Article Preview
Top1. Introduction
In July 2009, more than 20 years after the referendum repealing nuclear power, the Italian government proposed a bill to reintroduce nuclear power in Italy1. In addition to the objections on environmental grounds and social acceptance, this decision has raised many questions mainly for economic and financial reasons. Having abandoned the nuclear industry for such a long time, Italy is now in a weak position as regards infrastructure and has a technology gap, which could lead to expectations of cheap and easily producible nuclear power being seriously disappointed. Besides its social acceptance and the problem of handling radioactive waste, the economic appraisal of nuclear energy compared to coal and gas, its main competitors, is the most controversial item in the current debate on energy resources. Who would be prepared to invest in nuclear power in Italy? Who should fund the operation? What are the economic and financial risks involved? How long would it take to build the first reactor in Italy and how much would this cost? All questions to which answers have yet to be given.
The most prestigious international research bodies and universities have undertaken significant studies on costs and competitiveness. However, the conclusions drawn have very often been discordant and the inconsistent findings hint that estimating costs of nuclear energy is a task beset with uncertainty. The lack of consistent values applied to the various parameters (overnight cost, O&M cost, lifetime of plant, etc) means that the main problem is choosing which of these to use for the parameters employed in the appraisal of energy technologies. How can this uncertainty be handled in real terms? The most important papers in the literature that deal with the uncertainty of energy production costs are based on stochastic processes. A probability model was devised by Feretic and Tomsic (2005) and De Paoli and Gullì (2008) in order to overcome this ambiguity in part using the Monte Carlo analysis method. Others (Gollier et al., 2005; Naito et al., 2010; Takizawa & Suzuki, 2004) have contributed interesting ideas regarding appraisal under conditions of uncertainty.
This paper sets out another approach for dealing with uncertainty of costs based on fuzzy logic. The advantage of using fuzzy sets lies in its ability to handle uncertainty more easily and more transparently than traditional tools. Fuzzy sets capture the notion of “possibility” which is a wider concept than that of “probability”. A fuzzy-set is an estimate of an uncertain parameter with a value that may vary within a possible range while a probabilistic estimate implies forecasting a probable value. In our case, the use of fuzzy-sets thereby allowed us to represent the often divergent values of each parameter using a data interval (min-max).
Bellmann and Zadeh (1970) were the first researchers to use fuzzy sets in decision-making processes and launched the development of a vast body of literature in this field. A variety of studies have produced extremely interesting results and have contributed substantially to the advancement of knowledge on multi-criteria decision making (MCDM) processes notably: Aouam et al, (2003); Carlsson, (1982); Cheng, (1999); Fan et al., (2004); Fan et al., (2002); Li, (1999); Noci and Toletti, (2000); Roubens, (1997); Wang & Parkan (2005). Applications on solving MCDM problems using fuzzy sets theory have been published in professional journals covering a diversity of disciplines, such as transfer strategy selection in biotechnology (Chang & Chen, 1994), the automotive industry (Altrock & Krause, 1994), tool steel material selection (Chen, 1997), nature resource management (Liu & Stewart, 2004), assessment of climate change (Bell et al., 2003), selection of renewable energy alternatives (Kahraman et al., 2009), sustainable fishing development strategies evaluation (Chiou et al., 2005), site selection (Onut & Soner, 2007), accident management strategies (Jae & Moon 2002), robot selection (Chu & Lin, 2003)