Finding Life Cycle Assessment Research Direction with the Aid of Meta-Analysis


  • Alessandra Zamagni,

  • Paolo Masoni,

  • Patrizia Buttol,

  • Andrea Raggi,

  • Roberto Buonamici

Alessandra Zamagni, ENEA, Via Martiri di Monte Sole 4 40129 Bologna, Italy. E-mail:


Rapid growth in life cycle assessment (LCA) methodological developments has generated a large body of work that may appear to lack direction. In this article we developed and applied a structured approach, inspired by the meta-analysis concept, to examine literature and identify research thrusts on how to further develop LCA. The procedure consists of four steps: (1) definition of the research question; (2) carrying out a literature review concerning more than 280 articles, selected from about 2,000 articles according to predefined criteria, which resulted in the identification of some 60 main methodological topics; (3) research gap analysis, in which the methodological topics identified in the previous step were compared with the research priorities identified through a users’ needs survey; and (4) interpretation of results, in which the results of both the previous steps were evaluated and organized into coherent research thrusts.

Overall the analysis delivered two main research thrusts: one devoted to increase the practicability of LCA, the other to increase model fidelity. The former is aimed at making knowledge available in an easily usable way, while the latter focuses on better describing the complexity of the systems analyzed and those interrelations that are really meaningful. Specific research topics were identified for each thrust, which suggests that sophistication and practicability can and should coexist in the same method.


In recent years there has been an increasing awareness of the importance of life cycle thinking (LCT) (Fava et al. 2009) and, more specifically, of the standardized life cycle assessment (LCA) method as a way to face the challenges posed by sustainability questions (Huppes and Ishikawa 2009). This awareness is turning into a really striking effect at the European Union policy level: LCT and LCA are central themes of the recent Sustainable Consumption and Production (SCP) Action Plan (CEC 2008), as well as in the Eco-Design Directive (EC 2009), the Waste Framework Directive (EC 2006), and the Environmental Technologies Action Plan (CEC 2004). Moreover, in the Integrated Product Policy Communication, the European Commission states that “LCAs provide the best framework for assessing the potential environmental impacts of products currently available” (CEC 2003). The recently published International Life Cycle Data System (ILCD) Handbook (JRC-IES 2010), made available through the European Platform on LCA, is a further confirmation of the importance of LCA as a decision-supporting tool in contexts ranging from product development to policy making. In fact, the handbook,1 as a series of technical guidance documents to the International Organization for Standardization (ISO) 14040 and 14044 standards (ISO 2006a, 2006b), will serve as a basis for comparable and reliable LCA applications in business and public decision making.

This increasing interest in LCT has been driven also by the considerable methodological development of LCA in the last few years, as testified by the scientific articles published so far.2 Indeed, most of the developments have occurred in the last ten years, which was referred to by Guinée and colleagues (2011) as “the decade of elaboration.” Regarding the impact assessment phase, new impact categories, indicators, and characterization factors have been developed. Moreover, an increased sophistication of the existing methods of characterization has been also proposed (e.g., with the introduction of spatial and temporal differentiation). In the realm of the inventory analysis, the need to assess policies, technological and structural choices characterized by scarce reversibility, long-term effects, trade-offs between environment, economy, and society, and so on, has been driving an expansion of the scope of LCA, both in terms of the level of analysis (from products to systems, from micro to macro) and coverage of indicators (including also economic and social aspects). Consequential LCA, hybrid approaches combining LCA and input-output analysis (IOA), scenario modeling, and new efforts for developing social and economic assessment methods are only some examples of the developments LCA methodology has been going through. Overall, it may appear that LCA lacks direction on how to further develop.

Starting from the considerations above, we questioned whether it would be possible to structure the present developments and organize unexplored areas, questions, and research priorities to further develop LCA. For this purpose, the existing literature on LCA was examined using an approach that was inspired by the meta-analysis concept, but differs in a number of ways, which are discussed below.

Widely used in social and medical sciences, meta-analysis is usually a quantitative statistical technique aimed at combining the results of several studies in order to test the pooled data for statistical significance. Even when used as a component of a systematic review procedure, its focus is on the use of statistical methods as a way to combine evidence.

As far as the scope of this article is concerned, it is clear that the classic meta-analysis approach cannot be adopted, because the object of study is represented by methodological developments and not by analytical and quantitative studies whose results can be coded by means of statistical parameters. However, the rationale behind the approach is fully applicable, because meta-analysis is meant as the analyses “of a large collection of results from individual studies for the purpose of integrating the findings” (Glass 1976, 3). For this reason, inspired by the meta-analysis concept, we developed a structured approach to examine the existing literature on LCA. The approach is characterized by a shared subjectivity in review, rather than a true objectivity, which can never be assured when qualitative elements are dominant. In fact, the decisions made are public and transparent and open to criticism from other researchers.

