A survey of unresolved problems in life cycle assessment

Life cycle assessment (LCA) stands as the pre-eminent tool for estimating environmental effects caused by products and processes from ‘cradle to grave’ or ‘cradle to cradle.’ It exists in multiple forms, claims a growing list of practitioners, and remains a focus of continuing research. Despite its popularity and codification by organizations such as the International Organization for Standards and the Society of Environmental Toxicology and Chemistry, life cycle assessment is a tool in need of improvement. Multiple authors have written about its individual problems, but a unified treatment of the subject is lacking. The following literature survey gathers and explains issues, problems and problematic decisions currently limiting LCA’s goal and scope definition and life cycle inventory phases.

The review identifies 15 major problem areas and organizes them by the LCA phases in which each appears. This part of the review focuses on the first 7 of these problems occurring during the goal and scope definition and life cycle inventory phases. It is meant as a concise summary for practitioners interested in methodological limitations which might degrade the accuracy of their assessments. For new researchers, it provides an overview of pertinent problem areas toward which they might wish to direct their research efforts.

Multiple problems occur in each of LCA’s four phases and reduce the accuracy of this tool. Considering problem severity and the adequacy of current solutions, six of the 15 discussed problems are of paramount importance. In LCA’s first two phases, functional unit definition, boundary selection, and allocation are critical problems requiring particular attention.

Problems encountered during goal and scope definition arise from decisions about inclusion and exclusion while those in inventory analysis involve flows and transformations. Foundational decisions about the basis of comparison (functional unit), bounds of the study, and physical relationships between included processes largely dictate the representativeness and, therefore, the value of an LCA. It is for this reason that problems in functional unit definition, boundary selection, and allocation are the most critical examined in the first part of this review.