Analytic Network Process (ANP) for

Solar Power Plant Location Problem

 Introduction
 Every day, the sun sends out an enormous amount of energy, called solar energy.

 The sun produces energy in its inner core in a process called nuclear fusion. Only a
small part of the solar energy that the sun radiates into space reaches the earth, but that
is more than enough to supply all our energy needs.

 It takes the sun’s energy just a little over 8 minutes to travel the 93 million miles to
earth. It travels at a speed of 186,000 miles per second ( the speed of light).

 Today, people use solar energy to heat buildings and water and to generate electricity.

 Solar energy has great potential for the future:

 It is free, and unlimited.
 It is readily available and renewable.
 It does not pollute or damage the environment. It helps lessening the greenhouse
effect.
 It cannot be controlled by any one nation or industry.
 Introduction (cont.)
 Main problem of establishing a solar power plant is to determine its location.

 Solar power plant location problem is a typical multiple-criteria decision making (MCDM)
problem in the presence of various selection criteria and a set of possible alternatives.

 Among the available multi-attribute approaches, Analytic Hierarchy Process (AHP) (Saaty,
1981) and Analytic Network Process (ANP) (Saaty, 1996) can be used.

 AHP has the capability to combine different types of criteria in a multi-level decision
structure to obtain a single score for each alternative to rank the alternatives.

 AHP has a hierachical structure. It has a goal or source node or cluster. It also has a sink node
or cluster that representes alternatives. It has a linear top down structure with no feedback
from lower to higher levels. Alternatives are considered to be independent from each other.

 AHP is a special case of ANP with its assumptions about independence of upper levels from
lower levels and the independence of elements in a level.
 
 Introduction (cont.)

 The ANP provides a general framework to deal with decisions without

making assumptions about the independence of higher level elements from
lower level elelements and about the independence of elements within a level
as in a hierarchy.

 ANP uses a network without the need to specify levels. A network spreads
out in all directions and its clusters of elements are not arranged in a particular
order.

 In ANP ,(a more general form of AHP) all attributes and alternatives involved
are connected in a network system that accepts various dependencies. (inner
dependence, outer dependence)

 In this research, ANP is used to determine the best alternative (location of the
solar power plant ).
Figure: How a Hierarchy Compares to a Network (Saaty, 2005)
 Literature Review

 ANP was introduced by Saaty (1996)

 Hamalainen and Seppalainen (1996) presented ANP-based framework for a

nuclear power plant licensing problem in Finland.

 Meade and Presley (2002) used ANP for R&D project selection.

 Yurdakul (2003) used ANP to measure long-term performance of a

manufacturing company.

 Aragone´s-Beltran et al.(2010) suggested an ANP-based approach for the

selection of photovoltaics solar power plant investment projects.
ANP Network Diagram for Solar Power Plant Location Problem
 Application of ANP

 After constructing the ANP network diagram, the decision maker(s) compares the
elements at a given level on a pairwise basis to estimate their relative importance in
relation to the element at the immediate proceeding level.

 In conventional ANP, the pairwise comparison is made by using a ratio scale. A

frequently used scale is the nine-point scale developed by Saaty (1996) which
shows the participants` judgments or preferences.
 STEPS of ANP
 Step I. Model construction and problem structuring: In a typical ANP network, the
problem is defined using clusters and elements inside each cluster. The network also
defines the relationships and feedbacks among clusters, and among the elements in
each cluster, if applicable. (We have 4 clusters).

 Step II. Building pairwise comparison matrices:

 By using nine-point scale of Saaty, the decision maker(s) are asked to respond to
a series of pairwise comparisons with respect to an upper level “control”
criterion. These are constructed with respect to their relevant importance towards
the control criterion. (Note: aij=1/aji)

 In the case of interdependencies, components in the same level are viewed as

controlling components for each other. Levels may also be interdependent.

 Once the pairwise comparisons are completed, the local priority vector w(also
referred as e-Vector) is computed as the unique solution to; Aw ,  max w
where,  is the largest eigenvalue of A, and A is a pairwise decision matrix.
max
 STEPS of ANP
 Step III. Checking out consistency ratios (CR) for pairwise comparison matrices:
 After constructing all pairwise matrices, for each of them, the consistency ratio (CR)
should be calculated.
 The measure of inconsistency is called the consistency index (CI):  n
CI  max
n 1
 CR  CI is used to estimate directly the consistency of pairwise comparisons.
RI
 Random Consistency Index (RI), the average index for randomly generated weights.

n 2 3 4 5 6 7 8 9 10
RI than
 If the CR is less 0 10%,
0.58
the 0.90 1.12 1.24
comparisons 1.32 1.41otherwise
are acceptable, 1.45 1.51
they are not.

 Step IV. Pairwise comparison matrices of inter-dependencies: In order to reflect the

interdependencies in the network, pairwise comparisons among all the criteria are
constructed and their consistency ratios are calculated as we previously defined in Step
II and Step III.
 STEPS of ANP
 Step V. Limiting Supermatrix formation:
 A supermatrix is a two-dimensional matrix of elements by elements. The priority
vectors from the paired comparisons appear in the appropriate column of the
supermatrix.
 When an element has no influence on another element, its influence priority is
assigned zero.
 In the unweighted supermatrix, the sum of each column corresponds to the
number of comparison sets.
 Applying the cluster matrix numbers to their respective blocks in the unweighted
supermatrix yields the weighted supermatrix that is column stochastic (each
column sums to unity). This way, each elements’ relative overall weight is
determined among all elements (in all clusters).
 Raising the weighted supermatrix to powers gives the limiting global supermatrix.
This allows convergence of the interdependent relationships between the two
levels being compared, and a long-term, stable set of weights are obtained.
 Step VI. Selection of the best solar power plant alternative: The best alternative is the
alternative with the highest weight in the limiting supermatrix.
See Table 2
 Conclusions and Future Research

 Baskale (Van, Alternative A) is selected for the solar power plant location.

 In our study, a template of Microsoft EXCEL is used to make all necessary

calculations since we have a limited number of elements.

 For future study, a knowledge-based system or an expert system can be

integrated to ANP to help decision makers both make pair wise calculations
more concisely, and to interpret the results in each step of the ANP.