Operations Research For Sustainability PDF

European Journal of Operational Research 274 (2019) 1–21
Contents lists available at ScienceDirect
European Journal of Operational Research

journal homepage: www.elsevier.com/locate/ejor
Invited Review
Operations research for sustainability assessment of products: A

review
Christian Thies a,∗, Karsten Kieckhäfer a, Thomas S. Spengler a,b, Manbir S. Sodhi b
a
Institute of Automotive Management and Industrial Production, Technische Universität Braunschweig, Mühlenpfordtstr. 23, 38106 Braunschweig, Germany
b
Department of Mechanical, Industrial and Systems Engineering, University of Rhode Island, 103 Gilbreth Hall, Kingston, RI 02881, USA
a r t i c l e i n f o a b s t r a c t
Article history: The environmental and social impacts of products are being increasingly scrutinized. This necessitates
Received 4 April 2017 systematic assessment methods. Life Cycle Sustainability Assessment (LCSA) provides a framework to
Accepted 23 April 2018
addresses diverse sustainability issues over the product’s life cycle, but its application is complicated.
Available online 30 April 2018
Major challenges, such as the selection of relevant indicators, multi-criteria comparisons of alternatives,
Keywords: the treatment of uncertainties, or the integration of spatially differentiated data, can be facilitated by
OR in environment and climate change adopting advanced analytical methods from Operations Research. This paper reviews 142 articles that use
Product sustainability Operations Research methods for product-related sustainability assessments. The articles were selected
Multiple criteria analysis from peer-reviewed scientific literature in a systematic search and screening process. Descriptive analysis
Data envelopment analysis shows that related publication output is growing over time and originates mainly in journals related to
Literature review Environmental Management. While ecological indicators are considered by most articles, the integration
of economic and social indicators is emerging. Focusing on the contributions of Operations Research, a
detailed analysis shows that more than half of the articles adopt methods from Multi-Attribute Decision
Making (MADM), followed by Data Envelopment Analysis (DEA) and Multi-Objective Decision Making
(MODM). Uncertainties with regard to inventory data and decision makers’ preferences are addressed us-
ing fuzzy logic, stochastic models, or sensitivity analysis. The use of spatially differentiated data is not
frequently found in the reviewed articles. Research needs derived from this analysis comprise the inte-
gration of qualitative and semi-quantitative (social) indicators, the simultaneous consideration of global
and local sustainability objectives, and the application of systematic procedures to address uncertainty.
© 2018 Elsevier B.V. All rights reserved.
1. Introduction eration. Consequently, much attention is directed on methods to

trace and estimate the sustainability of products, capturing their
Sustainability assessment refers to the systematic compilation economic, environmental, and social impacts.
and evaluation of environmental, economic, and social impacts Systematic approaches to assess product sustainability are
related to a system. It aims at providing information that sup- based on two key principles: life cycle thinking and the triple
ports improving the system or comparing it to other systems. bottom line concept. Life cycle thinking requires assessing the
Sustainability assessments have been carried out for different impacts over the entire life cycle of a product “from cradle to
systems including countries, industries, technologies, companies, grave”. The triple bottom line concept stresses the importance of
and products (Munda, 2016; Ness, Urbel-Piirsalu, Anderberg, & simultaneously taking into account ecological, economic, and so-
Olsson, 2007; Zamagni et al., 2009). This article focuses on the cial sustainability impacts. Both principles are integrated into the
assessment of product-oriented systems (comprising goods and comprehensive framework of Life Cycle Sustainability Assessment
services), a branch of sustainability assessment that is receiving (LCSA) (Kloepffer, 2008; UNEP/SETAC, 2011).
increasing attention because the function, quality, availability, and LCSA builds on established methods like (environmental) Life
aesthetics of products are no longer the only determinants for Cycle Assessment (LCA), (economic) Life Cycle Costing (LCC), and
their value. Instead, their impact on the environment and the Social Life Cycle Assessment (SLCA). The procedure of the assess-
social consequences of their production are also taken into consid- ment is derived from the ISO 14040/14044 standards (Fig. 1). A
typical LCSA study starts with the definition of the goal and the
scope. Next, inventory data describing the exchanges between the
∗
Corresponding author. unit processes in the product’s life cycle and their executing orga-
E-mail address: ch.thies@tu-braunschweig.de (C. Thies). nizations as well as the external environment is collected. In the
https://doi.org/10.1016/j.ejor.2018.04.039
0377-2217/© 2018 Elsevier B.V. All rights reserved.
2 C. Thies et al. / European Journal of Operational Research 274 (2019) 1–21
(2009) discuss the methodological developments in LCA, and

Gaussin et al. (2013) survey existing approaches for assessing the
environmental footprint of manufactured products. Approaches
to monetary valuation in LCA are discussed by Pizzol et al.
(2015), and applications of LCC are reviewed by Korpi and Ala-
Risku (2008). With regard to the social dimension, Jørgensen, Le
Bocq, Nazarkina, and Hauschild (2008) give an early overview of
methodologies used in SLCA, and the more recent review by Petti
et al. (2016) investigates applications of SLCA.
An early discussion on the application of Operations Research
models in the context of Environmental Management is given in
Bloemhof-Ruwaard, van Beek, Hordijk, and van Wassenhove (1995).
Approaches in Management Sciences and Operations Research to
air pollution management with a focus on mathematical program-
Fig. 1. Framework for Life Cycle Sustainability Assessment (adapted from ISO ming are surveyed in Cooper et al. (1997). The state-of-the-art
14040:2006). in sustainability analysis methodologies for efficient decision sup-
port has been summarized by Liu, Leat, and Smith (2011). The
use of multi-criteria decision analysis in environmental sciences
subsequent impact assessment, the inventory data is assigned to is reviewed by Huang, Keisler, and Linkov (2011) and Cegan et al.
ecological, economic, and social impact categories. Finally, an eval- (2017). A review by Zanghelini et al. (2018) discusses how multi-
uation step integrates the various indicators from all three sustain- criteria decision analysis can aid results interpretation in LCA.
ability dimensions to support decision making, taking into account Recent reviews on Operations Research and Sustainable Supply
the decision makers’ preferences and the underlying uncertainty Chain Management have been published by Brandenburg, Govin-
(Grubert, 2017; Keller, 2015). All steps are accompanied by inter- dan, Sarkis, and Seuring (2014), Seuring (2013), Tang and Zhou
pretation and may be carried out in an iterative procedure as new (2012), and Barbosa-Póvoa, da Silva, and Carvalho (2018). While
insights, data limitations, or stakeholder views can lead to a redef- these papers focus on the integration of sustainability aspects into
inition of the study focus, goal, and methods (UNEP/SETAC 2011). Operations and Supply Chain Management models, little attention
Although sustainability assessments of products are quite has been given to structure the literature on the use of Operations
common in industry, and standard approaches are recognized in Research models within product-related sustainability assessment,
the scientific community, major unresolved problems in these despite the manifold contributions that have been made in this
assessments have been identified (for LCA see, e.g., Reap, Roman, area.
Duncan, & Bras, 2008). First, the selection of indicators and impact The remainder of this paper is organized as follows. The
categories that are relevant for the assessment is a delicate task methodology of this literature review is described in Section 2. Ag-
that involves tradeoffs between comprehensiveness and data gregate descriptive analyses of the review database are presented
availability and collection effort. Second, the comparison of differ- in Section 3 and detailed analyses of article clusters based on
ent products, alternative product designs, or supply chains of a Operations Research methods are carried out in Section 4. From
product is often difficult because it necessitates normalization and the results of these analyses, promising approaches and research
weighting of multiple and typically incommensurable assessment needs are identified in Section 5 and conclusions are presented in
criteria. Third, the handling of uncertainty and incomplete data is Section 6.
a major issue in sustainability assessment where little guidance
is available. Fourth, spatial differentiation to account for the het- 2. Review methodology
erogeneity in technologies and the environment across different
regions is often desired but rarely implemented as it requires A systematic search procedure was applied to ensure compre-
ample data and appropriate methods to process it. hensiveness and to minimize potential bias in selecting relevant ar-
To deal with the problems discussed above and to make sus- ticles. Based on the research questions defined in Section 1, appro-
tainability assessment an effective tool to support decision-making priate keywords were identified and combined into a multi-level
necessitates the application of advanced analytical methods. The search string. The structure of the search string is illustrated in
objective of this paper is therefore to review the use of Opera- Fig. 2. It includes keywords that are related to the context (sustain-
tions Research methods to facilitate sustainability assessment of ability assessment), the subject (products), and the methods (from
products. The review is based on 142 conceptual and application- Operations Research). With regard to the methods, two comple-
oriented articles from scientific literature that have been selected mentary search strategies were applied: Search I identifies articles
in a systematic search and screening process. Our analysis aims at that mention specific Operations Research methods in their title,
highlighting the boundaries of knowledge and identifying potential abstract, or keywords, and Search II includes articles that have been
research gaps. This review differs from previous publications in its published in journals that are listed in the top quartile of the Man-
systematic search procedure, the focus on product-related sustain- agement Science and Operations Research category in the Scimago
ability assessments, and the exclusive analysis of articles that ap- Journal and Country Rank (www.scimagojr.com). The searches
ply Operations Research methods to facilitate the assessment. We were first carried out in August 2016 and updated in November
seek to make a valuable contribution by providing a structured 2017 using Elsevier’s Scopus database (www.scopus.com), which
overview of the literature published in this field and to identify provides extensive coverage of peer-reviewed scientific literature
areas that require further consideration. and offers an advanced interface for detailed analysis and data
The existing literature on sustainability assessment has been export (Harzing & Alakangas, 2016; Moed et al., 2016).
summarized from different perspectives. Ness et al. (2007) provide The results from the database search were analyzed in a struc-
a general framework to categorize tools for sustainability assess- tured screening process. First, formal criteria were applied to in-
ment. Each tool is introduced briefly and selected applications clude only articles that are written in English and published in
are discussed. More detailed reviews can be found for specific peer-reviewed scientific journals. Next, the abstracts were checked
tools. Focusing on environmental sustainability, Finnveden et al. for content-related criteria, namely that the article is related to
C. Thies et al. / European Journal of Operational Research 274 (2019) 1–21 3
Fig. 2. Overview of search strategies and results.
sustainability assessment, that it addresses a product, and that Fig. 3 illustrates the number of articles in this review by pub-
the assessment is facilitated by Operations Research methods. lication year. The growth in the number of publications over time
Self-contained sustainability assessments of companies, industries, reflects an increasing deployment of Operations Research methods
countries, or policies are not in the scope of this review and to facilitate sustainability assessment of products. This parallels
articles covering these aspects were only considered if they are the development of LCA (Guinee et al., 2011). Before 1990, LCA was
strongly related to a product. For the articles remaining in the re- in the phase of conception and a common theoretical framework
view database, the full texts were retrieved and analyzed using was missing. The focus of the 1990s was then on standardization
a structured questionnaire to extract the relevant data. Additional and first scientific journal articles were published. Not surprisingly,
references that had been identified during this step were added to the earliest article on the use of Operations Research methods
the review database (Search III), resulting in a total of 142 articles in sustainability assessment of products found in the review
that are included for further analysis. database is from 1995 (Bloemhof-Ruwaard, Koudijs, & Vis, 1995).
The authors apply multiple Operations Research methods: Analytic
3. Descriptive analysis of the review database Hierarchy Process (AHP) to aggregate environmental effects into
a single score and linear programming to improve this score by
For a descriptive analysis of the review database, we have optimizing the product composition. The focus of the reviewed
tallied the publication output over time, compiled the contributing articles from this period is clearly on environmental issues, but
journals, reviewed the products considered, and listed the sustain- efforts to integrate economic aspects start to emerge (Azapagic &
ability dimensions and indicators used in the articles. An overview Clift, 1998, 1999).
on the application of Operations Research methods as well as the During the 20 0 0s, LCA received ever-increasing attention (e.g.,
treatment of uncertainties and spatial differentiation is also given UNEP/SETAC launch Life Cycle Initiative, European Commission
as a starting point for a more detailed elaboration on these topics communicates Integrated Product Policy), life cycle-based carbon
in Section 4.
Fig. 3. Distribution of articles by publication year.

Fig. 4. Distribution of articles by contributing journals.
footprint accounting was on the rise, and diverging approaches The application areas for sustainability assessment are quite
in LCA were developed, such as dynamic LCA, spatially differen- diverse (Fig. 5). Most often, electronic products have been assessed
tiated LCA, risk-based LCA, and consequential LCA. The introduc- (29 articles), which might be due to the high regulatory pressure
tion of SLCA and LCSA also took place in that decade, with con- from legislation in this field (e.g., Directive of Waste Electrical
siderable developments in the 2010s (Guinee et al., 2011). Over and Electronic Equipment (WEEE) in the European Union). Chem-
the past decade, sustainability assessment has become consider- icals (16 articles), materials (16 articles), electricity (15 articles),
ably complex (more indicators, broader scope), highlighting the food (13 articles), cars (10 articles), industrial machines (9 ar-
need for systematic decision support based on Operations Research ticles), packaging (8 articles), buildings and consumer products
methods. (7 articles each), and fuels (6 articles) have received significant
The integration of Operations Research methods into sustain- attention.
ability assessment has been advanced mainly by the Environmental A look at the sustainability dimensions and indicators ad-
Management community, as can be concluded from the analysis dressed by the articles reviewed reveals that almost all consider
of the contributing journals (Fig. 4). The 142 articles under review at least the ecological dimension of product-related impacts
were found in 74 different journals. While 17 journals contained (Fig. 6). Related articles concentrate on the ecological dimension
more than one related article, 57 journals contributed one single alone (60 articles) or integrate the ecological dimension with the
article each. The majority of all 74 journals is related to the economic (52 articles) or social dimension (2 articles). There are
Environmental Management domain. Only 7 journals belong to the 25 articles in which all three dimensions of sustainability are as-
Management Science and Operations Research field (according to sessed simultaneously. Two articles focus on the social dimension,
SJR journal ranking 2015), namely European Journal of Operations and one article addresses the social and the economic dimension
Research, International Journal of Production Research, Omega, of sustainability. Articles with a purely economic focus are not
International Journal of Production Economics, Journal of Loss considered in the scope of this review.
