Vandermeulen, V., Mettepenningen, E., Verspecht, A., 2007. Multifunctionality of agriculture: a review of definitions, evidence and instruments. Living Rev. Landsc. Res. 3 (1). Vaskan, P., Passuello, A., Guillén-Gosálbez, G., Schuhmacher, M., Jiménez, L., 2013. Combined use of GIS and mixed-integer linear programming for identifying optimal agricultural areas for sewage sludge amendment: a case study of Catalonia. Environ. Modell. Softw. 46, 163–169. Wanderer, T., Herle, S., 2015. Creating a spatial multi-criteria decision support system for energy related integrated environmental impact assessment. Environ. Impact Assess. Rev. 52, 2–8. World Commission on Environment and Development, 1987. Our Common Future. Oxford University Press. Yu, J., Wang, L., Gong, X., 2013. Study on the status evaluation of urban road intersections traffic congestion base on AHP-TOPSIS modal. Procedia—Soc. Behav. Sci. 96, 609–616. Zaidan, A.A., Zaidan, B.B., Al-Haiqi, A., Kiah, M.L.M., Hussain, M., Abdunlnabi, M., 2015. Evaluation and selection of open-source EMR software packages based on integrated AHP and TOPSIS. J. Biomed. Inform. 53, 390–404. Zhang, H., Chen, M., 2013. Research on the recycling industry development model for typical exterior plastic components of end-of-life passenger vehicle based on the SWOT method. Waste Manage. 33 (11), 2341–2353.