Page 113 - Materials Chemistry, Second Edition
P. 113

98                                                      A. Bjørn et al.

            8.5.4  Recommended Modelling Choices for the Identified
                   Decision Context


            ILCD provides recommendations for model choices for each of decision contexts
            (Situation A, B, C1 and C2) identified as part of the goal definition (see Chap. 7).
            These recommendations are the outcome of a comprehensive consultation process
            within the LCA community. Since different actors with different views have had a
            saying in this consensus process leading up to the ILCD recommendations, they are
            somewhat internally inconsistent, as pointed out by, for example, Ekvall et al.
            (2016). Below we present the recommendations for each decision context and make
            notes about the parts that are disputed. Table 8.4 summarises the recommendations.

            Situation A
            Situation A concerns micro-level decision support (see Chap. 7) and the conse-
            quence of a decision (e.g. the introduction of a new product on the market) is
            therefore of interest. Ideally, the marginal process should therefore be identified and
            used for all background processes (such as electricity supply) and cases of multi-
            functionality (e.g. of an incineration process) should be handled by system
            expansion with marginal processes, provided that subdivision is not possible (see
            Sect. 8.5.2). This ideal for Situation A is logically consistent with a consequential
            modelling framework. Yet, ILCD recommends using an average market con-
            sumption mix for background processes and in cases of system expansion in the
            background system. ILCD terms this attributional modelling, although system
            expansion was previously associated with consequential modelling, as mentioned
            above. The main reason for diverging from the ideal is that for the small changes
            studied under Situation A it can be very difficult to identify marginal processes, i.e.
            to understand the long- and short-term consequences on the market of introducing a
            small change in its composition of product systems. The actual market behaviour in
            response to small changes may also not be well-represented by simple mathematical

            Table 8.4 Summary of ILCD recommendations on LCI modelling choices
            Decision      LCI          Handling of       Modelling of background
            context       modelling    multifunctional   system
                          framework    processes when
                          (ILCD        subdivision is not
                          terminology)  possible
            Situation A   Attributional  System expansion  Average processes
            Situation B   Mix of       System expansion  Mix of long-term marginal
                          attributional                  processes for processes
                          and                            structurally changed.
                          consequential                  Average processes in all
                                                         other cases
            Situation C1  Attributional  System expansion  Average processes
            Situation C2  Attributional  Allocation      Average processes
   108   109   110   111   112   113   114   115   116   117   118