First, the article illustrates the approach, starting with an analysis of the existing literature on LCA. As an intermediate step, the main methodological aspects the scientific community is working on are defined, together with the research needs to fully develop them. Then, research gaps are identified with the support of a survey on users’ needs on how to further develop LCA. Finally, the results are discussed, organized in two research thrusts, and, in the last section, the main conclusions are drawn.

A Structured Approach to Examining Life Cycle Assessment Literature

When research results accumulate, it becomes increasingly difficult to understand the direction they point out and whether an unambiguous conclusion can be reached. Under those circumstances, a feeling of confusion and frustration might arise (Rudner et al. 2002) because consensus cannot be found in the literature,3 resulting in ambiguity and impossibility in making knowledge available to final users (policy makers, consumers, technicians) in a fruitful way.

Ambiguity arises whenever a contradiction exists on topics/studies on the same (and sometimes controversial) subject. This is quite common in the scientific literature on methodological developments that should explore, by definition, all the maze of knowledge and also investigate options that seem inconclusive or even unfeasible at first glance. However, researchers do not always aim to validate previous findings or to examine them critically. Some authors present approaches for solving a specific problem without duly taking previous works into consideration, and thus without any criticism of them. Moreover, it is not always true that a lack of debate on a specific issue is a symptom of consensus on that issue, or vice versa. In this way, it is not always possible to properly evaluate the consensus of the LCA scientific community on the (relative and absolute) relevance of an approach by means of a mere literature review.

We developed an approach to “extract knowledge from studies” (Glass 1976) and to investigate the methodological developments, in which the literature review is coupled with an analysis of users’ needs and with expert judgment. The purpose is to reduce the subjectivity in drawing conclusions. In fact, while every analysis requires a certain degree of subjectivity in making decisions—for example, in relation to the identification and selection of the relevant literature—in our approach choices are always explicitly stated, and therefore open to criticism.

The approach consists of the following steps:

  • • Definition of the research question
    The question raised by the authors is the following: given a situation of nonharmonized and sometimes diverging developments in the literature over the last decade (Guinée et al. 2011), how can the direction(s) in which LCA is pointing be identified? How can coherent research thrusts be identified?
  • • Carrying out the literature review
    Inclusion and exclusion criteria for the studies to be analyzed are set up. The purpose is to identify methodological approaches and related research needs on which the LCA community is working.
  • • Research gap analysis
    In this step, the methodological approaches previously identified are compared with research priorities, identified by a users’ needs survey (how users would like LCA to be developed in the future, to further support the decision-making process). The result consists of a list of priorities for research, not yet organized, however, into research thrusts.
  • • Reporting and interpretation of results
    The results of the literature review and of the users’ needs are evaluated and organized into research thrusts by means of expert judgment.

A representation of the approach is provided in figure 1, together with identification of the main elements that distinguish it from a meta-analysis.

Figure 1.

The proposed approach to examine life cycle assessment literature. The main elements that distinguish it from a meta-analysis are highlighted.

As can be seen in the right side of the figure, quantitative elements in the proposed method are missing. Compared to the classic meta-analysis, here the purpose is not to establish the presence of an effect or determine its magnitude, but to identify research priorities for further developing LCA. There are not quantitative variables to be measured, but trends to be identified, together with unexplored areas that deserve further investigative efforts. The meta-analysis concept was not used as a rigid procedure, but as a framework providing the conceptual rigor and objectivity necessary for the study. In this respect, the approach goes beyond a classic literature review in relation to two main aspects: (1) the broader-than-usual coverage of the literature review, and (2) the use of surveys on users’ needs, which make the analysis less subjective and responsive to what is happening in the real world, outside the (often considered) the so-called ivory tower of the research.

Literature Review

After defining the research question the approach starts out by performing a literature review.