Prevention in the Process Industries, Journal of the Operational Within the ecological dimension, the most commonly-used
Research Society, and Computers and Operations Research. With a indicator is the global warming potential (78 articles). Other pop-
total of 7 articles, the European Journal of Operational Research is ular ecological indicators are acidification potential (51 articles),
in the top 3 of contributing journals, being only surpassed by the toxicity potential (48 articles), photochemical oxidant formation
Journal of Cleaner Production (24 articles) and the International potential (41 articles), eutrophication potential (41 articles), and
Journal of Life Cycle Assessment (10 articles). This supports the resource depletion potential (38 articles). The choice of indicators
premise that sustainability assessment can be seen as an emerging is often related to the impact assessment method. Frequently-used
application area of Operations Research.
Fig. 5. Distribution of articles by product category. Fig. 6. Distribution of articles by sustainability dimension.
Fig. 9. Distribution of articles by level of spatial differentiation.
indicators. Furthermore, there are 13 articles that apply a variety

of methods that cannot be assigned to one of the above groups.
Fig. 7. Distribution of articles by Operations Research method.
Fig. 8 shows that uncertainty is considered in more than half
of the articles, mainly with regard to the inventory data (15 arti-
methods are, for instance, Eco-indicator 99, ReCiPe 2008, or TRACI. cles) and the decision makers’ preferences (34 articles), or both (22
The economic dimension is dominated by cost-oriented indicators articles). Uncertain inventory data is typically related to the vari-
(47 articles). A smaller number of articles uses price-oriented ability in input and output flows (e.g., spatial differences, seasonal
indicators (10 articles), profit-oriented indicators (7 articles), or changes) while uncertain preferences are deduced from the vari-
indicators considering the added value of the life cycle stages (5 ability across decision makers regarding the priority of assessment
articles). Regarding the social dimension, the variety of indicators criteria. The most common method used in the reviewed articles to
is even larger. Most often, indicators regarding the stakeholder address uncertainty is sensitivity analysis (45 articles), followed by
group “workers” (16 articles) are used, followed by “society” and fuzzy logic (26 articles) and (Monte Carlo) simulation (16 articles).
“local community” (10 articles each), “consumers” (7 articles), The level of spatial differentiation is addressed in Fig. 9. The
and “value-chain actors” (3 articles). Overall, the average number assessments in 102 articles are site-independent, meaning that av-
of indicators used in the articles is 9.5 and the median value is erage data not related to specific geographic locations is used. Only
7.0. While 103 articles consider 10 or less indicators, there is one in 20 articles, site-dependent data reflecting the regional condi-
article integrating 44 indicators and one article integrating 75 tions of the assessment is used, and in 20 articles, site-specific as-
indicators in the assessment. A detailed list of the indicators that sessments are carried out.
are applied in each article can be found in the supplementary Details on the Operations Research methods, the treatment of
material (Table S1). uncertainties, and the spatial differentiation are provided in the
The multiplicity of indicators considered in the reviewed supplementary material (Table S1) and discussed in Section 4.
articles underscores the need for the application of Operations
Research methods to obtain and interpret results when conducting 4. Application of Operations Research methods in sustainability
the sustainability assessment. These methods can be categorized assessment
into three main groups according to the purpose and the setting of
the assessment (Fig. 7). Multi-Attribute Decision Making (MADM) The following discussion is structured into four sections that
methods are applied to evaluate a finite set of alternatives based are related to the different families of Operations Research meth-
on multiple criteria (attributes), Multi-Objective Decision Making ods. Within each section, the typical application scenarios of the
(MODM) methods are applied to identify and evaluate Pareto methods in the context of sustainability assessment are described.
optimal solutions on the efficient frontier of a mathematically Moreover, we deliberate the methods’ capabilities to cope with
constrained solution space, and Data Envelopment Analysis (DEA) the challenges illustrated in Section 1 and highlight articles with
is applied to analyze the relative efficiency of a sample of al- promising approaches. Finally, we point at shortcomings of the
ternatives if the efficient frontier is unknown. MADM methods methods to derive research needs.
represent the largest group and are adopted by 86 articles to fa-
cilitate comparisons of alternative products, designs, or processes. 4.1. Multi-Attribute Decision Making (MADM)
Within this group, the most frequently applied methods are the
Analytic Hierarchy Process (AHP), Technique for Order Preference MADM models are used to make comparisons based on mul-
by Similarity to Ideal Solution (TOPSIS), outranking methods such tiple criteria (attributes) within a set of discrete alternatives. They
as Preference Ranking Organization Method for Enrichment Eval- are particularly useful when the number of attributes is large and
uation (PROMETHEE) or Elimination Et Choix Traduisant la Réalité the number of alternatives is moderate. A variety of MADM model
(ELECTRE), Simple Additive Weighting (SAW), and Multi-Attribute that differ with respect to input data and aggregation procedures
Value/Utility Theory (MAVT/MAUT). MODM methods are adopted has been developed (Guitouni & Martel, 1998; Seppälä, Basson, &
by 18 articles, mainly focusing on product design. DEA models Norris, 2001).
are used in 25 articles to assess the eco- or socio-efficiency of MADM models have been applied in 86 articles of this re-
products based on a limited number of typically incommensurable view (Table 1). Their ability to handle a large number of attributes
Fig. 8. Distribution of articles by (a) source of uncertainty and (b) method to address uncertainty.
Table 1
Overview of articles applying MADM methods.
Authors and Year Product ECOL ECON SOC Attributes Alternatives Methodological features
Weighted averages
Klein (2013) Electricity (solar power system) x x – 5 6 –
Ma and Okudan Kremer (2015) Gasoline engine x x x 5 7 Left-right fuzzy ranking
Miettinen and Hämäläinen (1997) Aluminum cans x – – 4 8 –
Milani et al. (2011) Plastic gears x x – 3 2 Signal-to-noise ratio
Park et al. (2006) Washing machine x x – 2 14 –
Pask et al. (2017) Industrial ovens x x x 7 3 Fuzzy sets, Monte Carlo
Simulation
Roth et al. (2009) Electricity (generation x x x 75 18 –
technologies)
Volkart et al. (2016) Electricity (generation x x x 18 26 –
technologies)
Wulf et al. (2017) Rare earth permanent magnets x x x 28 3 Linear and geometric
aggregation
MAVT/MAUT
Azapagic et al. (2016) Electricity (generation x x x 36 4 MAVT, integration of AHP
technologies)
Azkarate et al. (2011) Milling machine x x x 9 3 MAVT
Eddy et al. (2013) Charcoal grill x x – 10 7 MAUT
Rochat et al. (2013) PET bottles x x x 3 4 MAUT, material flow analysis
Scott et al. (2016) Carbon nanotubes for PV modules x x – 7 3 MAUT, Monte Carlo simulation
Subramanian et al. (2017) Nano-enabled biocidal paint – – x 12 1 MAVT
Tsang et al. (2014) Lumber treatment x x – 14 6 MAVT
AHP
Agarski et al. (2016) Food waste x – – 5 4 Fuzzy AHP
Ahmed Ali, Sapuan, Zainudin, and Natural fiber reinforced x x – 7 8 –
Othman (2015) composites
Avram et al. (2011) Machine tool x x – 6 24 –
Bereketli Zafeirakopoulos and Erol Hand blender x – – 11 6 ANP
Genevois (2015)
Bloemhof-Ruwaard, Koudijs et al. Fat blends x – – 9 8 AHP + Linear programming
(1995)
Chakroun, Gogu, Pacaud, and Spreader for composted products x – – 14 5 –
Thirion (2014)
Chan et al. (2014) Personal electronic products x x – 9 4 Fuzzy AHP
Chan et al. (2013) Electronic products x – – 5 4 Fuzzy AHP
De Luca et al. (2015) Citrus fruits x x x 44 9 Social LCA, participatory
approach
Eagan and Weinberg (1999) Aluminum x – – 6 2 –
García-Diéguez et al. (2015) Footwear x – – 3 4 Fuzzy AHP, Monte Carlo
simulation
Ghadimi et al. (2012) Car fuel filter x x x 7 2 Fuzzy AHP
Gloria et al. (2007) Building products x – – 12 – –
Hafizan, Noor, Abba, and Hussein Electricity (generation x – – 17 3 –
(2016) technologies)
Hermann, Kroeze, and Jawjit (2007) Pulp (for paper industry) x – – 5 1 –
Hosseinijou et al. (2014) Building materials – – x 30 2 –
Huang and Ma (2004) Packaging materials x – – 7 9 Cluster analysis
Jiang, Zhang, and Sutherland (2012) Valve system x – – 6 4 –
Kengpol and Boonkanit (2011) Air conditioners x x x 18 5 ANP
Khan et al. (2002) Urea x x – 17 4 Fuzzy Composite Programming
Kim et al. (2009) Home appliances x x – 10 6 –
Kim and Moon (2017) Cofee maker x x x 6 3 Grey relational analysis,
Bayesian network
Lipušček et al. (2010) Wood products x – — 28 – Delphi method
Lye et al. (2002) Coffee maker x x – 10 20 –
Myllyviita et al. (2013) Wood-based bioenergy x x x 8 3 Modified AHP
Myllyviita et al. (2014) Houses x – – 3 3 Comparison of SWING, SMART,
AHP
Ong, Koh, and Nee (2001) Plates x – – 30 2 –
Ordouei, Elkamel, Dusseault, and Gasoline blends x x – 6 3 –
Al-Hajri (2015)
Pineda-Henson et al. (2002) Pulp (for paper industry) x – – 7 4 –
Pineda-Henson and Culaba (2004) Semiconductors x – – 7 5 –
Ramanujan et al. (2014) Mining equipment x – – 6 8 Stochastic AHP
Ramzan et al. (2008) Distillation column x – – 11 1 –
Sadiq and Khan (2006) Urea x x – 17 4 Fuzzy composite programming
Sabaghi et al. (2016) Generic x x x 13 1 Fuzzy AHP
Shao, Taisch, and Ortega Mier Car x – x 16 – –
(2016)
(continued on next page)
Table 1 (continued)
Authors and Year Product ECOL ECON SOC Attributes Alternatives Methodological features
Wang, Chan, Lee, and Li (2015a) Consumer electronic products x – – 5 4 –

Wang et al. (2014) Electronic product x – – 5 4 Fuzzy ANP
Yu, Zhixian, and Zhiguo (2007) Chemical products x x – 11 2 –
TOPSIS
Chiang et al. (2011) Optical mouse x – – 5 5 Back-propagation neural
networks
Feng and Mai (2016) Milling machines x x x 19 6 Triangular fuzzy numbers
Gao et al. (2010) Soybean crusher x x – 11 8 Triangular fuzzy numbers,
integration of AHP
Huang, Zhang et al. (2011) PC housing materials x x – 2 8 Entropy method
Kucukvar et al. (2014) Pavements x x x 16 4 Intuitionistic fuzzy sets
Ng and Chuah (2012) Electric appliances x – – 5 4 Triangular fuzzy numbers,
integration of AHP
Onat et al. (2016) Alternative passenger vehicles x x x 16 7 Intuitionistic fuzzy sets
Pires et al. (2015) Packaging materials x – x 12 13 Integration of AHP
Tian et al. (2016) Personal electronic products x – – 5 4 Neutrosophic linguistic
numbers
Wang and Chan (2013) Automotive alternator x x x 4 4 Triangular fuzzy numbers,
fuzzy extent analysis
Wang et al. (2015b) Wireless electronic products x – – 20 4 Triangular fuzzy numbers
Outranking
Boufateh et al. (2011) T-shirt x – – 10 3 PROMETHEE I, GAIA
Canis et al. (2010) Carbon nanotubes x x – 5 4 Stochastic multi-attribute
analysis
Cinelli et al. (2017) Silver nanoparticles x – – 8 5 ELECTRE TRI, stoachstic
multi-criteria acceptability
analysis
Domingues et al. (2015) Light-duty vehicles x – – 10 6 ELECTRE TRI, constrained
weights
Geldermann and Rentz (2005) Mobile phones coating x x – 8 9 PROMETHEE I&II
Gheorghe and Xirouchakis (2007) Vacuum cleaners x x – 4 3 Fuzzy outranking
Le Téno and Mareschal (1998) Plaster wallboards x – – 3 4 PROMETHEE I&II with intervals,
GAIA
Parajuli et al. (2015) Biomass x x – 15 13 PROMETHEE I&II
Prado-Lopez et al. (2014) Laundry detergent x – – 18 2 Stochastic multi-attribute
analysis
Rajagopalan et al. (2017) Biofuels x – – 9 4 Stochastic multi-attribute
analysis
Rogers and Seager (2009) Transportation fuels x – – 6 5 PROMETHEE, stochastic
multi-attribute analysis
Samani et al. (2015) Composite construction materials x x – 6 5 PROMETHEE II
Schmitt et al. (2017) Food products x x 12 4 PROMETHEE I&II
Tervonen et al. (2009) Nanomaterials x – – 8 5 Stochastic multi-criteria
acceptability analysis
Vukelic et al. (2017) Knee support x x – 7 3 PROMETHEE II, GAIA
Other MADM
Bachmann (2013) Electricity (generation x x x 36 26 Dominating Alternative
technologies) Algorithm
Benetto et al. (2004) Electricity (from coal by-products) x – – 9 6 NAIADE
Dhouib (2014) Waste tires x x x 4 5 Fuzzy MACBETH
Dorini et al. (2011) Electricity (generation x x x 22 2 Compromise programming,
technologies) Monte Carlo simulation
Kadziński et al. (2018) Silver nanoparticles x – – 8 53 Ordinal regression
Naidu et al. (2008) Silica nanoparticles x x x 11 3 NAIADE
makes them a useful tool for the simultaneous assessment of Xirouchakis (2011), 24 different configurations of a machine tool
multiple sustainability dimensions. While ecological and economic system are compared, and in Bachmann (2013), 26 advanced elec-
criteria are considered in 23 articles, social criteria are addition- tricity generation technologies are investigated.
