Energy Decisions and the Environment A Guide to the Use of Multicrit
Planning, operating, and policy making in the electric utility and natural gas sectors involves important trade-offs among economic, social, and environmental criteria. These trade-offs figure prominently in ongoing debates about how to meet growing energ
- PDF / 3,810,925 Bytes
- 30 Pages / 439.37 x 666.14 pts Page_size
- 28 Downloads / 181 Views
THE APPLICATION OF MCDM METHODS Let all good things be done decently and in good order NEW TESTAMENT, I Corinthians 14: 40 We owe almost all our knowledge not to those who have agreed, but to those who have differed CHARLES CALED COLTON, Lacon 2:121 (1825)
In this and the following three chapters, we present a discussion of the sequence of steps involved in applying MCDM methods to a decision problem. Although most of the examples included here are drawn from energy resource selection, these steps are largely independent of the particular application involved, and apply to a wide spectrum of energy applications that range from power plant siting to bidding programs and the assessment of corporate environmental performance. l We begin with an overview of the process, and then discuss each ofthe steps in tum. The application of MCDM methods involves the following steps: Problem definition steps: 1. Selection and definition of attributes, say Ai, i= 1, ... 1, chosen to reflect important planning objectives and/or environmental concerns. System cost, reliability, effects on prices, air quality impacts, or impact on fisheries are examples: in this step we select which of B. F. Hobbs et al. (eds.), Energy Decisions and the Environment © Kluwer Academic Publishers 2000
ENERGY DECISIONS AND THE ENVIRONMENT
16
2. 3.
these will be used in an application, and precisely how they should be defined. Definition of the alternatives to be analyzed. Quantification of the levels Aij of the i attributes estimated for each of the j alternatives. This generally requires the application of a model to predict the impacts. Uncertainty and risk in attribute levels are quantified at this time.
Tradeoff analysis steps:
4. 5.
Construction of tradeoff curves. Dominance analysis.
Evaluation steps:
6. 7.
Preliminary screening of alternatives. Scaling of attributes, in which the level of an attribute is translated into a measure of value, vlAy), (also known as a single attribute value function). This sometimes involves normalization, such as using a scale of zero to one (in which the lowest value of the attribute value function is assigned zero, the highest is assigned one). 8. Selection of weights Wi for each attribute. 9. Determination and application of an amalgamation rule. Such rules combine the weights and value functions into a single overall value or ranking of the available options. Alternatively, a rule can merely reduce the number of options for further consideration to a smaller number of candidates. 10. Resolution of differences between methods, and between and among stakeholders. While the steps appear conceptually straightforward, in general most applications give relatively little attention to a number of theoretical requirements for the results to be valid. Equally important, even though we have presented these elements in the form of a sequential list, iterations among these steps will be necessary -- as we shall see in the subsequent discussions. The relationships between these steps are indicated in Figure 2.1. Alth
Data Loading...