The following criteria were adopted to ensure methodological soundness and that the selected studies were within the scope of the analysis:

  • • Identification of topics to be reviewed, that is, the main issues from the methodological point of view, like allocation, system boundary definition, uncertainty, and so on. This was done by interviewing LCA experts on what they considered to be the most challenging and emerging issues, and by examining the proceedings of conferences on LCA.4 Overall, about 2,000 articles were identified.
  • • Classification of the identified topics according to the phases of the LCA procedure (goal and scope definition, life cycle inventory [LCI], life cycle impact assessment [LCIA] and interpretation). Two additional separated sections were included: life cycle-based methods (life cycle costing and social life cycle assessment), and data quality and availability.
  • • Definition of criteria for screening the 2,000 references resulting from the literature search: (1) mainly officially peer-reviewed references published from 2000 to 2010 were considered, with a global geographic coverage. Review articles were included as well. (2) Case studies were reviewed only if they presented new methodological developments. This choice, which may seem too restrictive, was aimed at selecting the highest possible quality level of the studies. However, these criteria were not completely fulfilled for all approaches. In particular, grey literature was also consulted, especially for novel developments that were not yet published in scientific journals. Moreover, references before 2000 were also considered when they presented interesting approaches not taken into account in the more recent scientific literature. All in all, about 60 approaches, in more than 280 papers, with different degrees of maturity were analyzed. The complete list of references is available as supporting information on the Journal's Web site, organized according to the different topics.
  • • For each topic, identification of different approaches, that is, solutions proposed by the authors to deal with the problem. The approaches were selected trying to cover complementary authors and different schools of thought.
  • • Analysis of the approaches. An evaluation grid was developed, structured along three levels: (1) generalities, which include a description of the analyzed topic/approach, the deviations/developments with respect to ISO standards, and the relevant references; (2) analysis, with a description of the rationale (key principles of the approach, why it was implemented, which needs the authors addressed), main advantages, open questions, and practicability aspects; and (3) comments, in terms of research and development (R&D) needs and trends emphasized by the author(s).

The review of the literature was performed in the framework of the European Union's CALCAS (Co-ordination Action for Innovation in Life-Cycle Analysis for Sustainability) project, aimed at analyzing LCA approaches that have emerged during the last two decades and at indicating how the analysis might be improved by formulating research lines and roadmaps for sustainability decision support (Zamagni et al. 2008, 2009a). The work presented in this article was complemented by and benefited from the articles published in 2009, 2010, and partly in 2011, including the review articles by Bare (2010), Finnveden and colleagues (2009), and Reap and colleagues (2008a, 2008b), which the reader is invited to refer to for details.

Identification of Methodological Developments

About 10 LCA experts, selected according to their specific competencies on all the topics identified in the review, were invited to comment on a preliminary draft of the review's results. The results of this consultation, together with the outcome of the literature review, represented the basis for the identification of two main aspects: (1) the main methodological questions the scientific community has been working on, and (2) the research needs to fully develop them. These are described in table 1.

Table 1.  Methodological topics and related research lines. For each research line, a selection of references is provided. The full list is available as supporting information on the Web.
Topics Research lines
  1. Notes: CFs = characterization factors.

  2. aThe recently published environmental life cycle costing code (Swarr et al. 2011) represents a significant step forward in the development of LCC and it is expected that the number of applications and case studies will increase in coming years.

  3. b Moreno and colleagues 2011. Ontologies are (meta) data schemas, providing a controlled vocabulary of concepts, each with explicitly defined and machine processable semantics. By defining shared and common domain theories, ontologies help to communicate concisely, supporting the exchange of semantics, and not only syntax. Hence the cheap and fast construction of domain-specific ontologies is crucial for the success and proliferation of the semantic web.

  4. c Buttol and colleagues 2011. A system of data sharing within the whole supply chain would have a twofold approach: (1) to improve data availability, by allowing collecting data upstream and downstream in the supply chain with lower efforts; (2) to optimize and make the whole production system covered by the supply chain more efficient. Moreover, as several data useful for an LCA study are available in other sources and are managed by other tools (e.g., computer-aided design [CAD] systems), efforts should be spent in exploiting the synergies among the different tools.