ally integrated in 23 articles. MADM models are also used within A basic MADM method is the calculation of weighted aver-
Social Life Cycle Assessment. De Luca, Iofrida, Strano, Falcone, and ages, which is applied in 9 articles of this review. Miettinen and
Gulisano (2015) use a total of 44 mainly social indicators to com- Hämäläinen (1997) discuss how LCA can benefit from decision
pare different scenarios for citrus growing in Southern Italy, and analysis in both the planning of an LCA study and in the inter-
Hosseinijou, Mansour, and Shirazi (2014) assess building materials pretation and understanding of the results. They suggest the use
based on 30 social indicators related to five stakeholder groups. of a linear additive model for aggregating the criteria and the
In sustainabiliy assessments considering ecological, economic, and use of weight elicitations techniques like SMART (Simple Multi-
social criteria, up to 75 indicators have been applied (Roth et al., Attribute Rating Technique) or AHP to integrate the stakeholders’
2009). In contrast to the large number of attributes, the number preferences. Roth et al. (2009) and Volkart et al. (2016) compute
of alternatives is usually moderate and does not typically exceed weighted averages to aggregate sustainabiliy indicators in the
20. Only in Lye, Lee, and Khoo (2002), 20 components of a cof- context of electricity generation. Park, Tahara, Jeong, and Lee
fee machine are analyzed regarding their specific contribution to (2006) compare simple additive weighting as a representative of
the sustainability impact of the product, in Avram, Stroud, and MADM methods to three other methods that integrate environ-
mental and economic aspects in the end-of-life stage of a washing is used (Seppälä et al., 2001). AHP has been applied for product
machine. Milani, Eskicioglu, Robles, Bujun, and Hosseini-Nasab sustainability assessment in 38 articles of this review. It is most
(2011) use simple additive weighting for combining environmental often used in combination with LCA or SLCA to support the
and economic criteria for plastic gear material selection into a determination of weights when aggregating multiple indicators
single score. They adopt a signal-to-noise ratio to limit the direct (e.g., Agarski, Budak, Vukelic, & Hodolic, 2016; Hafizan et al., 2016;
compensation among criteria by means of their variability whereby Myllyviita et al., 2014). However, AHP can also be applied on its
a material with a high average score and a low standard deviation own when LCA results are replaced by expert judgements (e.g.,
over all criteria receives highest preference. Wulf et al. (2017) test Eagan & Weinberg, 1999; Lipušček et al., 2010).
the influence of different normalization, weighting, and aggrega- While the common AHP formulation is based on unidirectional
tion procedures when combining indicators of all three sustain- hierarchical relationships among decision levels, more complex in-
ability dimensions for the case of rare earth permanent magnets. terrelationships among decision levels and components can be
Taking into account the uncertainty in design preferences, Ma and modeled using the Analytic Network Process (ANP) (Saaty, 1996).
Okudan Kremer (2015) present a fuzzy logic-based approach to ANP is also based on pairwise comparisons but allows for mod-
assess the sustainability of end-of-life options for components in eling more sophisticated decisions involving interactions and de-
the product design phase considering economic, environmental, pendencies between alternatives and attributes that exist in real
and social indicators. Fuzzy operations and a left-right fuzzy rank- life problems (Bereketli Zafeirakopoulos & Erol Genevois, 2015).
ing method (Chen and Hwang, 1992) are applied to derive weights ANP is applied by Bereketli Zafeirakopoulos and Erol Genevois
for all sustainability criteria. Similarly, Pask, Lake, Yang, Tokos, and (2015) to identify the most significant environmental aspect of a
Sadhukhan (2017) use fuzzy set theory and Monte Carlo simulation hand blender in order to guide improvements in product design.
to compute an aggregate fuzzy desirability indicator that allows Kengpol and Boonkanit (2011) use ANP for assessing product de-
for comparing alternative design options of an industrial oven. sign improvement options for an air conditioner, and Wang, Chan,
In Multi-Attribute Value Theory (MAVT) and Multi-Attribute and White (2014) implement ANP to select the most environmen-
Utility Theory (MAUT), alternatives are first evaluated against each tally sustainable design of electronic products.
criterion using partial value or utility functions and the results AHP has also been combined with other Operations Research
are then aggregated over all criteria in an additive, multiplicative, methods to increase its usefulness in specific applications. For
or mixed way to obtain a global value or utility function. While example, Bloemhof-Ruwaard, Koudijs et al. (1995) apply LCA and
MAVT is used for value measurement when there is no uncertainty AHP to evaluate the environmental impact of refined oils and
about the consequences of the alternatives, MAUT explicitly con- incorporate the results into a linear programming blending model
siders that the consequences of the alternatives may be uncertain to determine optimal fat blends under consideration of certain
(Keeney & Raiffa, 1976; Seppälä et al., 2001). availability constraints. Huang and Ma (2004) use cluster analysis
Seven of the articles reviewed employ MAVT/MAUT for sustain- to combine the rather qualitative judgments from AHP with the
ability assessment. Tsang, Bates, Madison, and Linkov (2014) use quantitative results from LCA. Kim and Moon (2017) combine
MAVT to evaluate the benefits and risks of conventional and nano- AHP with grey relational analysis to compute a total sustainability
enabled lumber treatment products. While their focus is on en- indicator for a family of coffeemakers and use Bayesian network
vironmental and economic aspects, Subramanian et al. (2017) use to analyze the risk to product redesign with regard to changes in
MAVT in combination with SLCA to analyze the social aspects components and sub-systems.
of a novel nano-copper oxide-based paint with biocidal function- In some situations, it is difficult to express the pairwise com-
ality. Indicators from all three sustainability dimensions are in- parisons of alternatives and criteria with crisp numbers due to
tegrated into an aggregate sustainability index using MAVT by lack of data or uncertainty in the decision maker’s preferences.
Azapagic, Stamford, Youds, and Barteczko-Hibbert (2016) to com- Nine of the articles analyzed integrate fuzzy logic into AHP. Quite
pare alternative electricity generation scenarios and by Azkarate, frequently, linguistic terms used for pairwise comparisons are
Ricondo, Pérez, and Martínez (2011) to compare alternative de- translated into fuzzy numbers that are further processed based on
signs of a machine tool. Eddy, Krishnamurty, Grosse, Wileden, fuzzy logic. Agarski et al. (2016) propose a fuzzy-based method-
and Lewis (2013) develop a Normative Decision Analysis Method ology for impact category weighting in LCA based on both default
for the Sustainability-Based Design of Products (NASDOP) based and subjective weighting factors. Chan, Wang, White, and Yip
on MAUT, allowing taking into account the uncertainty associated (2013) use fuzzy AHP to evaluate the performance of alternative
with material and energy flows. They use this method to analyze product designs, and Chan, Wang, and Raffoni (2014) combine
potential alternatives of a charcoal grill based on both ecological LCA with fuzzy AHP as a tool for pre-screening product designs
and economic criteria. Rochat, Binder, Diaz, and Jolliet (2013) com- before a full LCA is conducted. García-Diéguez et al. (2015) de-
bine material flow analysis, LCA, and MAUT to support the selec- velop a fuzzy AHP-based eco-design tool, which they apply to
tion of the best end-of-life scenario for polyethylene terephthalate children’s footwear. Ghadimi, Azadnia, Mohd Yusof, and Mat
(PET) bottles in Colombia. In this work, MAUT integrates ecolog- Saman (2012) use fuzzy AHP to weight selected elements and
ical, economic, and social aspects in an analysis of the problem sub-elements for a more precise assessment and better decision
from a regional perspective. Scott et al. (2016) apply MAUT to the support. Khan, Sadiq, and Husain (2002) and Sadiq and Khan
output of LCA to better understand the tradeoffs that exist in the (2006) propose a risk-oriented methodology for process plant
adoption of different nanomaterials for photovoltaic modules from design based on LCA and fuzzy composite programming involving
the public decision maker’s perspective. By integrating Monte Carlo AHP. Sabaghi, Mascle, Baptiste, and Rostamzadeh (2016) employ
simulation, they are able to analyze the extent to which MAUT fuzzy AHP accompanied by Shannon’s entropy formula to deter-
results are driven by scenario uncertainty resulting from the de- mine the relative importance of each element in the hierarchy.
cision maker’s perspective, production process uncertainties, and Wang et al. (2014) integrate fuzzy logic into ANP to support the
economic uncertainties. selection of environmentally sustainable product designs.
The Analytic Hierarchy Process (AHP) is a popular MADM method The inherent uncertainty can be modeled using stochastic or
developed by Saaty (1980) to elicit weights for each element in a simulation-based approaches. García-Diéguez et al. (2015) test the
hierarchical structure through pairwise comparisons. To this end, sensitivity of their fuzzy AHP model by means of Monte Carlo
a pre-specified nine-point scale for quantifying verbal descriptions simulation. They investigate uncertainty in the fuzzy membership
of preference among alternatives and importance among attributes function’s parameters and in the decision maker’s preferences
(criteria weights). As a result, the sustainability scores of each Outranking methods are based on pairwise comparisons of
alternative are represented as probability distributions. Ramanujan the alternatives against each other or a predefined norm. To this
et al. (2014) use a stochastic simulation module within their AHP end, preference relations are established based on performance
model to account for the variability in preferences among decision differences between the alternatives in each of the assessment
makers. criteria. A variety of outranking methods with different aggre-
When evaluating the relative importance of the criteria, it gation procedures has been developed. Two popular groups are
may be desirable to involve the affected stakeholders. In this ELECTRE and PROMETHEE. In ELECTRE, concordance and discor-
regard, AHP has the advantage that the scale used for pairwise dance matrices, which determine the strengths and weaknesses
comparisons is linked to easily understandable verbal descriptions. of alternatives based on the considered attributes, are analyzed
De Luca et al. (2015) present a participatory approach to SLCA to determine the order of alternatives, and in PROMETHEE, pos-
that involves three stakeholder groups (workers, local community, itive and negative outranking flows are calculated to obtain the
and society) who are interviewed to select relevant criteria and ranking. Outranking methods are partially compensatory because
to make the pairwise comparisons. Ramanujan et al. (2014) also the preference relations prevent compensation between particular
incorporate judgements of expert groups and integrate a stochastic attributes once a preference threshold of an attribute is exceeded.
simulation module within AHP to account for the variability in Comprehensive literature reviews on methodologies and applica-
preferences among the decision makers. The utility functions used tions are provided by Govindan and Jepsen (2016) for ELECTRE
for aggregation of parameter values in Lipušček et al. (2010) are and by Behzadian, Kazemzadeh, Albadvi, and Aghdasi (2010) for
based on experts’ findings that were gathered in a three-round PROMETHEE.
Delphi method. Fifteen articles of this review are related to outranking meth-
The Technique for Order Preference by Similarity to Ideal Solution ods. ELECTRE TRI, a special version of ELECTRE that enables the
(TOPSIS) introduced by Hwang and Yoon (1981) is based on the categorization of alternatives into a set of pre-defined classes, is
principle that good alternatives should be as close to the ideal used by Domingues et al. (2015) to classify light-duty vehicles
solution as possible and as far from the negative-ideal solution as according to their environmental performance and by Tervonen
possible. The ideal and the negative-ideal solution are generated et al. (2009) as well as Cinelli et al. (2017) for the classification
as composites of the best and worst performance values exhibited of nanoparticles. PROMETHEE is used by Geldermann and Rentz
by all alternatives. The proximity to each of these performance (2005) to investigate alternative scenarios for industrial coating of
poles is measured via Euclidean distance measured with optional mobile phones and automotive components. Two methodological
weighting of each attribute (Seppälä et al., 2001). A recent review variants of PROMETHEE are applied: PROMETHEE I, resulting in
of TOPSIS applications is provided by Behzadian, Khanmohammadi a partial preoder of alternatives based on positive and negative
Otaghsara, Yazdani, and Ignatius (2012). outranking flows, and PROMETHEE II, resulting in a total preoder
TOPSIS-based approaches have been applied in 11 articles of of alternatives based on net outranking flows. Both PROMETHEE
this review. Several useful extensions of TOPSIS and combinations variants are applied by Le Téno and Mareschal (1998) for com-
with other methods have been proposed. AHP is integrated into paring the design of building products, by Parajuli, Knudsen,
TOPSIS by Gao, Liu, Hu, Zhang, and Gu (2010), Ng and Chuah and Dalgaard (2015) for the pre-screening of potential biomasses
(2012), and Pires, Martinho, Ribeiro, Mota, and Teixeira (2015) for as feedstocks for biorefineries, and by Schmitt et al. (2017) for
a systematic determination of criteria weights. Back-propagation comparing the sustainability of local and global food products.
neural networks (BPNN) are used by Chiang, Che, and Wang PROMETHEE I is used by Boufateh, Perwuelz, Rabenasolo, and Des-
(2011) to estimate the material and energy consumption of dif- odt (2011) to obtain a ranking of different fiber types for a t-shirt.
ferent design options for an optical mouse based on the product PROMETHEE II is used by Samani, Mendes, Leal, Miranda Guedes,
attributes and previous life cycle inventory data. A triple bottom and Correia (2015) to identify the most sustainable sandwich-
line sustainability assessment model based on economic input– structured composite for novel housing solutions and Vukelic et
output analysis is adopted by Kucukvar, Gumus, Egilmez, and al. (2017) to select an optimal knee support. A stochastic approach
Tatari (2014) and Onat, Gumus, Kucukvar, and Tatari (2016). Group to PROMETHEE is proposed by Rogers and Seager (2009) to rank
TOPSIS is applied in the same articles to integrate the preferences alternative transportation fuels.
of multiple decision makers. The life cycle sustainability triangle To enhance the interpretation of the assessment results, Le
(Finkbeiner, Schau, Lehmann, & Traverso, 2010; Hofstetter et al., Téno and Mareschal (1998) suggest using PROMETHEE with the
1999) is used by Onat et al. (2016) for a better visualization Geometrical Analysis for Interactive Assistance (GAIA) (Mareschal
of the results and an analysis how the decision maker’s prior- and Brans, 1988). The idea behind GAIA is to project the multi-
ities can influence the ranking and the perceived sustainability dimensional alternatives onto a two-dimensional plane that is
performance. calculated using Principal Component Analysis. This graphical
Different fuzzy TOPSIS approaches are used to deal with vague representation, which allows for an easy identification of families
and imprecise data. Most often, criteria values and weights are of similar alternatives as well as groups of conflicting criteria, is
described by linguistic terms that are linked to fuzzy numbers. also applied by Boufateh et al. (2011) and Vukelic et al. (2017).
