2015 年 11 巻 3 号 p. 278-291
Background and Objective. Though weighting in life cycle impact assessment (LCIA) is widely used in practice, it is still a controversial issue. There are many kinds of calculation methods of weights such as distance-to-target (DtT), panel, and monetization methods and the choice of the calculation method has an effect on the decision-making. From another aspect, weighting methods can be classified into midpoint and endpoint approaches. There exist previous studies that compare different LCIA methods including weighting phases. However no studies have compared weighting methods themselves considering the difference in frameworks of LCIA methods. The purpose of this study is, by calculating multiple sets of weighting factors using different methods under the identical LCIA framework, to investigate how differences in calculation methods and viewpoints among weighting methods have influence on midpoint and endpoint weighting factors.
Methods. Weighting factors were calculated for midpoint and endpoint impact categories using the panel method as well as the DtT method under the framework of LIME2. The questionnaire surveys for panel methods were conducted for respondent groups who have different viewpoints, such as LCA specialists and practitioners, and weighting factors based on two different approaches, “paired comparison” and “classification of importance”, were calculated by two styles of questionnaires. By comparing the results, differences between respondent groups, between panel and DtT methods, and between different approaches of the panel method were discussed.
Results and Discussion. The variability of calculated weighting factors among specialists was larger than the LCA practitioners. Compared with the weighting factors based on the DtT methods, there was less difference among the weights of impact categories calculated based on panel methods. The choice of the approach of the panel method had an influence on the calculated weight of impact categories. For example, noise, as one of the midpoint impact categories of LIME2, was evaluated to be more important by paired comparison than classification of importance.
Conclusions. These results suggest that weighting factors can be affected by various factors such as the choice of calculation method of weights. It was found that even weighting factors of a particular respondent were variable depending on the approach of the panel method. When practicing LCIA, interpreting or using the results of LCIA, it is important to understand the underlying concepts and viewpoints of weighting methods behind the variability of weighting factors.