**2. Methods**

#### **2.1 The study area**

Negeri Sembilan is located between 2°.43<sup>0</sup> 54.5268″ N latitude and 102°.15<sup>0</sup> 9.0072″ E longitude, occupying about 6645 km<sup>2</sup> in Malaysia with the population of about 1.7 million people in 2012. It borders Selangor at the northeast, Pahang at the north, Johor at the east, and Melaka at the south with the capital in Seremban. The people are engaged in farming paddy rice at the valley of the steep hills, rubber trees are extensively cultivated, and oil palms are grown. Prawn farming is one of the major livelihoods for the inhabitant. The State contributed about 22% of prawn production in Malaysia in 2014 [3].

Fisheries (DoF) who were knowledgeable in the field of prawn farming and have relevant information to assign weight to each criterion [10, 11]. However, out of the 30 experts, only 20 were found consistent. These pairwise comparisons were then applied as the input to generate a ratio matrix, and the relative weights are created

*Site Suitability Analysis of Infrastructure Facilities for Giant Freshwater Prawn Farming*

The AHP pairwise comparison matrix calculates the weight for each criterion and factor (*wi*) by taking the eigenvector corresponding to the largest eigenvalue of the matrix and normalising the sum of the component to 1 as expressed below:

Then the final importance of each criterion was calculated. The main input is the pairwise comparison matrix 'A' of *n* criteria proposed by Saaty's scale, in the order

where A is a matrix with elements aij. The matrix usually has a reciprocity

in which B is the normalised matrix of A with the elements bij expressed as

*<sup>i</sup>*¼<sup>1</sup>*aij* <sup>¼</sup> 1, 2, 3, … *<sup>n</sup>*

A mistake may be made in preference during the survey stage [13]. Therefore, Saaty [14] introduced a single mathematical index to make sure that the pairwise comparison matrix is consistent by applying the consistency ratio (CR). The consistency ratio is computed by dividing the *CI* by the *RI.* The equation is expressed as

*CR* <sup>¼</sup> *CI*

where (*CR*) = consistency ratio; (*CI*) = consistency index; (*RI*) = random index

To ensure the reliability of the relative importance applied, the AHP provides a certain measure to determine inconsistency of the judgments. Based on the priorities of the reciprocal matrices, the consistency ratio can be calculated. CR < 0.1 indicates that the level of consistency in the pairwise comparison is acceptable. But if CR > 0.10 it means that there is inconsistency in the evaluation process and the

(mean value) depending on the computed matrix order set by Saaty [8].

*bij* <sup>¼</sup> aij P*<sup>n</sup>*

> P*<sup>n</sup> <sup>i</sup>*¼<sup>1</sup>*bij*

P*<sup>n</sup> i*¼1 P*<sup>n</sup>*

After creating this matrix, it is then normalised as matrix B:

Each weight value is computed as shown below:

*wi* ¼

*wi* ¼ 1 (1)

A ¼ ½ � aij , ij ¼ 1, 2, 3 … n (2)

aij ¼ 1*=*aji ð Þ *n n*ð Þ � *1 =2* (3)

B ¼ ½ � Bij , ij ¼ 1, 2, 3 … n (4)

*<sup>i</sup>*¼<sup>1</sup>*bij* , *<sup>i</sup>*, *<sup>j</sup>* <sup>¼</sup> 1, 2, 3 … , *<sup>n</sup>* (6)

*RI* (7)

(5)

X*n i*¼1

as the output [12].

defined as.

follows:

**95**

of (*nxn*) as described in Eq. (2):

*DOI: http://dx.doi.org/10.5772/intechopen.90659*

#### **2.2 Factors and class weights (standardisation)**

The actual factors and class weights of the parameters involved in the study are needed to generate the land suitability map. The AHP was systematically used to determined theses. The AHP criteria were developed based on the expert's survey interviews. The experts were asked to determine the relative importance of each factor. The process assesses the relative importance of all the parameters by allocating weights for each of them in the hierarchy order, and the suitability weight for each class of the factors was assigned in the last level of the hierarchy. Usually, the priority of each factor involved in the AHP analysis is calculated based mainly on the opinions of the experts [4, 5]. Prioritisation is the determination of the relative significance of the criteria which needs brain stomping among experts to assign values on a Saaty's nine-point scale [6] for a pairwise comparison of criteria.

The pairwise comparison matrix was applied to determine the weight and consistency of each criterion at each level of hierarchy by relative rating. The ninepoint rating scale was used where 1 represents equal importance (i.e., two factors contributing equally to the objective), 3 represents moderate importance (one factor slightly favoured over another), 5 indicates strongly important, 7 stands for very strongly important, and 9 stands for extreme importance (as earlier mentioned in Chapter 3 of this study). It is a score systematically indicating the relative rating from most important (9–1) or the least important (1/9, 1/8, 1/7 … ..1/2, 1). The AHP results in combination with information collected from other methods were used to describe the land suitability analysis for prawn farm identification, as well as the opportunities and challenges for prawn farming in the current site.