Triangular fuzzy numbers based on traditional fuzzy set theory Different approaches to deal with uncertainties within outrank-
(Zadeh, 1965) are employed by Feng and Mai (2016), Gao et al. ing methods have been implemented. Rather than selecting precise
(2010), Ng and Chuah (2012), Wang and Chan (2013), and Wang, criteria weights, Domingues et al. (2015) define feasible weight
Chan, and Li (2015b). As an extension, intuitionistic fuzzy num- vectors that obey certain constraints concerning, for example, the
bers (Atanassov, 1986), which quantify degrees of membership maximum acceptable ratio between any two weights. Thus, the
and non-membership to a set, are employed by Kucukvar et al. result of their analysis is not a crisp classification of each alter-
(2014) and Onat et al. (2016). Neutrosophic linguistic numbers, an native into a single class but rather a set of possible classes. This
even more generalized but also debated concept that integrates allows the decision maker to draw more robust conclusions that
differences in semantics, are applied by Tian, Wang, Wang, and are independent of specific assumptions on the weights. Gheorghe
Zhang (2016). An alternative approach to modeling uncertainty is and Xirouchakis (2007) present a fuzzy outranking method for
pursued by Huang, Zhang, Liu, and Sutherland (2011) who adopt evaluating product designs at their early stage of development
the entropy method to modify the weighting factors based on according to different criteria. The imprecision about the design
unreliability and disorder in the data. alternatives is modeled using triangular or trapezoidal fuzzy num-
bers. Alternatives are first compared within each criterion. Then, The use of spatially differentiated data can be found in 21
a fuzzy aggregation operator is applied to obtain global fuzzy articles that apply MADM methods. Most of these articles use
outranking relations. weighting factors that reflect the preferences of particular stake-
The Stochastic Multi-Attribute Life Cycle Impact Assessment holder groups. For example, site-specific weighting factors based
(SMA-LCIA), an outranking method described by Rogers and Sea- on local preferences are used by De Luca et al. (2015), Hosseini-
ger (2009), also addresses the uncertainty in criteria weights but jou et al. (2014), and Lipušček et al. (2010), and site-dependent
uses a stochastic representation in combination with Monte Carlo weighting factors reflecting the priorities in different countries are
simulation to explore feasible weight spaces. The method results used by Kim, Hwang, and Park (2009) and Rochat et al. (2013). In
in a rank acceptability index representing the probability that an this regard, MADM methods and in particular AHP have the advan-
alternative will be ranked first, second, third (and so on) as well tage of systematic weight elicitation procedures. Site-dependent
as a central weights vector required for an alternative to be ranked inventory data is used by Wulf et al. (2017) who compare al-
first. This allows the decision maker to analyze the sensitivity of ternative supply chains of rare earth permanent magnets and in
the rank ordering results to uncertainty, to eliminate alternatives articles that are based on national economic input-output tables
with poor performance in all situations, and to prioritize research (Kucukvar et al., 2014; Onat et al., 2016). National normalization
resources for reducing uncertainty. Rajagopalan, Venditti, Kelley, references are used in Domingues et al. (2015).
and Daystar (2017) apply SMA-LCIA to interpret the results of The MADM methods that have been applied in the reviewed
an LCA study on different feedstocks for biofuels. Motivated by articles have proven to be useful for the aggregation of multiple
the technological uncertainty associated with engineered nano- criteria into a single score. One advantage of such systematic
materials, Canis, Linkov, and Seager (2010) expand the SMA-LCIA aggregation procedures can be seen in their ability to combine in-
methodology by modeling uncertainty not only with regard to dicators from multiple sustainability dimensions that may even be
the criteria weights but also with regard to the performance measured on different scales. Furthermore, the systematic deter-
attributes. Prado-Lopez et al. (2014) adopt the SMA-LCIA method- mination of criteria weights allows for integrating the preferences
ology for a comparative assessment of liquid and powder laundry of different decision makers and stakeholders while simultane-
detergents. They use the Pedigree matrix (Weidema & Wesnaes, ously ensuring transparency in the assessment. The explicit use of
1996) to model the uncertainty in inventory data with regard to spatially differentiated data or preferences within MADM models
reliability, completeness, sample size, and temporal, geographical, is sparsely found in the reviewed articles, opening options for
and technological correlation. Finally, extensions to ELECTRE-TRI future research.
with the added capability of using imprecise measurement values,
thresholds, weights, and class profiles are proposed by Tervonen 4.2. Multi-Objective Decision Making (MODM)
et al. (2009) and Cinelli et al. (2017).
Another way to analyze the robustness of a ranking is proposed Being a sub-discipline of MCDM, MODM is concerned with
by Geldermann and Rentz (2005) who calculate sensitivity inter- decision-making problems where multiple conflicting objectives
vals that specify the domain in which the weighting factor for each have to be considered simultaneously and the set of alterna-
criterion can be changed so that the overall ranking of alternatives tives is large and implicitly defined by constraints. Contrary to
remains unchanged. To address the imprecision that may prevail MADM, MODM rather deals with design problems and not with
in the attribute values, Le Téno and Mareschal (1998) propose a choice problems. Hwang, Paidy, Yoon, and Masud (1980) classify
modification to PROMETHEE that is based on intervals. In fact, this multi-objective optimization methods according to the stage of
procedure is similar to fuzzy approaches and a defuzzification step the optimization procedure at which the decision maker has
is required to transform the results into a crisp ranking. to specify the preferences on the different objectives: never, a
There are 6 remaining articles that apply other MADM methods, priori, a posteriori, or interactively during the procedure. Most
which cannot be matched to any of the groups above. Kadziński commonly, optimization methods requiring an a priori or an a
et al. (2018) propose an MADM model that incorporates ordinal posteriori articulation of the preference information are used. A
regression to assess the implementation of green chemistry prin- priori approaches (e.g., weighted-sum method, utility-function
ciples in nanoparticle synthesis protocols. Bachmann (2013) dis- method, fuzzy-logic method, goal programming methods) scalarize
cusses how the development of LCSA can benefit from Multi- the different objectives into one single objective function before
Criteria Decision Analysis (MCDA). The dominating alternative al- the optimization problem is solved and only one solution is gen-
gorithm, which selects the best alternative out of a set by compar- erated. In contrast, approaches with a posteriori information (e.g.,
ing two alternatives at a time using a dominance index, is com- multiple-run weighted-sum method; ε -constraint method, multi-
pared to a total costs approach for a ranking of 26 future power objective genetic algorithms) determine a set of Pareto-optimal
generation technologies. Dhouib (2014) assesses reverse logistics solutions. The selection of the “best” alternative out of the efficient
options for waste tires using a fuzzy MACBETH (Measuring Attrac- solutions is then subject to the decision maker’s individual and
tiveness by a Categorical Based Evaluation Technique) model that subjective preferences (Hwang et al., 1980).
allows for combining qualitative and quantitative information on The articles that use MODM for sustainability assessment of
the attractiveness of alternatives. Benetto, Rousseaux, and Blondin products are listed in Table 2. In these articles, MODM is used for
(2004) compare six scenarios for electricity production from an en- product and/or process design decisions. Carreras, Boer, Cabeza,
vironmental point of view, and Naidu, Sawhney, and Li (2008) eval- Jiménez, and Guillén-Gosálbez (2016); Doi, Chujo, Yoshimura,
uate three manufacturing processes for nanoparticles based on Nishiwaki, and Izui (2009); Michaud et al. (2009), and van Mierlo,
eleven sustainability metrics using NAIADE (Novel Approach to Im- Rohmer, and Gerdessen (2017) consider optimization of the
precise Assessment and Decision Environments), a multi-criteria design of a single product, Mangun and Thurston (2002) and
method based on fuzzy pairwise comparisons, to integrate LCA re- Thurston and de la Torre (2007) focus on a portfolio of products,
sults and to analyze uncertainty. Finally, Dorini et al. (2011) dis- Tang, Wang, and Ullah (2016) on the selection of modules, and
cuss the handling of uncertainty in multi-criteria decision making Manzardo et al. (2014) and Zhou, Yin, and Hu (2009) consider the
related to sustainability assessment. They propose a compromise selection of materials. Process design decisions mainly focus on the
programming model in which uncertainty is characterized using a determination of an optimal life cycle pathway for a given prod-
probabilistic approach and propagated using Monte Carlo simula- uct. For instance, Azapagic and Clift (1998, 1999) optimize the life
tion. cycle of boron products, Herrmann et al. (2014) and Tan, Culaba,
Table 2
Overview of articles applying MODM methods.
Authors and Year Product ECOL ECON SOC Objectives Approach Methodological features
Ameli et al. (2016) Waste cell phones x x – 2 a posteriori Stochastic optimization:

simulation-based enumeration
Azapagic and Clift (1998) Boron products x x – 3 a posteriori ε -constraint
Azapagic and Clift (1999) Boron products x x – 3 a posteriori ε -constraint
Carreras et al. (2016) Buildings x x – 2 a priori Multi objective genetic algorithm
NSGA II with monetized
objective function (eco cost)
Doi et al. (2009) Milling machine x x – 3 a posteriori Normal-boundary intersection
Herrmann et al. (2014) Biodiesel x – – 5 a posteriori Hypervolume estimation taboo
search
Khan et al. (2001) Vinyl chloride monomer x x – 3 a posteriori ε -constraint
Komly et al. (2012) PET bottles x – – 3 a posteriori Multi objective genetic algorithm
NSGA II + modified TOPSIS to
identify preferred solution
Kravanja and Čuček (2013) Biogas x x – 2 a priori/a ε -constraint/monetization (eco
posteriori profit)
Mangun and Thurston (2002) Personal computers x x – 3 a priori Utility function
Manzardo et al. (2014) Pulp in paper industry x x – 2 a priori Weighted sum + TOPSIS to define
objective function
Michaud et al. (2009) Wood-plastic components x – 3 a posteriori Particle swarm optimization
Ramadhan et al. (2014) Palm-based biomass – x x 2 a priori Fuzzy logic
Tan et al. (2008) Electricity, biofuels x – – 2 a priori Fuzzy logic
Tang et al. (2016) Coal sample preparation system x x – 2 a priori Genetic algorithm with
weighted-sum objective function
Thurston and de la Torre (2007) Personal computers x x – 3 a priori Utility function
van Mierlo et al. (2017) Meat replacers x – – 4 a priori Minsum/minmax
Zhou et al. (2009) Drink containers x x – 3 a priori Genetic algorithm + artificial neural
network to forecast fitness
and Aviso (2008) of biofuels, and Komly, Azzaro-Pantel, Hubert, 2016). Thereby, Zhou et al. (2009) use an artificial neural network
Pibouleau, and Archambault (2012) of polyethylene terephthalate to forecast the fitness of material properties. Tang et al. (2016) de-
(PET) bottles. The approach typically requires the conduction of fine a weighted-sum objective function to balance GHG emissions
an LCA for a specific product or process first. Afterwards, the and a customer satisfaction index. A similar approach is applied by
product/process system under consideration is optimized taking Manzardo et al. (2014). They minimize the weighted distance from
into account economic, environmental, and/or social criteria in an ideal solution as known from TOPSIS where the ideal solution is
order to improve the system’s overall sustainability performance determined by optimizing every single objective individually. Van
(Azapagic & Clift, 1999; Ramadhan et al., 2014). Ameli, Mansour, Mierlo et al. (2017) first determine a payoff matrix by minimizing
and Ahmadi-Javid (2016) integrate product and process design each indicator separately and then derive a minsum and a minmax
decisions by setting up an optimization model involving three objective function. Carreras et al. (2016) scalarize the objective
different design options and four alternative end-of-life options for function by means of monetization using the eco-costs indicator to
a cell phone (reuse, remanufacture, recycle, and dispose of) in the quantify the cost of preventing a certain amount of environmental
design stage of the product for a hypothetical company. burden (Vogtländer et al., 2001). Monetization is additionally
The number of objectives simultaneously considered in the applied by Kravanja and Čuček (2013) who compute total profit
optimization models ranges between two and five. Most often, as the sum of economic profit and eco-profit (difference between
ecological objectives (e.g., global warming potential, energy use, eco-benefits of unburdening the environment and eco-costs of
environmental impact) are balanced against economic (e.g., profit, burdening the environment).