Scenario analysisUncertainty: the knowledge generated by scenarios should be evaluated together with the associated uncertainty (Zurek and Henrichs 2007);
 Relevance, that is, when the use of scenarios is relevant in life cycle assessment (LCA) and in which situation one approach to scenario analysis is more suited than others (Börjeson et al. 2006; Rebitzer and Ekvall 2004);
 Development of consistent and generic scenarios of different types that could be used by different practitioners (Höjer et al. 2008).
Consequential LCA (CLCA) Affected process (Lund et al. 2010; Mathiesen et al. 2009; Schmidt 2008; Weidema et al. 2009)
 — identification of the affected process(es) and related uncertainty evaluation
 — collection of marginal data
  Economic models in CLCA (Earles and Halog 2011; Ekvall and Andrae 2006)
 — use of partial equilibrium models (PEMs):
 ○ when they are relevant;
 ○ data on price elasticities has to be estimated for many products: they should be included in databases posted in connection to ordinary LCI databases.
 — use of multimarket, multiregion PEMs, use of general equilibrium models
 — use of experience curves
 ○ how transfer of knowledge and experience among technologies and geographical regions should be accounted for when applying experience curves;
 ○ experience curves need to be established for more technologies. Data can be included in databases posted in connection to ordinary LCI databases.
Hybrid approaches combining input-output analysis (IOA) and LCA— Work on methodological shortcomings of IOA: approximation of the product of interest by its commodity; proportionality between market price and environmental impacts; noncompleteness of the product life cycle stages (Suh and Huppes 2005)
 — Data issues: improve the consistency among the different IO tables and build reliable and publicly available environmental intervention databases for the IO tables (activities are in progress) (Suh and Huppes 2005; Tukker et al. 2009)
 — Focus on uncertainty evaluation (uncertainty factors for IO data) (Nielsen and Weidema 2001)
 — Further explore new approaches of IOA combined with LCA for analyses with a regional resolution (Cicas et al. 2007; Yi et al. 2007)
 — Accumulate experience from successful integrated hybrid analysis (Mattila et al. 2010; Rowley et al. 2009; Suh and Nakamura 2007)
Life cycle impact assessment Improvements of existing characterization methods
 — Spatial differentiation (Finnveden and Nilsson 2005; Gallego et al. 2010; Seppälä et al. 2006[B34]; Shah and Ries 2009; Wegener Sleeswijk 2011; Wegener Sleeswijk and Heijungs 2010)
 ○ development of methodology for the definition of the optimal regions for site-dependent characterization factors (CFs)
 ○ identification of impact categories for which more detailed regional data are necessary
 ○ availability of regional emissions and environmental data to calculate regional CFs
 ○ guidelines for practitioners and manageability quality control of location/country-specific elementary flows
 — Improvement of existing approaches for the impact categories of human toxicity and ecotoxicity (Gandhi et al. 2010; Rosenbaum et al. 2008).
  Development of midpoint and damage-oriented methods in a common framework (Goedkoop et al. 2009; Itsubo and Atsushi 2007; Jolliet et al. 2004)
 — Provision of consistent and operational sets of methods and factors for LCIA
 — Uncertainty evaluation of damage function modeling
  Development of new characterization methods and new impact categories (Althaus et al. 2009a, 2009b; Hellweg et al. 2009; Milà I Canals et al. 2007a, 2007b; Núñez et al. 2010; Pfister et al. 2009; Zhang et al. 2010)
 Developments in this field are growing exponentially. Besides the categories of resource depletion, noise, land use, ionizing radiation, water use, and indoor and occupational exposure, recently the following categories have been pointed out as relevant: odor, genetic pollution (due to genetically modified organisms), nonionizing radiation (electromagnetic waves), light (e.g., from greenhouses), and thermal pollution. For these impact categories, there is no consensus on their relevance of inclusion in LCAs, on the inventory item(s) to which characterization models would attach, on the exact impact pathways, and on the impact indicators at the midpoint level. The detailed list of references available as supporting information on the Web provides all the necessary information.
(Environmental) Life cycle costing (LCC)— Conduct case studies, to increase the knowledge and the practice (Hunkeler et al. 2008; Swarr et al. 2011)a
 — Further deepen the question of a formal computational structure for LCC, as for the opportunity to strengthen the linkage with LCA (Settanni 2008)
Social life cycle assessment (S-LCA)— Further analysis of the relevance of S-LCA (which questions to be answered and which stakeholders to be addressed) (Swarr 2009; Zamagni et al. 2011)
 — Deepen the methodological questions related to the scope of the analysis, further elaborating the consequential concept (Jørgensen et al. 2009)
 — Selection and formulation of indicators (Jørgensen et al. 2008; Kim and Hur 2009)
 — Data collection and availability (Benoît et al. 2010; Ciroth and Franze 2011; Franze and Ciroth 2011)
 — Characterization of social indicators (Dreyer et al. 2010; Jørgensen et al. 2010; Reitinger et al. 2011)
Uncertainty— More investigation of correlations and interdependencies among parameters (Bojacá and Schrevens 2010)
 — Development of guidelines for supporting practitioners in the uncertainty analysis (Basson and Petrie 2007; Lloyd and Ries 2007)
 — Major effort should be spent on scenario uncertainty: it should become routine practice, due to the influence of choices on the final LCA results (Benetto et al. 2006; Huijbregts 1998; Huijbregts et al. 2003).
Data availability, quality, and management— Approaches for developing inventory models for those products whose data availability is a critical aspect (Geisler et al. 2004; Hischier et al. 2005; Wernet et al. 2009)
 — Approaches to data transposition, i.e., the use of European background data for countries outside Europe (Colodel et al. 2009)
 — Deepen the semantic web approach for LCA applications: development of interfaces between the environmental data published on the semantic web and existing LCA tools (Moreno et al. 2011)
 — Finalization of core ontology necessary for LCA publication on the web (Moreno et al. 2011)b
 — Approaches for integrating data along the whole supply chain and from different sources (Buttol et al. 2011)c

At the end of this step, a very fragmented picture, with no uniform view, is obtained. Therefore the research needs have to be further analyzed and coded according to the characteristics defined in the next step.