The AHP as a multi-criteria evaluation method was applied to determine the weight of each criterion. The principle behind AHP is in the construction of a threelevel hierarchy model with the goal, the criteria (objectives), and the sub-criteria (attributes) which are at the bottom layer of the hierarchy [7]. Inputs of experts are considered as the pairwise comparison, and the best criteria will be selected according to the highest rank between the criteria.

Multi-criteria evaluation is a process that incorporates multiple and conflicting criteria, which allow solving a wide range of complex problems and transforming them into decision-making. The AHP developed by Saaty [8] for doing pairwise comparison matrix is a tool required for comparing alternatives with respect to a set of criteria. The criteria were ranked according to the order of importance. Some relative weights were assigned to the criteria indicating the degree of importance or preference of each criterion with respect to the other criterion.

Expert opinion was usually required to rank the criteria by assigning a score to each criterion [9]. For this study 30 experts were drawn from the Department of

*Site Suitability Analysis of Infrastructure Facilities for Giant Freshwater Prawn Farming DOI: http://dx.doi.org/10.5772/intechopen.90659*

Fisheries (DoF) who were knowledgeable in the field of prawn farming and have relevant information to assign weight to each criterion [10, 11]. However, out of the 30 experts, only 20 were found consistent. These pairwise comparisons were then applied as the input to generate a ratio matrix, and the relative weights are created as the output [12].

The AHP pairwise comparison matrix calculates the weight for each criterion and factor (*wi*) by taking the eigenvector corresponding to the largest eigenvalue of the matrix and normalising the sum of the component to 1 as expressed below:

$$\sum\_{i=1}^{n} wi = \mathbf{1} \tag{1}$$

Then the final importance of each criterion was calculated. The main input is the pairwise comparison matrix 'A' of *n* criteria proposed by Saaty's scale, in the order of (*nxn*) as described in Eq. (2):

$$\mathbf{A} = [\mathbf{a}\mathbf{i}], \mathbf{i}\mathbf{j} = \mathbf{1}, \mathbf{2}, \mathbf{3} \dots \mathbf{n} \tag{2}$$

where A is a matrix with elements aij. The matrix usually has a reciprocity defined as.

$$\mathbf{a}\ddot{\mathbf{i}} = \mathbf{1}/\mathbf{a}\ddot{\mathbf{j}} \ (n(n-1)/2) \tag{3}$$

After creating this matrix, it is then normalised as matrix B:

$$\mathbf{B} = [\mathbf{B}\mathbf{i}] / \mathbf{i} \mathbf{j} = \mathbf{1}, \mathbf{2}, \mathbf{3} \dots \mathbf{n} \tag{4}$$

in which B is the normalised matrix of A with the elements bij expressed as

$$bij = \frac{\text{aij}}{\sum\_{i=1}^{n} aij = 1, 2, 3, \dots n} \tag{5}$$

Each weight value is computed as shown below:

$$wi = \frac{\sum\_{i=1}^{n} bij}{\sum\_{i=1}^{n} \sum\_{i=1}^{n} bij}, i, j = 1, 2, 3 \dots, n \tag{6}$$

A mistake may be made in preference during the survey stage [13]. Therefore, Saaty [14] introduced a single mathematical index to make sure that the pairwise comparison matrix is consistent by applying the consistency ratio (CR). The consistency ratio is computed by dividing the *CI* by the *RI.* The equation is expressed as follows:

$$CR = \frac{CI}{RI} \tag{7}$$

where (*CR*) = consistency ratio; (*CI*) = consistency index; (*RI*) = random index (mean value) depending on the computed matrix order set by Saaty [8].

To ensure the reliability of the relative importance applied, the AHP provides a certain measure to determine inconsistency of the judgments. Based on the priorities of the reciprocal matrices, the consistency ratio can be calculated. CR < 0.1 indicates that the level of consistency in the pairwise comparison is acceptable. But if CR > 0.10 it means that there is inconsistency in the evaluation process and the

**2. Methods**

**2.1 The study area**

production in Malaysia in 2014 [3].

**2.2 Factors and class weights (standardisation)**

Negeri Sembilan is located between 2°.43<sup>0</sup> 54.5268″ N latitude and 102°.15<sup>0</sup> 9.0072″ E longitude, occupying about 6645 km<sup>2</sup> in Malaysia with the population of about 1.7 million people in 2012. It borders Selangor at the northeast, Pahang at the north, Johor at the east, and Melaka at the south with the capital in Seremban. The people are engaged in farming paddy rice at the valley of the steep hills, rubber trees are extensively cultivated, and oil palms are grown. Prawn farming is one of the major livelihoods for the inhabitant. The State contributed about 22% of prawn

*Emerging Technologies, Environment and Research for Sustainable Aquaculture*

The actual factors and class weights of the parameters involved in the study are needed to generate the land suitability map. The AHP was systematically used to determined theses. The AHP criteria were developed based on the expert's survey interviews. The experts were asked to determine the relative importance of each factor. The process assesses the relative importance of all the parameters by allocating weights for each of them in the hierarchy order, and the suitability weight for each class of the factors was assigned in the last level of the hierarchy. Usually, the priority of each factor involved in the AHP analysis is calculated based mainly on the opinions of the experts [4, 5]. Prioritisation is the determination of the relative significance of the criteria which needs brain stomping among experts to assign values on a Saaty's nine-point scale [6] for a pairwise comparison of criteria.