life cycle costs) and/or rather technical objectives such as static Most of the work related to the a posteriori approach applies
compliance (Doi et al., 2009) or reliability (e.g., Mangun and the ε -constraint method for the generation of a set of Pareto-
Thurston, 2002). Five articles fully concentrate on different ecolog- optimal solutions (e.g., Azapagic & Clift, 1999; Khan, Natrajan,
ical objectives (Herrmann et al., 2014; Komly et al., 2012; Michaud, & Revathi, 2001). To this end, one objective is selected for op-
Castéera, Fernandez, & Ndiaye, 2009; Tan et al., 2008; van Mierlo timization and the others are reformulated as constraints with
et al., 2017). Only Ramadhan et al. (2014) consider a social objec- varying ε defining the lower bound on the objective value. The
tive by balancing work-related fatalities against life cycle costs. starting value is derived by solving the corresponding single-
About half of the MODM articles in this review make use objective optimization problems (Cohon and Marks, 1975). For
of an a priori approach and generate a single solution. To this their nonlinear multi-objective optimization problem of designing
end, Mangun and Thurston (2002) and Thurston and de la Torre a milling machine, Doi et al. (2009) use the normal-boundary
(2007) maximize a normative multi-attribute utility function, intersection method that allows determining evenly spread Pareto-
trading-off the attributes cost, reliability, and environmental im- optimal solutions. These solutions are identified at the intersection
pact. Fuzzy logic is utilized by Ramadhan et al. (2014) and Tan between the boundary of the set of feasible solutions and the
et al. (2008). Here, a continuous fuzzy degree of satisfaction is normal to any point in the so-called convex hull of individual
maximized subject to predetermined upper and lower limits of minima (Das & Dennis, 1998). Both the ε -constraint method and
each objective/constraint. These limits are obtained by solving the normal-boundary intersection method require an iterative
the corresponding single-objective optimization problems. For process to identify the Pareto front. An alternate approach is the
non-linear programs, which optimize complex design problems, Non-dominated Sorting Genetic Algorithm II (NSGA II) adopted
single-objective (Tang et al., 2016; Zhou et al., 2009) or even by Komly et al. (2012) in which the Pareto front is obtained in
multi-objective genetic algorithms are applied (Carreras et al., one single run because of its ability to consider a population of
possible solutions (Deb et al., 2002). Michaud et al. (2009) develop products are compared based on their efficiency (e.g., Barba-
a multi-objective particle swarm optimization to determine a Gutiérrez, Adenso-Díaz, & Lozano, 2009; Lozano, Adenso-Díaz, &
set of Pareto-optimal solutions for eco-design problems. For the Barba-Gutiérrez, 2011; Zhu, Wang, & Zhang, 2014). Assessments
selection of a preferred solution out of the non-dominated ones, a of alternative product designs, production processes, and supply
modified TOPSIS approach is proposed by Komly et al. (2012). chains for the same product are carried out by, e.g., Chen, Zhu,
Only Ameli et al. (2016) present a stochastic optimization Yu, and Noori (2012), Martín-Gamboa, Iribarren, Susmozas, and
model in which uncertain parameters such as usage duration, Dufour (2016), and Sanjuan et al. (2011).
return time, and revenue generated from recovered products DEA models can be input- or output-oriented. Input-oriented
are taken into account when deciding about optimal design and models seek improvements in efficiency by minimizing the inputs
end-of-life options for cell phones. In this model, three different levels while keeping output levels constant. In contrast, output-
risk measures are considered (expected value, value-at-risk, and oriented models aim at maximizing the output levels while inputs
conditional value-at-risk). The model is solved by means of an are fixed. The majority of the articles analyzed use input-oriented
enumerative method that is based on Monte Carlo simulation and DEA models. Only two articles apply an output-oriented approach
obtains a set of Pareto optimal solutions for a limited number of (Hwang, Chen, Chen, Lee, & Shen, 2012; Rebolledo-Leiva, Angulo-
design alternatives. In all other articles, uncertainty is handled Meza, Iriarte, & González-Araya, 2017), and three articles apply
using sensitivity analyses, if at all. For instance, cost factors for both input- and output approaches (Korhonen and Luptacik, 2004;
the monetization of ecologic impacts (Carreras et al., 2016), objec- Limleamthong et al., 2016; Picazo-Tadeo, Beltrán-Esteve, & Gómez-
tives that are simultaneously integrated in the objective function Limón, 2012). Furthermore, DEA models can also be differentiated
(Khan et al., 2001; Kravanja & Čuček, 2013), weighting factors, or into constant returns-to-scale (CRS), if changes in inputs are
customer preferences (Mangun and Thurston, 2002; Thurston and proportional to changes in outputs, and variable returns-to-scale
de la Torre, 2007) are altered even though the uncertainty is not (VRS). CRS is the predominant variant occurring in articles covered
considered in a systematic manner. in this review which is usually assumed when the DMUs function
Spatially differentiated data is used in 4 out of the 18 MODM in a competitive market.
articles. Carreras et al. (2016) optimize the thermal insulation of a Most articles using DEA address both the environmental and
building envelope for five different locations with varying climate the economic dimension of sustainability. Eco-efficiency is, in
conditions. Manzardo et al. (2014) and Tang et al. (2016) allow for these cases, defined as the ratio of the economic product value
sourcing the same materials or modules from different suppliers (often approximated by its price) to the environmental impacts
that vary in their performance regarding economic (e.g., price), related to the product (Barba-Gutiérrez et al., 2009; Lozano et al.,
ecological (e.g., water footprint, GHG emissions), and other factors 2011). Social aspects are additionally integrated by Galán-Martín
(e.g., quality). For their case of producing green products from et al. (2016) and Izadikhah and Saen (2017).
palm-based biomass, Ramadhan et al. (2014) distinguish between The basic DEA methodology can be enhanced by several ex-
several plantations as a source of biomass, mills and biorefineries tensions. For a further differentiation of efficient DMUs, super-
for the conversion of the biomass, as well as different customers efficiency analysis (Andersen & Petersen, 1993) is applied by
for the green products. Thereby, the biomass potential, conversion Irribaren et al. (2010, 2010, 2011), Limleamthong et al. (2016), and
rates, transport distances, customer demands, and especially the Vázquez-Rowe, Villanueva-Rey, Iribarren, Teresa Moreira, and Fei-
number of fatalities at every process stage are assumed to differ joo (2012). It allows for ranking the efficient DMUs by assign-
according to their location. ing them efficiency scores greater than one. A ranking of efficient
In summary, determining an optimal product design or an DMUs can also be obtained by integrating the order-of-efficiency
optimal life cycle pathway for a given product under consideration concept (Das, 1999), which is particularly useful when a large
of economic, ecologic, and/or social objectives typically results number of criteria is analyzed (Galán-Martín et al., 2016). The con-
in complex, non-linear, stochastic multi-objective optimization cept is based on the repeated application of a standard DEA model
problems. The review of the articles indicates that MODM can for different combinations of indicators in each sustainability di-
generally facilitate sustainable product design and process deci- mension. While the endogenous optimization of weighting fac-
sions although stochastic optimization and spatial differentiation tors within DEA reduces external bias, it may lead to unrealistic
have drawn minor interest in literature so far. A posteriori meth- weighting factors. Therefore, Tatari and Kucukvar (2012) and Wier
ods have proven to be capable of obtaining a set of Pareto-optimal et al. (2005) add additional constraints to the DEA model in order
solutions when trading-off conflicting objectives related to the to enforce more realistic bounds while still allowing for some flex-
triple bottom line. However, the number of objectives that can ibility. Finally, the product systems that are represented as DMUs
be considered simultaneously is limited due to the increase in in DEA models are often viewed as black boxes with inputs and
computational effort, making the selection of relevant objectives outputs, assuming a rather aggregate perspective. More detailed
an extremely important task. The development of more efficient analyses of internal system structure are enabled by Network DEA
heuristics might be a promising approach to generating good (Färe & Grosskopf, 20 0 0; Kao, 2014). A two-stage Network DEA is
solutions to such problems in future. applied by Chen et al. (2012) to analyze the sustainable design
performance of an “industrial design” module and a “bio-design”
4.3. Data Envelopment Analysis (DEA) module. Similar approaches are presented by Mahdiloo, Jafarzadeh,
Saen, Tatham, and Fisher (2016) for car designs and by Zhu et al.
DEA is a non-parametric approach for efficiency analysis of (2014) for the production and use stage of pesticides.
decision making units (DMUs), which seeks to compare DMUs to Although stochastic and fuzzy approaches to DEA are available
their best peers (efficient frontier). The basic model, first proposed (Dyson & Shale, 2010; Hatami-Marbini, Emrouznejad, & Tavana,
by Charnes, Cooper, and Rhodes (1978), has been extended and 2011; Olesen & Petersen, 2016), only few of the articles in this re-
applied in many ways. The application of DEA for sustainability view consider uncertainty or vagueness. Most notably, Ewertowska
assessment considers the sustainability indicators as inputs and et al. (2017) use a stochastic DEA model to evaluate the environ-
outputs, and the DMUs are the alternatives to be assessed. mental efficiency of products under uncertainty. Their results show
Table 3 lists the articles in this review that use DEA to analyze that the efficiency scores obtained from the stochastic model can
the eco- or socio-efficiency of products such as food and bever- differ significantly from the scores obtained from the deterministic
ages, household appliances, and cars. In these studies, different model. Izadikhah and Saen (2017) incorporate stochastic data into
Table 3
Overview of articles applying DEA.
Article Product/DMUs ECOL ECON SOC DMUs Inputs Outputs Input- Output- CRS VRS Methodological features
oriented oriented
Barba-Gutiérrez et al. Electric/electronic x x – 9 3 1 x – — x –
(2009) appliances
Chen et al. (2012) Car designs x – – 23 4+2 2+4 n.a. n.a. x x Two-stage network DEA
Ewertowska et al. (2017) Electricity (generation x – – 11 9 1 x – – x Stochastic DEA
technologies)
Galán-Martín et al. (2016) Electricity (generation x x x 8 18 1 x – x – Order of efficiency concept
technologies)
González-García et al. Fishing vessels x x – 20 3 1 x – x – –
(2015)
Hwang et al. (2012) Car designs x – – 2500 6 2+4 – x – x Undesirable outputs
Iribarren et al. (2011) Dairy farms x x – 72 7 6 x – x – Super-efficiency analysis,
undesirable outputs
Iribarren, Electricity (wind farms) x x – 25 5 1 x – x – –
Martín-Gamboa, and
Dufour (2013)
Iribarren et al. (2010) Mussel cultivation sites x – – 67 9 1 x – x – Super efficiency analysis
Fishing vessels x – – 21 3 1 x – x – Intra- and intra sectorial
assessment
Fishing vessels x – – 15 3 1 x – x – Enhanced economic
dimension
Fishing vessels x – – 10 3 1 x – x – Window analysis
Izadikhah and Saen Pasta x x x 27 2+5 2+5 n.a. n.a. x – Chance-constrained two-stage
(2017) DEA
Korhonen and Luptacik Electricitiy (power plants) x x – 24 1 1+3 x x x x Multiple DEA models,
(2004) undesirable outputs
Kortelainen and Car models x x – 88 5 1 x – x x Absolute shadow prices
Kuosmanen (2007)
Limleamthong et al. Solvents x – – 125 9 1 x x – x Super-efficiency analysis
(2016)
Lozano et al. (2011) Electric/electronic x x – 9 3 1 x – – x –
appliances
Lozano et al. (2009) Mussel cultivation sites x – – 62 14 1 x – x x –
Mahdiloo et al. (2016) Car designs x – – 23 4+2 2+4 n.a. n.a. x x Multi-criteria two-stage DEA
Martín-Gamboa et al. Biohydrogen x – – 19 5 1 x – x – –
(2016)
Picazo-Tadeo et al. (2012) Olive farms x x – 55 4 1 x x x – Directional distance functions
Rebolledo-Leiva et al. Organic blueberries x x 14 5 2 – x – x –
(2017)
Sanjuan et al. (2011) Cheese production x x – 16 3 1 x – x – Monte Carlo Simulation for
scenarios sensitivity analysis
Tatari and Kucukvar Wall finishes x x – 11 9 1 x – x – Weight restrictions
(2012)
Vázquez-Rowe et al. Vineyards x – – 40 10 1 x – x – Super-efficiency analysis
(2012)
Wier et al. (2005) Commodities in x – – 11 7 1 x – x – Weight restrictions
households
Yang, Lu, Guo, and Refrigerators x x – 4 12 3 x – x – –
Yamamoto (2003)
Zhu et al. (2014) Pesticides x x – 10 3+1 1+1 n.a. n.a. x – Network DEA
∗
= multiple (not specified); n.a. = not applicable
a two-stage DEA model with undesirable outputs. Sanjuan et al. Based on the articles reviewed, we observe that DEA is a useful
(2011) carry out sensitivity analysis using Monte Carlo simulation tool for eco-efficiency assessment for large sets of alternatives but
to test the influence of price changes on the eco-efficiency of a limited number of criteria. It is applied to identify efficient pro-
different production scenarios for Mahón-Menorca cheese. They duction sites, to set target efficiency levels for currently inefficient
perform 10,0 0 0 simulation runs with varying prices and record sites, and to estimate the environmental and economic benefits
the number of times each production scenario is considered eco- of moving towards efficiency. Taking the endogenously derived
efficient. Korhonen and Luptacik (2004) investigate how different efficient frontier as a reference for improvement, however, also
modeling approaches within DEA influence the eco-efficiency of has some drawbacks. It neither allows for deriving improvement
power plants. In their case, the choice of a model variant influ- recommendations for already efficient units nor does it allow for
ences the eco-efficiency scores, but efficient units are efficient in judgements whether a state of sustainability would be achieved if
all model variants. Finally, Zhu et al. (2014) analyze the shadow all units operated efficiently.
prices from the DEA model to reveal its sensitivity to changes in
the input and output indicators. 4.4. Other methods
DEA is used for site-specific sustainability assessments in
12 articles. Most often, different production sites and sup- Various other Operations Research methods that facilitate
ply chains for a given product are compared. For example, sustainability assessment have been applied in the remaining
González-García et al. (2015) compare the efficiency of 20 Por- 13 articles (Table 4). The subsequent discussion of these articles
tuguese fishing vessels, Picazo-Tadeo et al. (2012) analyze a sample is structured according to the challenges which they address.
of 55 olive-growing farms in Southern Spain, and Vázquez-Rowe Beginning with the selection of relevant indicators, Gutierrez,
et al. (2012) analyze the grape production in 40 specific vineyards Lozano, Moreira, and Feijoo (2010) propose the use of statistical
in Northwestern Spain. The surrounding conditions at specific multivariate analysis techniques such as Principal Component
locations are also considered by Tatari and Kucukvar (2012) who Analysis (PCA) and Multi-Dimensional Scaling (MDS). PCA allows
analyze the eco-efficiency of construction materials. for reducing the dimensionality of a data set with several related
Table 4
Overview of articles applying other methods.