Research Gap Analysis

The next step consists of identifying gaps in the research topics listed in table 1, with respect to the needs and expectations of LCA users. For this purpose, the authors analyzed the needs of different groups of users, at the European level, in terms of how stakeholders want LCA to develop in the future to make it more useful in sustainability decision making.

More specifically, user needs were considered as a sort of litmus test to evaluate how much methodological developments are in line with users’ expectations and needs. As a result, priorities in research developments were established, and new and innovative proposals were suggested to address research gaps related to unsatisfied needs.

The analysis, performed within the CALCAS project (Rydberg et al. 2008, 2009), was carried out by means of a survey targeted to different stakeholder groups. The scope of the decision-making processes analyzed was quite wide, ranging from high-level strategy choices to detailed technology and product choices. The survey was conducted by means of semistructured interviews with key actors as well as questionnaires to stakeholders.

The following main groups of stakeholders were surveyed (Rydberg et al. 2008):

  • • Public authorities. Life cycle analysis is not the reference approach in this context, but they foresee a potential use of life cycle approaches in activities such as system choice (e.g., transport systems) and technology choices (e.g., waste management).
  • • Business (industry, retailers). They mainly use LCA as a decision support tool in product development (77%), selection of raw materials (66%), and choices of technologies (55%).
  • • Nongovernmental organizations (NGOs, including consumer associations). They make use of life cycle approaches for communication purposes.
  • • R&D programmers (national funding organizations and research institutes), those who will finance future LCA research.

The findings of the survey were discussed in a workshop with 37 invited experts, where the requests that had been identified in the users’ needs survey as important for future development of life cycle approaches were prioritized. Moreover, further needs and wishes about LCA development were discussed and proposed.

For both the survey and the workshop, the experts were selected according to the following criteria:

  • • good knowledge of the life cycle thinking and of the life cycle assessment method;
  • • stakeholders who have responsibility in relation to sustainability decision making; and
  • • stakeholders who work in an organization where the life cycle approach is considered to play an important role in supporting the decision-making process.

The overall results for each stakeholder category are summarized in table 2.

Table 2.  Priority research needs expressed by different stakeholders’ groups
Stakeholder group Research needs
  1. Source: Adapted from Rydberg and colleagues (2008).

Public authoritiesInclude additional economic performance in LCA, to increase the acceptance and the success of the methods
 Include social effects
 Work on making the results easy to understand
BusinessIncrease transparency in models
 Develop methods together with the industry, not by the scientists alone
 Develop strategies on how LCA can be communicated within the supply and value chains
 Make consumers understand their role in the life cycle of products
 Include future impacts in the models
 Include risk analysis in environmental assessments
 Increase validity and credibility of simplified life cycle approaches
 Develop tools tailored to specific industrial sectors
 Introduce regional variations in the models
Nongovernmental organizations (NGOs)Work on simplifying methods and making them less expensive
 Strengthen educational activities on life cycle approaches
 Increase the reliability of methods
Research and development (R&D) programmersDevelop models for sustainability evaluations, integrating economic, environmental, and social dimensions
 Conduct more research to introduce sustainability parameters in life cycle approaches for all the main mid- and long-term technological and strategic choices, such as energy and the environment, transport systems, urban policy issues, planning, and quality control in the service sector
 Develop reference systems for quantitative knowledge (data)

The comparison between the research topics of table 1, identified by means of the literature review, and the research needs of table 2, expressed by the surveyed users, shows a partial overlap. In fact, there is a leitmotiv in all the stakeholders’ categories, which is presently not fully addressed by the research community: the need for simplified LCA methods and tools. This in turn involves

  • • simpler interfaces and less time-consuming models, and
  • • methods and software tailored to specific industrial sectors.

Sustainability assessment is the other main topic pointed out, in which there is a growing interest in the scientific community (and not only). More specifically, main needs and expectations relate to

  • • integration of economic and social elements in life cycle approaches, and
  • • introduction of sustainability parameters in life cycle approaches for all the main mid- and long-term technological and strategic choices, such as energy and the environment, transport systems, urban policy issues, planning, and quality control in the service sector.