The pairwise comparison matrix was applied to determine the weight and consistency of each criterion at each level of hierarchy by relative rating. The ninepoint rating scale was used where 1 represents equal importance (i.e., two factors contributing equally to the objective), 3 represents moderate importance (one factor slightly favoured over another), 5 indicates strongly important, 7 stands for very strongly important, and 9 stands for extreme importance (as earlier mentioned in Chapter 3 of this study). It is a score systematically indicating the relative rating from most important (9–1) or the least important (1/9, 1/8, 1/7 … ..1/2, 1). The AHP results in combination with information collected from other methods were used to describe the land suitability analysis for prawn farm identification, as well as the

The AHP as a multi-criteria evaluation method was applied to determine the weight of each criterion. The principle behind AHP is in the construction of a threelevel hierarchy model with the goal, the criteria (objectives), and the sub-criteria (attributes) which are at the bottom layer of the hierarchy [7]. Inputs of experts are

Multi-criteria evaluation is a process that incorporates multiple and conflicting criteria, which allow solving a wide range of complex problems and transforming them into decision-making. The AHP developed by Saaty [8] for doing pairwise comparison matrix is a tool required for comparing alternatives with respect to a set of criteria. The criteria were ranked according to the order of importance. Some relative weights were assigned to the criteria indicating the degree of importance or

Expert opinion was usually required to rank the criteria by assigning a score to each criterion [9]. For this study 30 experts were drawn from the Department of

considered as the pairwise comparison, and the best criteria will be selected

opportunities and challenges for prawn farming in the current site.

preference of each criterion with respect to the other criterion.

according to the highest rank between the criteria.

**94**


#### **Table 1.**

*Table of random index (RI).*

process needs to be recomputed or else the AHP may not yield meaningful results [6].

Consistency ratio simplifies the assessment of possible events and measures logical inconsistencies of the decision-maker and judgement [15]. It denotes the probability where the matrix judgments were randomly formed [16].

The consistency ratio depends upon the eigenvector (λMa**x**) and the consistency index (CI). Therefore, one needs to find the vector *w* of the order *n* such that *A* � *w =* λ � w.

where *w* is the eigenvector (i.e., weight vector) and λ is the eigenvalue.

where λma**<sup>x</sup>** � *w* ≥ n and λma**<sup>x</sup>** is the principal eigenvalue of the matrix. Therefore, the inconsistency of the judgement is reflected in the differences between λma**<sup>x</sup>** and *n*. The process computes a consistency index (CI) to check the consistency of the pairwise comparison matrix:

$$\text{CI} = \frac{\lambda\_{\text{max}} - n}{n - 1} \tag{8}$$

facilities, prawn farming will not be successful [17]. Major differences in the factors ranking for giant freshwater prawn farming were the relative importance placed on infrastructure where they were higher weights in relation to water and soil as

*Site Suitability Analysis of Infrastructure Facilities for Giant Freshwater Prawn Farming*

The infrastructural facilities aspects of the study comprised of distance to road, distance to market, distance to electricity, distance to fry source, or hatchery layers which were overlay in the GIS environment to generate the overall infrastructure

Prawn farming operations were affected by infrastructural factors [18]. One of the requirements for successful prawn farming is a good road network. Foods and other necessary equipment are transported to the farm and market. Therefore, prawn farms should be close to the road for easy and quick access. The distance to road suitability classification showed that 90.5% of Negeri Sembilan was close to access roads within less than 2 km. Road accessibility was limited in the hilly and

Electricity is a vital factor in determining the success of prawn farming due to the power supply to power the farm's machines for prawn production [19]. Any area lying within less than 3 km to the main electric power supply line are considered the most suitable site for prawn farming. Major roads and cities coincided with areas with good electricity supply in the study area. The area between 7 and 12 km was considered moderately suitable. In this study, more than half of the area have electricity distribution within the suitable range. The not suitable area falls within areas greater than 12 km away from the main source of electric power line occupy-

infrastructure plays a vital role in farming in the study area.

*DOI: http://dx.doi.org/10.5772/intechopen.90659*

dense forested area where roads construction was difficult.

ing 0.4% of the study area at the extreme north of the study area.

*The infrastructure facilities criteria suitability map for prawn farming in Negeri Sembilan,*

facilities map (**Figure 1**).

**Figure 1.**

**97**

**3.1 Infrastructural facilities criteria map for prawn farming**

where (*CI*) refers to the consistency index, (*n*) refers to the number of criteria used (**Table 1**), and (λMa**x)** refers to the average value of the consistency vector (the highest eigenvector).