Authors and Year Product ECOL ECON SOC Attributes Methodological features
Bovea and Wang (2003) Office table x – – 1 Fuzzy QFD

Gutierrez et al. (2010) Mussels x – – 10 Principal Component Analysis,
Multi-Dimensional Scaling
Gonzáles et al. (2002) Lamp x – – 6 Fuzzy inference system
Halog (2004) Light fitting system x x – 3 Fuzzy linguistic decision support
system
Hanes et al. (2015) Ethanol x – – 1 Comprehensive allocation investigation
strategy
Jeong and Lee (2009) LCD panel x x – 2 Monetization: Conjoint Analysis
Krieg et al. (2013) Mulit-use plastic crates x x – 3 Monetization: Shadow prices from
SIMPLEX
Kuo et al. (2016) Network router x – – 9 Depth-first search
Kuo et al. (2009) Toner cartridge x – – 3 Fuzzy QFD
Prado-Lopez et al. (2016) Photovoltaic panels x – – 18 Overlap area approach
Santhanam and Gopalakrishnan Road pavement x – – 6 Qualitative approach with logical
(2013) inference
Su et al. (2012) Office chair x x – 2 Bi-level optimization: genetic
algorithm + dynamic programming
Tambouratzis et al. (2014) Liquid container x – – 4 Artifical neural networks, genetic
algorithms
variables into a smaller data set of uncorrelated variables. MDS Cruze, Goel, and Bakshi (2015) address the problem of arbitrarily
generates graphical representations of relationships among stimuli chose allocation criteria by developing a Comprehensive Allocation
that can be used for identifying and summarizing similar indica- Investigation Strategy (CAIS) based on linear model theory to ex-
tors. Prado-Lopez et al. (2016) use a probability distribution-based amine any given inventory under all possible allocation decisions.
approach to assess the significance of performance differences Only one article proposes a notable approach for spatial
among alternatives that allows LCA practitioners to focus their differentiation. Namely, Krieg et al. (2013) determine organization-
analysis on those aspects that are most influential to the decision. specific shadow prices for environmental impacts.
The comparison of alternative products is facilitated in different
ways. Monetary valuation is applied to convert sustainability 5. Promising approaches and research needs
impacts (measured in different units) to a common monetary
scale. To determine the monetary value of environmental impacts, In this section, we synthesize the typical application scenarios
Jeong and Lee (2009) use Conjoint Analysis and Krieg, Albrecht, for product sustainability assessment, highlight the promosing
and Jäger (2013) use shadow prices that can be obtained from approaches from Operations Research, and explain how they can
the SIMPLEX algorithm. An alternative approach that compares be used to address the key challenges in sustainability assessment.
products based on qualitative judgements of the decision maker This is followed by a discussion where further research is needed.
is proposed by Santhanam and Gopalakrishnan (2013). It adopts
logical inference to avoid subjective weighting and to facilitate the 5.1. Application scenarios
interpretation of the results by the decision maker.
Addressing the issue of uncertain and incomplete data, methods The typical application scenarios of product sustainability
to obtain preliminary estimates of the product-related sustainabil- assessment can be subsumed under three groups (Table 5):
ity impacts are proposed. Gonzáles, Adenso-Díaz, and González- the development of new products, the improvement of existing
Torre (2002) develop a streamlined method for impact assessment products, and the evaluation of products. The development of
in LCA making use of fuzzy logic to avoid the need for in-depth new products is in the focus of many articles because the future
environmental knowledge and extremely accurate data, making it sustainability performance of a product can be influenced signifi-
a practicable tool for small- and medium-sized enterprises with cantly during its development stage. Since conducting LCSA at this
limited resources. Streamlined assessments also integrate Quality stage is difficult due to limited data availability and significant
Function Deployment (QFD) to identify customer preferences and uncertainties, methods to obtain rough estimates of the results are
fuzzy logic to deal with imprecise information regarding the prod- applied. For example, the sustainability impacts of a new product
uct’s environmental, financial, and technical performance (Bovea & are approximated from previous product designs or material prop-
Wang, 2003; Halog, 2004; Kuo, Wu, & Shieh, 2009). The prediction erties using neural networks (Chiang et al., 2011; Tambouratzis et
of environmental impacts of new products based on depth-first al., 2014) or depth-first-search (Kuo et al., 2016). Often, multiple
search is proposed by Kuo, Smith, Smith, and Huang (2016). Their design alternatives of the new product are developed and the best
methodology facilitates early stages of an eco-design process by alternative from a sustainability perspective must be selected. This
matching new product designs to previous designs, searching is usually done by MADM models. In some cases, it is also possible
similarity graphs, separating designs into groups, and predicting to find functional relationships between the product attributes
environmental impacts from previous designs. The approach of and its sustainability impacts, which allows for the generation of
Tambouratzis et al. (2014) estimates the environmental impacts optimal product designs by application of MODM models (e.g., Doi
of different materials from their properties using artificial neural et al., 2009; Michaud et al., 2009; Tang et al., 2016).
networks and identifies the properties of a maximally sustainable The improvement of existing products usually starts with an
material for a given application using genetic algorithms. A de- assessment of their improvement potential. Specific improvement
cision support system to estimate the CO2 emissions and cost of targets can be formulated by comparing the current sustainability
product designs adopting an evolutionary-based genetic algorithm performance with the efficient frontier derived in DEA models
and dynamic programming is developed by Su, Chu, and Wang (e.g., Galán-Martín et al., 2016; Iribarren, Vazquez-Rowe, Moreira,
(2012). Uncertainty in inventory analysis also arises when multi- & Feijoo, 2010; Lozano, Iribarren, Moreira, & Feijoo, 2009). For the
functional processes are involved and allocation is required. Hanes, exploration of improvement options, MODM models are used to
Table 5
Application scenarios for Operations Research methods in sustainability assessment.
find a range of Pareto-optimal solutions (e.g., Azapagic & Clift, yet there remains considerable need for further research. The key
1998, 1999; Carreras et al., 2016). For the selection of the best im- challenges to sustainability assessment, as discussed in Section 1,
provement option, MADM models (e.g., Pineda-Henson, Culaba, & are the selection of relevant indicators/impact categories, the
Mendoza, 2002; Ramzan, Degenkolbe, & Witt, 2008) or approaches normalization and weighting of different criteria, the handling of
based on monetization (e.g., Jeong & Lee, 2009) are used. uncertain and incomplete data, and the spatial differentiation in
The evaluation of products is the most prevalent application impact assessment.
scenario in the articles reviewed. Some of the proposed methods The selection of sustainability indicators is not addressed system-
focus on prioritizing the sustainability drivers of a product with atically by most articles in this review. While hundreds of different
regard to the sustainability impacts, the product components, or indicators could be identified over the entire sample, only 9.5 in-
its life cycle phases. To this end, MADM methods like AHP (e.g., dicators are used on average within a particular assessment. The
Bereketli Zafeirakopoulos & Erol Genevois, 2015) or other methods average number is slightly higher in MADM models (11.9 indica-
like Principal Component Analysis (e.g., Gutierrez et al., 2010) are tors) and lower in DEA models (7.8 indicators) and MODM models
applied. The comparison of alternative products is often facilitated (3.7 indicators). The indicators are often selected without justifi-
by MADM and DEA models. Products can not only be compared to cation, although the selection may influence the outcome of the
other products but also be assigned to a predefined set of classes assessment significantly. For example, from a comparison of the
using outranking methods (e.g., Cinelli, Coles, & Kirwan, 2014; solutions to an MODM model with different sets of indicators as
Domingues et al., 2015). objective functions, Khan et al. (2001) find that the optimization of
The application of Operations Research methods fulfills differ- selected indicators does not necessarily improve the performance
ent purposes in relation to the life cycle-based sustainability as- of other indicators. A systematic approach to analyze the relation-
sessment methods that are integrated under the LCSA framework. ships between sustainability indicators is proposed by Gutierrez et
In some articles, the Operations Research methods are used to en- al. (2010) where multivariate statistical methods like PCA and MDS
able LCSA, i.e., the Operations Research methods become an integral are applied to reduce the number of indicators without distorting
part of LCSA and solve particular problems such as allocation or the main features of the decision problem. In this regard, future
weighting (e.g., Agarski et al., 2016; Azapagic & Clift, 1998; Hanes research should attempt to better understand how the selection
et al., 2015). In most cases, the Operations Research methods are of indicators influences the assessment results and adopt more
applied to complement LCSA, i.e., the results from an LCSA study are systematic selection procedures. The analysis of the indicators
used as an input to Operations Research methods, which facilitate selected also reveals a bias towards environmental and economic
their analysis and interpretation or combine them with other in- indicators. Social aspects are only considered in 30 of 142 articles,
dicators in order to compute aggregate sustainability indices (e.g., all of which were published after 2008. This underscores the
Geldermann and Rentz, 2005; Lozano et al., 2009; Onat et al., emerging integration of the triple bottom line of sustainability.
2016). Finally, some authors apply Operations Research methods to However, because the set of potential social indicators is very large,
substitute LCSA in the sense of a streamlined assessment, for exam- justification of relevant indicators becomes even more important.
ple, by generating rankings based on expert judgements as a start- The comparison of alternative products based on multiple crite-
ing point for more detailed life cycle-based analyses (e.g., Parajuli ria has been facilitated by Operations Research methods in many
et al., 2015; Tian et al., 2016; Wang & Chan, 2013). Especially for ways. MADM methods, which have been applied in 86 articles, are
comprehensive assessments of complex product systems, the ap- used to evaluate finite sets of alternatives. In most cases, a ranking
plication of Operations Research Methods to enable or complement of alternatives is obtained by comparing the aggregated sustain-
the established life cycle-based methods appears to be a promis- ability scores. The aggregation procedures used in different MADM
ing approach that not only receives considerable attention in many methods typically involve normalization and weighting steps that
articles of this review but also in the conceptual advancement of integrate the decision maker’s preferences in a systematic way. By
sustainability assessment methods (Cinelli et al., 2014; Onat, Ku- far the most popular method for weight elicitation is AHP (applied
cukvar, Halog, & Cloutier, 2017; Zamagni et al., 2009). in 38 articles). Furthermore, it is even possible to include the pref-
erences of multiple decision makers that may represent different
stakeholder groups or contribute different expertise (e.g., De Luca
5.2. Key challenges and research needs
et al., 2015; Dorini et al., 2011; Hosseinijou et al., 2014; Kucukvar
et al., 2014; Lipušček et al., 2010; Tian et al., 2016). An emerging
The application of Operations Research methods contributes
challenge is in the integration of semi-quantitative and qualitative
to solving some of the challenges in sustainability assessment,
indicators, which are often used for assessing social sustainability site-independent inventory data and generic criteria weights. Only
issues. For this purpose, the application of ordinal classification 20 articles use site-specific and 20 articles use site-dependent data.
methods like the recently proposed ELECTRE TRI-nB (Fernández, Site-specific inventory data is mainly used in DEA models where
Figueira, Navarro, & Roy, 2017) appears useful. DEA models, which explicit production sites of a product are compared (e.g., Iribarren
have been applied in 25 articles, are useful to compare alternatives et al., 2010; Lozano et al., 2009; Vázquez-Rowe et al., 2012). Site-
without the explicit knowledge of the decision maker’s prefer- specific inventory data is also used in some MADM and MODM ar-
ences. However, discrimination among the efficient alternatives ticles that address social sustainability (De Luca et al., 2015; Hos-
requires the integration of additional methods like super-efficiency seinijou et al., 2014; Ramadhan et al., 2014). In addition to that,
analysis. The application of such methods can sparsely be found in site-specific weighting factors elicited from local sustainability ex-
the reviewed articles (e.g., Iribarren et al., 2011; Vázquez-Rowe et perts have been integrated into MADM models (De Luca et al.,
al., 2012), and especially the application of more recent approaches 2015; Hosseinijou et al., 2014; Lipušček et al., 2010; Myllyviita et
to increase the discriminatory power of DEA based on cooperative al., 2013; Ramanujan et al., 2014). On a coarser spatial resolution,
games (Li, Xie, Wang, & Liang, 2016) or weighted goal program- site-dependent data has mainly been integrated into MADM mod-
ming (Rubem et al., 2017) appears promising and deserves further els. For example, LCA data is derived from country-specific input-
investigation. From the 18 articles in which MODM methods have output tables (Kucukvar et al., 2014; Onat et al., 2016; Tatari &
been applied, about half uses a priori approaches whereby the Kucukvar, 2012), national reference values are used for normaliza-
different objectives are scalarized based on the decision makers tion (Domingues et al., 2015), and weighting factors reflecting the
preferences to determine one optimal solution, and half uses perspectives of different makers are investigated (e.g., Agarski et
a posteriori approaches that generate a set of Pareto-optimal al., 2016; Bloemhof-Ruwaard, Koudijs et al., 1995; Gloria, Lippiatt,
solutions from which the preferred solution is selected by the & Cooper, 2007; Hermann et al., 2007). Despite these efforts, the
decision maker. Although there is no theoretical limitation, the spatial distribution of sustainability impacts is hardly considered
number of objectives that can practically be handled in MODM at the valuation stage. For example, the environmental impact of
models is limited. In fact, the number of objectives considered by electricity use depends on the regional electricity mix, or the loss
the MODM articles in this review ranges between 2 and 5, which of manual labor in developing countries would have different con-
is much smaller than the number of criteria that are typically sequences than in industrialized countries. Furthermore, a prod-
considered in sustainability assessments. Hence, there is a need uct that is evaluated as globally sustainable due to a lower over-
to employ advanced MODM solution procedures like the multi- all impact score may aggravate the situation in a particular region.
objective branch and bound algorithm (Przybylski & Gandibleux, Therefore, appropriate methods that evaluate both global and local
2017) that are capable of handling more objectives simultane- sustainability need to be developed.
ously while keeping the computational effort at a reasonable
level.
