weighting. The Resistance-to-Change Methodology results in decision-makers automatically placing the criteria into a hierarchy of relative importances. This hierarchy
can then be used to compute directly the weighting for each criterion.
The key to using this method lies in deriving two terms for each criterion – one
expressing its most desirable aspect and the other expressing its least desirable
aspect. This is termed expressing the criterion in ‘bipolar form’. The six criteria
from the last three examples can be expressed in bipolar form as follows:
• C1 – economical waste treatment/costly waste treatment
• C2 – low level of construction disturbance/high level of construction disturbance
• C3 – negligible impact on birdlife/significant impact on birdlife
• C4 – negligible visual impact/significant visual impact
• C5 – high-quality water/low-quality water
• C6 – highly acceptable politically/politically unacceptable
The left-hand side of each of the above bipolar expressions represents the
criterion at its most desirable, with the right-hand side representing it at its least
desirable. For an economic criterion, the preferred side is assumed to be the one
minimising cost. For an environmental/social criterion, the preferred side is assumed
to be the one minimising environmental/social impact.
In order to derive a pairwise comparison matrix of the type illustrated in
Table 15.15 for each pair of criteria, the decision-maker is required to examine the
bipolar expressions for both. He is then asked the question that, if he had to change
one of these criteria from its desirable side to its undesirable side, which one would
he be least willing to change? For each comparison, the criterion resisting change is
given the score ‘1’ while the criterion which the decision-maker is willing to change
to its undesirable side is given the score ‘0’. If the decision-maker has an equal
resistance to changing either criterion, each is given the score ‘0.5’. When a given
criterion has been compared for resistance to change with all others, its scores are
totalled to give a final weighting. All weightings obtained are then summed and each
is divided by this figure to attain normalisation.
The theoretical basis for the grid originated from Hinkle (1965) who found that
the more important a criterion is, the more likely it will be that the decision-maker
resists changing it to its more undesirable state.
An example of the ‘resistance-to-change’ grid that yields weightings identical to
the basic pairwise technique in Example 15.6 is illustrated in Table 15.16. Looking
in detail at criterion C 3 (impact on birdlife), the overall score of 2.5 out of a possible
5 indicates the following:
• The decision-maker preferred the idea of an economical waste plant having a
significant impact on birdlife than a costly one having minimal impact on birdlife.
Hence, in this case, C 3 was assigned the score ‘0’.
• The decision-maker preferred the idea of a waste plant having a minimal impact
on birdlife but creating significant construction impacts rather than one having
significant impact on birdlife but creating minimal construction disturbance.
Hence, in this case, C 3 was assigned the score ‘1’.
432
15 Multi Objective Evaluation Criteria for Infrastructure
can then be used to compute directly the weighting for each criterion.
The key to using this method lies in deriving two terms for each criterion – one
expressing its most desirable aspect and the other expressing its least desirable
aspect. This is termed expressing the criterion in ‘bipolar form’. The six criteria
from the last three examples can be expressed in bipolar form as follows:
• C1 – economical waste treatment/costly waste treatment
• C2 – low level of construction disturbance/high level of construction disturbance
• C3 – negligible impact on birdlife/significant impact on birdlife
• C4 – negligible visual impact/significant visual impact
• C5 – high-quality water/low-quality water
• C6 – highly acceptable politically/politically unacceptable
The left-hand side of each of the above bipolar expressions represents the
criterion at its most desirable, with the right-hand side representing it at its least
desirable. For an economic criterion, the preferred side is assumed to be the one
minimising cost. For an environmental/social criterion, the preferred side is assumed
to be the one minimising environmental/social impact.
In order to derive a pairwise comparison matrix of the type illustrated in
Table 15.15 for each pair of criteria, the decision-maker is required to examine the
bipolar expressions for both. He is then asked the question that, if he had to change
one of these criteria from its desirable side to its undesirable side, which one would
he be least willing to change? For each comparison, the criterion resisting change is
given the score ‘1’ while the criterion which the decision-maker is willing to change
to its undesirable side is given the score ‘0’. If the decision-maker has an equal
resistance to changing either criterion, each is given the score ‘0.5’. When a given
criterion has been compared for resistance to change with all others, its scores are
totalled to give a final weighting. All weightings obtained are then summed and each
is divided by this figure to attain normalisation.
The theoretical basis for the grid originated from Hinkle (1965) who found that
the more important a criterion is, the more likely it will be that the decision-maker
resists changing it to its more undesirable state.
An example of the ‘resistance-to-change’ grid that yields weightings identical to
the basic pairwise technique in Example 15.6 is illustrated in Table 15.16. Looking
in detail at criterion C 3 (impact on birdlife), the overall score of 2.5 out of a possible
5 indicates the following:
• The decision-maker preferred the idea of an economical waste plant having a
significant impact on birdlife than a costly one having minimal impact on birdlife.
Hence, in this case, C 3 was assigned the score ‘0’.
• The decision-maker preferred the idea of a waste plant having a minimal impact
on birdlife but creating significant construction impacts rather than one having
significant impact on birdlife but creating minimal construction disturbance.
Hence, in this case, C 3 was assigned the score ‘1’.
432
15 Multi Objective Evaluation Criteria for Infrastructure