The choice of system boundaries and allocation methods can have decisive effects on the results and conclusions of a life cycle assessment (LCA). System expansion makes it possible to model the indirect effects of a decision; however, this modelling is often based on inaccurate assumptions. Subdivision and/or allocation based on physical, causal relationships model the consequences of a decision that affects the internally used functions but not the exported functions of a multi-function process. Allocation based on gross sales value model the causes of environmental burdens. System expansion and marginal data can be used in most LCA applications. It can be expected to contribute to individual decisions and actions that result in a lower level of environmental burdens per functional unit than would have been the case without the LCA; however, the modelling of indirect and marginal effects may be restricted by other methodological requirements, such as acceptability, feasibility, and, in some applications, the need for a detailed methodological standard. This thesis includes a new approach to the allocation problem in open-loop recycling. This approach models the indirect effects of a change in the supply of, or demand for, the recycled material. It can be used for system expansion as well as for allocation. It takes important mechanisms into account, but the precision in the model can be poor. The 50/50 allocation method presented in an earlier paper can be regarded as an approximation of the new approach. Important methodological issues in the environmental comparison between recycling and waste incineration of wastepaper, old corrugated board, etc. include the modelling of indirect effects and the choice of data for electricity production. The indirect effects depend on which energy source competes with energy from wastepaper incineration, on which material is replaced by the recycled paper, and on the alternative use of forest resources that are not required for pulpwood production. All of these, as well as the marginal electricity production, depend on the time-perspective and on other waste management, forestry, and energy policies.
Chalmers University of Technology, Gothenburg, Sweden
Prof. Anne-Marie Tillman