The two major aspects highlighted above are dealt with in the next section, in which the last step of the procedure is described. The purpose is to organize the results of the literature review and of the users’ needs into coherent research thrusts, within a uniform vision of where LCA is going and should go.

Reporting and Interpretation of Results: Definition of Research Thrusts

The analysis described in the previous paragraphs confirms that two are the main research thrusts, which take the plurality of needs and the vision of the scientific community into consideration: one devoted to an increased practicability and another to sophistication aimed at increasing model fidelity.

In these terms such results are not surprising and can be considered even trivial.5 However, not so trivial is the identification of the specific research topics that can contribute to increase practicability and model fidelity, also considering that these two different requirements could seem conflicting.

The results and considerations are based on the judgment of LCA experts. This step could be affected by higher subjectivity than the previous ones. However, to limit it and to avoid results that are biased by the personal interests of the authors, LCA experts were selected who were not involved in the development of specific approaches.

Increasing Practicability

Practicable is an adjective that refers to the capability of being put into effect or being used. Thus it involves the capability of making knowledge available in an easily usable way and one that is economically viable. In the LCA context, this can be translated in terms of making available user-friendly, low resource-demanding tools (methods and software), reliable background and foreground data to operate them, and guidance about how to use them in the variety of applications.

To reach these objectives, research activities should be developed along two main lines:

  • • simplification, and
  • • the development of integrated approaches (i.e., methods, tools, databases, guidelines, and training) supporting the application of LCA.

Simplification is a crucial question for an extensive use of life cycle information, especially for small- and medium-size enterprises that rarely have the knowledge and resources necessary to implement LCA. Facilitating access to reliable, accurate, and relevant life cycle information means reducing the costs of the process of a product's environmental improvement and better communication along the supply chain. Users mainly pose the question of simplification in terms of simpler interfaces and less time-consuming models. However, the topic deserves attention also from the methodological point of view, considering that, at present, comprehensive approaches to simplification have not yet been developed.

Simplification can be achieved in three main ways: (1) reducing the number of environmental impact indicators, (2) intervening at the level of methodological choices in the inventory phase, and (3) working on data availability.

The first approach is the most applied, as testified by the widespread use of the carbon footprint and by the ongoing development of the water footprint. The use of such a reduced set of indicators increases communicability to final users, and offers an answer to the growing concern over climate change and water scarcity raised by several stakeholders. However, the risk of bias exists, as the environmental implications of the assessment could be misleading, and the basic principle of LCA—its comprehensiveness that prevents burdens shifting—could be lost (SETAC Europe LCA Steering Committee 2008).

Possible strategies at the inventory step are the exclusion of phases of the life cycle and a reduction in the number of elementary flows. These choices should be made very carefully and cannot be generalized because of the risk of reducing reliability and/or completeness of the assessment.

Finally, development of reliable and comprehensive databases, which started almost in parallel with the first LCA applications, facilitates the work of practitioners and allows them to maintain the completeness of the assessment, and to preserve the life cycle principles. Some aspects deserve particular attention for future and ongoing research initiatives:

  • • Enlargement of the number of products covered, and activities aimed at a global consistency of databases. The latter aspect is addressed in the “Global guidance principles for life cycle assessment databases” (Sonnemann and Vigon 2011). The document is expected to provide global guidance on LCI data for widespread use, thus contributing to increasing the credibility and accessibility of existing LCA data (Sonnemann et al. 2011).
  • • Further investigation of analytical models to produce LCIs of products. There are some sectors (e.g., chemicals) in which, due to the vast number of chemicals in production and the problem of data confidentiality, a detailed inventory analysis is difficult, if not unfeasible. Different studies have been produced, aimed at developing methods for generating generalized data, such as parameterized inventories (Mueller et al. 2004), LCIs for groups of materials (Rydh and Sun 2005), LCIs of chemicals (Geisler et al. 2004; Hischier et al. 2005; Wernet et al. 2009), but agreed solutions are still lacking.
  • • Investigation of approaches to data transposition, that is, use of European background data for countries outside Europe. Although, in the long term, national LCI databases will become available, in the meantime it is necessary to provide practitioners with an intermediate solution to properly take regional issues into account. First approaches under development (Colodel et al. 2009) should be further investigated and tested in real case studies.

Not only does the issue of practicability have a methodological connotation, it also refers to the capability to set up an integrated approach supporting the application of LCA and overcoming or reducing obstacles at the application level, such as costs, users’ knowledge, and the availability of tools (software and databases). Such a system should be aimed at optimizing the use of resources for the production of LCI data and specialized databases, and at offering training and guidance for the application of LCA (Buttol et al. 2011).