The numerous uncertainties inherent to sustainability assess-
ment are only considered in about half of the reviewed articles. 6. Conclusion
The main sources of uncertainty are the inventory data and the
decision makers’ preferences. Both sources of uncertainty are In this paper, we survey how Operations Research methods
most often addressed by means of sensitivity analysis. In addi- have been applied to facilitate sustainability assessment of prod-
tion, MADM models have been complemented with fuzzy logic ucts. From the analysis of 142 articles, which were identified by
and simulation-based approaches. Fuzzy logic allows expression a systematic search process, we observe that many methods from
of imprecise inventory data and, more importantly, vague pref- Operations Research are increasingly used to address challenges
erence information as linguistic variables that are mapped to in sustainability assessment. The use of MADM, MODM, DEA, and
fuzzy numbers. However, the parameters used for this procedure other methods from Operations Research has introduced new per-
induce additional uncertainty, which requires further investigation spectives into traditional sustainability assessment, increasing its
(García-Diéguez et al., 2015). Therefore, it appears more reason- comprehensiveness and providing new ways of communicating re-
able to analyze feasible weight sets directly using Monte Carlo sults, instead of merely presenting a list of performance indicators
simulation and to present the resulting sustainability scores as box for the alternatives considered. This opportunity has been mainly
plots or probability distributions, or to calculate the probabilities recognized by the Environmental Management community. In
that an alternative achieves a certain rank (Prado-Lopez et al., contrast, the field of product-related sustainability assessment has
2014; Ramanujan et al., 2014; Rogers & Seager, 2009; Scott et al., received little attention by the Operations Research community
2016). DEA models eliminate one potential source of uncertainty and there is significant potential in the identification and analysis
as they do not require the decision maker to express preferences of practically relevant issues in this domain. By classifying the re-
regarding the importance of the criteria. However, the remaining lated articles according to their setting and the applied Operations
uncertainties related to input and output parameters are barely Research methods, promising approaches and important research
addressed by the articles in this review despite the availability needs have been highlighted. The increasing requirement for the
of stochastic DEA approaches (Olesen & Petersen, 2016). Hence, integration of social indicators in the assessment, leading to chal-
there is a need for bridging the gap from the theoretical DEA lenges with regard to the selection of indicators, the integration
works to the specific context the sustainability assessment. Sim- of semi-quantitative and qualitative indicators, and the simulta-
ilarly, a posteriori MODM methods determine all Pareto efficient neous consideration of global and local sustainability objectives,
solutions independent of the decision maker’s preferences. Nev- are all avenues for future research. Furthermore, there remains
ertheless, more systematic approaches to address the remaining considerable potential for the application of fuzzy and stochastic
uncertainties, such as stochastic and robust optimization mod- approaches to MODM and DEA to address the uncertainty inherent
els like the recently proposed approaches by Jornada and Leon in the inventory data and the decision maker’s preferences ap-
(2016) or Hanks, Weir, and Lunday (2017), have not found broad propriately when conducting a sustainability assessment. Finally,
application in sustainability assessment yet and deserve closer it should also be recognized that challenges such as the selection
attention. of indicators, balancing incommensurable assessment criteria, and
Spatial differences have generally received little attention by the data quality issues require additional action outside the realm of
articles in this review. The assessments in 102 articles are based on Operations Research.
Acknowledgments Bovea, M. D., & Wang, B. (2003). Identifying environmental improvement options
by combining life cycle assessment and fuzzy set theory. International Journal of
Production Research, 41(3), 593–609. doi:10.1080/0020754021000033878.
The authors wish to thank the editor and the anonymous Brandenburg, M., Govindan, K., Sarkis, J., & Seuring, S. (2014). Quantitative models
reviewers for their valuable comments and suggestions. for sustainable supply chain management: Developments and directions. Euro-
pean Journal of Operational Research, 233(2), 299–312. doi:10.1016/j.ejor.2013.09.
032.
Supplementary materials Canis, L., Linkov, I., & Seager, T. P. (2010). Application of stochastic multiat-
tribute analysis to assessment of single walled carbon nanotube synthesis
processes. Environmental Science and Technology, 44(22), 8704–8711. doi:10.1021/
The review database with detailed information on bibliographic es102117k.
data, sustainability dimensions and indicators, Operations Research Carreras, J., Boer, D., Cabeza, L. F., Jiménez, L., & Guillén-Gosálbez, G. (2016). Eco-
costs evaluation for the optimal design of buildings with lower environmental
methods, treatment of uncertainties, spatial differentiation, prod-
impact. Energy and Buildings, 119, 189–199. doi:10.1016/j.enbuild.2016.03.034.
ucts, and application scenarios of each article can be found in the Cegan, J. C., Filion, A. M., Keisler, J. M., & Linkov, I. (2017). Trends and applications of
online version, at doi: http://dx.doi.org/10.1016/j.ejor.2018.04.039. multi-criteria decision analysis in environmental sciences. Literature review. En-
vironment Systems & Decisions, 37(2), 123–133. doi:10.1007/s10669- 017- 9642- 9.
Chakroun, M., Gogu, G., Pacaud, T., & Thirion, F. (2014). Eco-innovative design ap-
References proach: Integrating quality and environmental aspects in prioritizing and solv-
ing engineering problems. Frontiers of Mechanical Engineering, 9(3), 203–217.
Agarski, B., Budak, I., Vukelic, D., & Hodolic, J. (2016). Fuzzy multi-criteria-based doi:10.1007/s11465-014-0308-8.
impact category weighting in life cycle assessment. Journal of Cleaner Production, Chan, H. K., Wang, X., & Raffoni, A. (2014). An integrated approach for green de-
112, 3256–3266. doi:10.1016/j.jclepro.2015.09.077. sign: Life-cycle, fuzzy AHP and environmental management accounting. British
Ahmed Ali, B. A., Sapuan, S. M., Zainudin, E. S., & Othman, M. (2015). Implemen- Accounting Review, 46(4), 344–360. doi:10.1016/j.bar.2014.10.004.
tation of the expert decision system for environmental assessment in compos- Chan, H. K., Wang, X., White, G., & Yip, N. (2013). An extended fuzzy-AHP approach
ite materials selection for automotive components. Journal of Cleaner Production, for the evaluation of green product designs. IEEE Transactions on Engineering
107, 557–567. doi:10.1016/j.jclepro.2015.05.084. Management, 60(2), 327–339. doi:10.1109/TEM.2012.2196704.
Ameli, M., Mansour, S., & Ahmadi-Javid, A. (2016). A multi-objective model for se- Charnes, A., Cooper, W. W., & Rhodes, E. (1978). Measuring the efficiency of decision
lecting design alternatives and end-of-life options under uncertainty: A sustain- making units. European Journal of Operational Research, 2(6), 429–444. doi:10.
able approach. Resources, Conservation and Recycling, 109, 123–136. doi:10.1016/ 1016/0377-2217(78)90138-8.
j.resconrec.2016.01.011. Chen, S.-J., & Hwang, C.-L. (1992). Fuzzy multiple attribute decision making. Methods
Andersen, P., & Petersen, N. C. (1993). A procedure for ranking efficient units and applications. Berlin, Heidelberg: Springer (Lecture Notes in Economics and
in data envelopment analysis. Management Science, 39(10), 1261–1264. doi:10. Mathematical Systems, 375). doi:10.1007/978- 3- 642- 46768- 4.
1287/mnsc.39.10.1261. Chen, C., Zhu, J., Yu, J.-Y., & Noori, H. (2012). A new methodology for evaluat-
Atanassov, K. T. (1986). Intuitionistic fuzzy sets. Fuzzy Sets and Systems, 20, 87–96. ing sustainable product design performance with two-stage network data en-
doi:10.1016/S0165-0114(86)80034-3. velopment analysis. European Journal of Operational Research, 221(2), 348–359.
Avram, O., Stroud, I., & Xirouchakis, P. (2011). A multi-criteria decision method for doi:10.1016/j.ejor.2012.03.043.
sustainability assessment of the use phase of machine tool systems. Interna- Chiang, T.-A., Che, Z. H., & Wang, T.-T. (2011). A design for environment methodol-
tional Journal of Advanced Manufacturing Technology, 53(5-8), 811–828. doi:10. ogy for evaluation and improvement of derivative consumer electronic product
10 07/s0 0170-010- 2873- 2. development. Journal of Systems Science and Systems Engineering, 20(3), 260–274.
Azapagic, A., & Clift, R. (1998). Linear programming as a tool in life cycle assess- doi:10.1007/s11518- 011- 5168- 8.
ment. International Journal of Life Cycle Assessment, 3(6), 305–316. Cinelli, M., Coles, S. R., & Kirwan, K. (2014). Analysis of the potentials of multi cri-
Azapagic, A., & Clift, R. (1999). Life cycle assessment and multiobjective optimisa- teria decision analysis methods to conduct sustainability assessment. Ecological
tion. Journal of Cleaner Production, 7(2), 135–143. Indicators, 46, 138–148. doi:10.1016/j.ecolind.2014.06.011.
Azapagic, A., Stamford, L., Youds, L., & Barteczko-Hibbert, C. (2016). Towards sus- Cinelli, M., Coles, S. R., Nadagouda, M. N., Błaszczyński, J., Słowiński, R., Varma, R. S.,
tainable production and consumption. A novel decision-support framework in- & Kirwan, K. (2017). Robustness analysis of a green chemistry-based model for
tegrating economic, environmental and social sustainability (DESIRES). Comput- the classification of silver nanoparticles synthesis processes. Journal of Cleaner
ers & Chemical Engineering, 91, 93–103. doi:10.1016/j.compchemeng.2016.03.017. Production, 162, 938–948. doi:10.1016/j.jclepro.2017.06.113.
Azkarate, A., Ricondo, I., Pérez, A., & Martínez, P. (2011). An assessment method Cohon, J. L., & Marks, D. H. (1975). A review and evaluation of multiobjective
and design support system for designing sustainable machine tools. Journal of programing techniques. Water Resources Research, 11(2), 208–220. doi:10.1029/
Engineering Design, 22(3), 165–179. doi:10.1080/09544820903153570. WR011i0 02p0 0208.
Bachmann, T. M. (2013). Towards life cycle sustainability assessment: Drawing on Cooper, W. W., Hemphill, H., Huang, Z., Li, S., Lelas, V., & Sullivan, D. W. (1997).
the NEEDS project’s total cost and multi-criteria decision analysis ranking meth- Survey of mathematical programming models in air pollution management. Eu-
ods. International Journal of Life Cycle Assessment, 18(9), 1698–1709. doi:10.1007/ ropean Journal of Operational Research, 96(1), 1–35. doi:10.1016/S0377-2217(97)
s11367- 012- 0535- 3. 86747-1.
Barba-Gutiérrez, Y., Adenso-Díaz, B., & Lozano, S. (2009). Eco-efficiency of electric Das, I. (1999). A preference ordering among various Pareto optimal alternatives.
and electronic appliances: A data envelopment analysis (DEA). Environmental Structural Optimization, 18(1), 30. doi:10.10 07/s0 01580 050 065.
Modeling and Assessment, 14(4), 439–447. doi:10.1007/s10666- 007- 9134- 2. Das, I., & Dennis, J. E. (1998). Normal-boundary intersection. A new method for gen-
Barbosa-Póvoa, A. P., da Silva, C., & Carvalho, A. (2018). Opportunities and challenges erating the pareto surface in nonlinear multicriteria optimization problems. The
in sustainable supply chain. An operations research perspective. European Jour- SIAM Journal on Optimization, 8(3), 631–657. doi:10.1137/S1052623496307510.
nal of Operational Research, 268(2), 399–431. doi:10.1016/j.ejor.2017.10.036. De Luca, A. I., Iofrida, N., Strano, A., Falcone, G., & Gulisano, G. (2015). Social life
Behzadian, M., Kazemzadeh, R. B., Albadvi, A., & Aghdasi, M. (2010). PROMETHEE. A cycle assessment and participatory approaches: A methodological proposal ap-
comprehensive literature review on methodologies and applications. European plied to citrus farming in Southern Italy. Integrated Environmental Assessment
Journal of Operational Research, 200(1), 198–215. doi:10.1016/j.ejor.2009.01.021. and Management, 11(3), 383–396. doi:10.1002/ieam.1611.
