Civil Engineering Reference
In-Depth Information
MCDA approaches
alternatives are
explicitly known a
priori
Simulation-based
approaches
Building retrofit
approaches
alternatives are
implicitly defined
by an optimization
model
MOP approaches
Fig. 5
Categorization of methodological approaches for building retrofit
promoted the use of multi-criteria decision analysis methods. A wide range of
MCDA methods have been applied in the energy-planning area (Diakoulaki et al.
2005
). In an MCDA approach, it is necessary to define the problem clearly,
identify the actors involved in the decision-making process and their values,
develop a coherent set of evaluation criteria, and establish realistic alternatives. An
MCDA method is selected to aggregate the performance of each alternative
according to the set of criteria using the preferences elicited from the DM through
technical parameters. Most MCDA methods require weighting of the criteria,
although the meaning of weights may be very different from method to method.
The application of MCDA methods may provide a selection of the best alternative,
a ranking of the alternatives, or a sorting of the alternatives in predefined ordered
categories of merit. Most representative MCDA methods may be included in the
broad classifications of methods, developing an overall synthesis value (e.g.,
multi-attribute
value/utility
function
approaches,
AHP)
and
outranking-based
approaches (e.g., ELECTRE, PROMETHEE).
Blondeau et al. (
2002
) used both combinatorial method based on the multiple
attribute utility theory (MAUT) and outranking methods to determine the most
suitable ventilation strategy of a university building, that is, to ensure the best
possible indoor air quality and thermal comfort of the occupants, and the lower
energy consumption in case of accelerated diurnal or nocturnal ventilation and/or
air-conditioning. It was shown that the results of the analysis by combinatorial
method strongly depend on the definition of the total utility function, and the
pernicious effects may affect its validity. On the other hand, outranking method
most probably allows to best fit the DM's way of thinking, but their result is not
always as clear as the one obtained with combinatorial method.
Roulet et al. (
2002
) used principal component analysis, as well as multi-criteria
ranking method, based on ELECTRE III and VI algorithms to develop a method
for ranking office buildings (ORME—office rating methodology) according to an
extended list of parameters, including energy use for heating, cooling, and other
appliances, impact on external environment, indoor environment quality, and cost.
Outranking methods are also used by Rey. The ELECTRE III method is used to
rank office building retrofitting strategies.