Increasing Model Fidelity

Fidelity refers to the degree to which a model is able to represent the reality described so that it is able to capture the complexity and those interrelations within the system that are really meaningful. This concept is fundamental for all applications, and becomes crucial for those with a broad scope and/or object of analysis, like mid- and long-term technological and strategic choices on energy, the environment, urban policy issues, and so on. In these cases, interactions and feedback in the system are significant, and generally perturbations introduced by the functional unit cannot be neglected. Because classic LCA is a linear steady-state model with a focus on environmental aspects, it does not consider the complexities related to mechanisms and relations besides the mere technological and environmental ones (Heijungs et al. 2010). However, researchers have started investigating the feasibility of a more sophisticated LCA, and the literature review shows that methodological developments are going on in this direction.

Based on these considerations, two main research topics are considered relevant:

  • • strengthening the integration among LCA, life cycle costing (LCC), and social life cycle assessment (S-LCA); and
  • • broadening and deepening the analysis, addressing the problem at different scales of application and effects (from micro to macro).

A graphical representation of these two research topics is provided in figure 2.

Figure 2.

Representation of the research topics under the research thrusts “increasing model fidelity.” Adapted from Zamagni and colleagues (2009b).

Performing sustainability evaluations to accumulate knowledge about the interrelations among environmental, economic, and social effects is a clear need expressed by users within public authorities.

The concepts of sustainability and sustainable development are very controversial and are disputed at both the scientific and social level. In fact, sustainability is a multidimensional concept that involves different areas (economic, environmental, and social), normative positions, and empirical knowledge. It covers more aspects than LCA. Thus, in order to move from LCA to a life cycle-based analysis for sustainability, it is necessary, first, to broaden the scope by including economic and social dimensions. This is the present state of the art in LCA, in which life cycle-based sustainability evaluations are performed according to the formula expressed by Kloepffer (2008):


In figure 2, this is represented by the horizontal axis, labeled “broadening the scope of indicators.” The three methods, which have a different degree of development, are applied at the product level, independently, one from another, under specific consistency requirements, but without considering the mutual relations that can arise.

As a consequence, synergies, linkages, and side effects across different dimensions and across the boundaries of systems are not lost while they represent the core of any sustainability problems (Graedel and van der Voet 2010). Thus research should focus on accumulating knowledge about the interrelations among environmental, economic, and social effects, answering the question of how to deal with the trade-offs among them. As depicted in figure 2, two main components are necessary to describe such complexities (Guinée et al. 2011):

  • • Deepening. This can be achieved by adding further sophistication to the modeling, for example, by adopting spatially differentiated models, and/or by including more relations in the analysis. In the figure this is represented by the arrow that crosses the three layers. An example of deepening is given by the consequential approach. In fact, the rationale behind it requires thinking about the consequences of the actions—the interrelations—and thus projecting the problem at the market level, with all its dynamics. Thus consequential LCA represents a way to further add sophistication to LCI modeling by giving the opportunity to introduce market mechanisms into LCA. Another example is represented by the use of hybrid approaches combining IOA and LCA. In fact, they share modeling and data structure, even with different routes of data supply. With their flexible computational structure, they potentially open new perspectives not only towards adding realism by including a more complete system, but also towards expanding the scope of LCA applications to higher scales of analysis, from “micro questions on specific products, to meso questions on life styles up to macro questions in which the societal structure is part of the analysis” (Heijungs et al. 2010, 422).
  • • Broadening. Besides extending the number of environmental indicators or also including the economic and social ones in the analysis in an integrated way, broadening is achieved by shifting the analysis from individual product systems to sectors, baskets of commodities, markets, or whole economies (from the micro to meso and macro levels). In figure 2, it is represented by the vertical axis “broadening the object of the analysis.”

An approach to this type of analysis has been developed by Guinée and colleagues (2011), leading to the proposal of a life cycle sustainability analysis, a transdisciplinary integration framework that works with several models. The framework, drafted at a methodological level, but not fully developed, requires several research efforts, as identified in the article by Guinée and colleagues (2009).

Discussion and Conclusion

Rapid growth in LCA methodological developments has generated a large body of work that may appear to lack direction. We developed and applied a structured approach, inspired by the meta-analysis concept, to examine the literature and to derive research thrusts for LCA in the future. The main characteristic of this approach consists of coupling a literature review with an analysis of users’ needs and with expert judgment. In this respect, the approach goes beyond a classic literature review because it reduces the subjectivity of a judgment based entirely on the opinions of the articles’ authors.

Overall the analysis delivered two main research thrusts: one devoted to increase the practicability of LCA and the other to increase model fidelity.

