Life cycle analysis methodology incorporating private and social costs

A key question for life cycle analysis has been the translation of the inventory phase of the analysis into results usable in a business environment to make product decisions. The methods routinely used to translate the many dimensions of a life cycle inventory into a single ''score'' or measure of performance have been regularly challenged on a wide range of bases. However, these criticisms fail to recognize the inherent normative nature of the exercise, where the only relevant criterion for success is the ability make a discrete choice from a limited set of alternatives, rather than identifying the ''best'' possible action. The need to discriminate among alternatives, rather than to optimize globally, should allow the analyst to approach the problem of life cycle analysis from a more practical perspective. This premise is tested through the application of a proposed methodology that provides a set of broad ranges of value, in $/kg of every emission, based on aggregate willingness to pay to avoid the environmental impacts of each pollutant. The ranges include scientific uncertainty, variation in context or location, and a large range of possible values for parameters that have a subjective component. The $/kg ranges can be applied to the life cycle inventories of products to produce measures of overall environmental impact. Often, these ranges are sufficient to determine which is the best of several product or process alternatives. Finally, the methodology provides a quick guide to which pollutants and parameters are the most important, which makes conclusions more transparent, and it helps focuses further analysis for decision making, if no conclusion emerges initially. This kind of methodology can be very useful if the $/kg estimates are reliable. The first half of this paper develops how the ranges were derived. Any potential user needs to understand the notions of value employed; which effects are included and which are not, what the assumptions are, and how the answers can be modified. The second half of the paper demonstrates the application of the methodology to an automotive fender case study.