Practicability is the basic requirement to support real-world decisions in business and public policy making. The main topics to be dealt with under this umbrella are those that were at the core of the debate in the late 1990s. Data quality and availability at reduced costs, simplified tools tailored to users’ requirements, and approaches to uncertainty analysis, just to mention a few, are nowadays still the object of further development. In this regard, the harmonization work of the ILCD Handbook, coordinated by the European Platform on LCA, supports the defining of methodology recommendations for LCA use in business and public policy contexts. Further steps should be devoted to making the ILCD Handbook fully applicable, with detailed provisions on how to solve the main methodological problems for a broad range of industrial applications.

Meanwhile research should proceed, and the question of sophistication and complexities should be considered. There are plenty of examples in the literature of interesting approaches. Scientists have started going back to LCA's foundations (Zamagni et al. 2008) and have rediscovered its nature, experimented with novel approaches, and reproposed previous ones with a new appearance. Even if historically the LCA framework works with the support of other models and disciplines, as clearly demonstrated in the impact assessment phase, it seems that only in the last decade have scientists become fully aware of that. This attitude in the past could have been caused by the concern that the contribution of other disciplines to further developing LCA methodology violated its inherent principles as defined in the ISO standards. As LCA was highly criticized in the past and, only with standardization did it regain its reputation, there would be the fear that any major change could again endanger its credibility.

Only recently have scientists started using LCA in combination/integration with economic models, ecological models, and social theories (e.g., Joshi 1999; Venkatesh et al. 2009) to make the methodology more complete, better model the system, reduce uncertainty, collect more representative data, define scenarios, and include other mechanisms than the environmental and technological ones commonly considered in LCA (Heijungs et al. 2010).

These main findings on the evolution of LCA show an apparently contradictory and diverging situation because, on the one side, further research is suggested towards increased model fidelity, and on the other side, a focus on increasing usability is suggested. However, we believe that these two different requirements are not conflicting, but they represent two sides of the same coin. In fact, a correct and scientifically sound development of the simplification techniques requires a complete understanding of the complexities of the problems as a preliminary condition to decide where and how to introduce simplifications.

Consequently the future challenge of research is to work on complexity, taking advantage of the contributions from other disciplines and making knowledge available with tolerable uncertainty. If such an expanded LCA can still be considered LCA is a question we open for debate in the research community.


This article is partially based on work done for the CALCAS (Co-ordination Action for Innovation in Life Cycle Analysis for Sustainability) project, which was funded by the European Union as part of the 6th Framework Programme (Project no. 037075; see The CALCAS partners are gratefully acknowledged for their contribution. Furthermore, three anonymous reviewers are gratefully acknowledged for their useful suggestions.

Moreover, we came to the conclusions by applying a transparent procedure, based on an extensive review of the literature over the last ten years, coupled with an analysis of users’ needs. Even if the final result is a confirmation of the impressions the LCA community has been aware of for quite some time, this article is aimed at providing the supporting evidence for this, opening the door to the criticisms of other researchers.


  • 1

    The Handbook includes explicit and goal-specific methodological recommendations, a multilanguage terminology, a nomenclature, a detailed verification/review frame, and further supporting documents and tools.

  • 2

    Research done in the abstract and citation database SCOPUS using the keyword “life cycle assessment” delivered 745 articles published in 2010, compared with 416 in 2007 and 188 in 2000.

  • 3

    A curious example can be mentioned that shows why ambiguity in science is so frustrating for decision makers. U.S. Senator Walter Mondale expressed his concerns to the American Psychological Association in 1970: “For every study, statistical or theoretical, that contains a proposed solution or recommendation, there is always another equally well-documented study, challenging the assumptions or conclusions of the first. No one seems to agree with anyone else's approach. But more distressing: no one seems to know what works” (Rudner et al. 2002, 4).

  • 4

    Conferences are the place in which new and novel approaches are presented, aimed at advancing the state of the art. For this reason they were considered an important and reliable source of information for a first selection of the methodological topics.

  • 5

    However, what we find surprising is that the topic of practicability, despite being addressed since the origins of LCA, is still one of the main shortcomings that hampers a broad application of the method, especially in the industrial context.

About the Authors

Alessandra Zamagni, a PhD candidate at University “G. d’Annunzio,” Pescara, Italy, is a researcher at the Italian National Agency for New Technology, Energy and Sustainable Economic Development (ENEA), Bologna, Italy. Patrizia Buttol and Roberto Buonamici are, respectively, first researcher and research director at ENEA, Bologna, Italy. Paolo Masoni, is a research director, and leads the LCA and Ecodesign Laboratory at ENEA, Bologna, Italy. Andrea Raggi is full professor in industrial ecology at the University “G. d’Annunzio,” Pescara, Italy.