Behzadian, M., Khanmohammadi Otaghsara, S., Yazdani, M., & Ignatius, J. (2012). A Deb, K., Pratap, A., Agarwal, S., & Meyarivan, T. (2002). A fast and elitist multiobjec-
state-of the-art survey of TOPSIS applications. Expert Systems with Applications, tive genetic algorithm. NSGA-II. IEEE Transactions on Evolutionary Computation,
39(17), 13051–13069. doi:10.1016/j.eswa.2012.05.056. 6(2), 182–197. doi:10.1109/4235.996017.
Benetto, E., Rousseaux, P., & Blondin, J. (2004). Life cycle assessment of coal by- Dhouib, D. (2014). An extension of MACBETH method for a fuzzy environment
products based electric power production scenarios. Fuel, 83(7-8), 957–970. to analyze alternatives in reverse logistics for automobile tire wastes. Omega
doi:10.1016/S0016- 2361(03)00258- 8. (United Kingdom), 42(1), 25–32. doi:10.1016/j.omega.2013.02.003.
Bereketli Zafeirakopoulos, I., & Erol Genevois, M. (2015). An analytic network pro- Doi, K., Chujo, Y., Yoshimura, M., Nishiwaki, S., & Izui, K. (2009). Construc-
cess approach for the environmental aspect selection problem - A case study tion of an optimum system design method considering product lifecycle.
for a hand blender. Environmental Impact Assessment Review, 54, 101–109. doi:10. International Journal of Sustainable Engineering, 2(3), 171–183. doi:10.1080/
1016/j.eiar.2015.05.002. 19397030903171740.
Bloemhof-Ruwaard, J. M., Koudijs, H. G., & Vis, J. C. (1995). Environmental impacts Domingues, A. R., Marques, P., Garcia, R., Freire, F., & Dias, L. C. (2015). Applying
of fat blends - A methodological study combining life cycle analysis, multiple multi-criteria decision analysis to the life-cycle assessment of vehicles. Journal
criteria decision making and linear programming. Environmental & Resource Eco- of Cleaner Production, 107, 749–759. doi:10.1016/j.jclepro.2015.05.086.
nomics, 6(4), 371–387. doi:10.10 07/BF0 0691820. Dorini, G., Kapelan, Z., & Azapagic, A. (2011). Managing uncertainty in multiple-
Bloemhof-Ruwaard, J. M., van Beek, P., Hordijk, L., & van Wassenhove, L. N. (1995). criteria decision making related to sustainability assessment. Clean Technologies
Interactions between operational research and environmental man- and Environmental Policy, 13(1), 133–139. doi:10.1007/s10098-010-0291-7.
agement. European Journal of Operational Research, 85(2), 229–243. Dyson, R. G., & Shale, E. A. (2010). Data envelopment analysis, operational research
doi:10.1016/0377-2217(94)00294-M. and uncertainty. Journal of the Operational Research Society, 61(1), 25–34. doi:10.
Boufateh, I., Perwuelz, A., Rabenasolo, B., & Desodt, A. M. J. (2011). Multiple criteria 1057/jors.2009.145.
decision-making for environmental impacts optimisation. International Journal of Eagan, P., & Weinberg, L. (1999). Application of analytic hierarchy process techniques
Business Performance and Supply Chain Modelling, 3(1), 28. doi:10.1504/IJBPSCM. to streamlined life-cycle analysis of two anodizing processes. Environmental Sci-
2011.039972. ence and Technology, 33(9), 1495–1500. doi:10.1021/es9807338.
Eddy, D. C., Krishnamurty, S., Grosse, I. R., Wileden, J. C., & Lewis, K. E. (2013). A Harzing, A.-W., & Alakangas, S. (2016). Google scholar, scopus and the web of sci-
normative decision analysis method for the sustainability-based design of prod- ence. A longitudinal and cross-disciplinary comparison. Scientometrics, 106(2),
ucts. Journal of Engineering Design, 24(5), 342–362. doi:10.1080/09544828.2012. 787–804. doi:10.1007/s11192- 015- 1798- 9.
745931. Hatami-Marbini, A., Emrouznejad, A., & Tavana, M. (2011). A taxonomy and review
Ewertowska, A., Pozo, C., Gavaldà, J., Jiménez, L., & Guillén-Gosálbez, G. (2017). Com- of the fuzzy data envelopment analysis literature: Two decades in the making.
bined use of life cycle assessment, data envelopment analysis and Monte Carlo European Journal of Operational Research, 214(3), 457–472. doi:10.1016/j.ejor.2011.
simulation for quantifying environmental efficiencies under uncertainty. Journal 02.001.
of Cleaner Production, 166, 771–783. doi:10.1016/j.jclepro.2017.07.215. Hermann, B. G., Kroeze, C., & Jawjit, W. (2007). Assessing environmental perfor-
Färe, R., & Grosskopf, S. (20 0 0). Network DEA. Socio-Economic Planning Sciences, mance by combining life cycle assessment, multi-criteria analysis and environ-
34(1), 35–49. doi:10.1016/S0038-0121(99)00012-9. mental performance indicators. Journal of Cleaner Production, 15(18), 1787–1796.
Feng, C., & Mai, Y. (2016). Sustainability assessment of products based on fuzzy doi:10.1016/j.jclepro.20 06.04.0 04.
multi-criteria decision analysis. International Journal of Advanced Manufacturing Herrmann, I. T., Lundberg-Jensen, M., Jørgensen, A., Stidsen, T., Spliid, H., &
Technology, 85(1-4), 695–710. doi:10.10 07/s0 0170-015- 7978- 1. Hauschild, M. (2014). Enabling optimization in LCA: From “ad hoc” to
Fernández, E., Figueira, J. R., Navarro, J., & Roy, B. (2017). ELECTRE TRI-nB. A new “structural” LCA approach - Based on a biodiesel well-to-wheel case study.
multiple criteria ordinal classification method. European Journal of Operational International Journal of Life Cycle Assessment, 19(1), 194–205. doi:10.1007/
Research, 263(1), 214–224. doi:10.1016/j.ejor.2017.04.048. s11367-013-0615-z.
Finkbeiner, M., Schau, E. M., Lehmann, A., & Traverso, M. (2010). Towards life Hofstetter, P., Braunschweig, A., Mettier, T., Müller-Wenk, R., & Tietje, O. (1999).
cycle sustainability assessment. Sustainability, 2(10), 3309–3322. doi:10.3390/ The mixing triangle: Correlation and graphical decision support for LCA-
su2103309. based comparisons. Journal of Industrial Ecology, 3(4), 97–115. doi:10.1162/
Finnveden, G., Hauschild, M. Z., Ekvall, T., Guinee, J., Heijungs, R., Hellweg, S., et al. 108819899569584.
(2009). Recent developments in life cycle assessment. Journal of Environmental Hosseinijou, S. A., Mansour, S., & Shirazi, M. A. (2014). Social life cycle as-
Management, 91(1), 1–21. doi:10.1016/j.jenvman.2009.06.018. sessment for material selection: A case study of building materials.
Galán-Martín, Á., Guillén-Gosálbez, G., Stamford, L., & Azapagic, A. (2016). Enhanced International Journal of Life Cycle Assessment, 19(3), 620–645. doi:10.1007/
data envelopment analysis for sustainability assessment: A novel methodology s11367-013-0658-1.
and application to electricity technologies. Computers and Chemical Engineering, Huang, C.-C., & Ma, H.-W. (2004). A multidimensional environmental evaluation of
90, 188–200. doi:10.1016/j.compchemeng.2016.04.022. packaging materials. Science of the Total Environment, 324(1-3), 161–172. doi:10.
Gao, Y., Liu, Z., Hu, Di, Zhang, L., & Gu, G. (2010). Selection of green product design 1016/j.scitotenv.2003.10.039.
scheme based on multi-attribute decision-making method. International Journal Huang, H., Zhang, L., Liu, Z., & Sutherland, J. W. (2011). Multi-criteria decision mak-
of Sustainable Engineering, 3(4), 277–291. doi:10.1080/19397038.2010.516371. ing and uncertainty analysis for materials selection in environmentally con-
García-Diéguez, C., Herva, M., & Roca, E. (2015). A decision support system based scious design. International Journal of Advanced Manufacturing Technology, 52(5-
on fuzzy reasoning and AHP-FPP for the ecodesign of products: Application to 8), 421–432. doi:10.10 07/s0 0170-010-2745-9.
footwear as case study. Applied Soft Computing Journal, 26, 224–234. doi:10.1016/ Huang, I. B., Keisler, J., & Linkov, I. (2011). Multi-criteria decision analysis in envi-
j.asoc.2014.09.043. ronmental sciences: Ten years of applications and trends. The Science of the Total
Gaussin, M., Hu, G., Abolghasem, S., Basu, S., Shankar, M. R., & Bidanda, B. (2013). Environment, 409(19), 3578–3594. doi:10.1016/j.scitotenv.2011.06.022.
Assessing the environmental footprint of manufactured products. A survey of Hwang, C. L., Paidy, S. R., Yoon, K, & Masud, A. S. (1980). Mathematical programming
current literature. International Journal of Production Economics, 146(2), 515–523. with multiple objectives: A tutorial. Computers and Operations Research, 7(1-2),
doi:10.1016/j.ijpe.2011.12.002. 5–31. doi:10.1016/0305-0548(80)90011-8.
Geldermann, J., & Rentz, O. (2005). Multicriteria analysis for technique assess- Hwang, C.-L., & Yoon, K. (1981). Multiple attribute decision making methods
ment:case study from industrial coating. Journal of Industrial Ecology, 9(3). and applications. A state-of-the-art survey. Berlin, Heidelberg: Springer (Lecture
doi:10.1162/1088198054821591. Notes in Economics and Mathematical Systems, 186) http://dx.doi.org/10.1007/
Ghadimi, P., Azadnia, A. H., Mohd Yusof, N., & Mat Saman, M. Z. (2012). A weighted 978- 3- 642- 48318- 9.
fuzzy approach for product sustainability assessment: A case study in automo- Hwang, S.-N., Chen, C., Chen, Y., Lee, H.-S, & Shen, P.-D. (2012). Sustainable design
tive industry. Journal of Cleaner Production, 33, 10–21. doi:10.1016/j.jclepro.2012. performance evaluation with applications in the automobile industry: Focusing
05.010. on inefficiency by undesirable factors. Omega (United Kingdom), 41(3), 553–558.
Gheorghe, R., & Xirouchakis, P. (2007). Decision-based methods for early phase sus- doi:10.1016/j.omega.2012.07.002.
tainable product design. International Journal of Engineering Education, 23(6), Iribarren, D., Hospido, A., Moreira, M. T., & Feijoo, G. (2011). Benchmarking en-
1065–1080. vironmental and operational parameters through eco-efficiency criteria for
Gloria, T. P., Lippiatt, B. C., & Cooper, J. (2007). Life cycle impact assessment weights dairy farms. Science of the Total Environment, 409(10), 1786–1798. doi:10.1016/
to support environmentally preferable purchasing in the United States. Environ- j.scitotenv.2011.02.013.
mental Science and Technology, 41(21), 7551–7557. doi:10.1021/es070750. Iribarren, D., Martín-Gamboa, M., & Dufour, J. (2013). Environmental benchmarking
Gonzáles, B., Adenso-Díaz, B., & González-Torre, P. L. (2002). A fuzzy logic approach of wind farms according to their operational performance. Energy, 61, 589–597.
for the impact assessment in LCA. Resources, Conservation and Recycling, 37, 61– doi:10.1016/j.energy.2013.09.005.
79. doi:10.1016/S0921-3449(02)0 0 069-1. Iribarren, D., Vazquez-Rowe, I., Moreira, M. T., & Feijoo, G. (2010). Further po-
González-García, S., Villanueva-Rey, P., Belo, S., Vázquez-Rowe, I., Moreira, M. T., Fei- tentials in the joint implementation of life cycle assessment and data en-
joo, G., & Arroja, L. (2015). Cross-vessel eco-efficiency analysis. A case study velopment analysis. The Science of the Total Environment, 408(22), 5265–5272.
for purse seining fishing from North Portugal targeting European pilchard. doi:10.1016/j.scitotenv.2010.07.078.
International Journal of Life Cycle Assessment, 20(7), 1019–1032. doi:10.1007/ Izadikhah, M., & Saen, R. F. (2017). Assessing sustainability of supply chains by
s11367- 015- 0887- 6. chance-constrained two-stage DEA model in the presence of undesirable fac-
Govindan, K., & Jepsen, M. B. (2016). ELECTRE. A comprehensive literature review tors. Computers and Operations Research. doi:10.1016/j.cor.2017.10.002.
on methodologies and applications. European Journal of Operational Research, Jeong, I.-T., & Lee, K.-M. (2009). Assessment of the ecodesign improvement op-
250(1), 1–29. doi:10.1016/j.ejor.2015.07.019. tions using the global warming and economic performance indicators. Journal
Grubert, E. (2017). The need for a preference-based multicriteria prioritization of Cleaner Production, 17(13), 1206–1213. doi:10.1016/j.jclepro.2009.03.017.
framework in life cycle sustainability assessment. Journal of Industrial Ecology, Jiang, Z., Zhang, H., & Sutherland, J. W. (2012). Development of an environmen-
80(1), 64. doi:10.1111/jiec.12631. tal performance assessment method for manufacturing process plans. Interna-
Guinee, J. B., Heijungs, R., Huppes, G., Zamagni, A., Masoni, P., Buonamici, R., et al. tional Journal of Advanced Manufacturing Technology, 58(5-8), 783–790. doi:10.
(2011). Life cycle assessment: Past, present, and future. Environmental Science & 10 07/s0 0170-011-3410-7.
Technology, 45(1), 90–96. doi:10.1021/es101316v. Jørgensen, A., Le Bocq, A., Nazarkina, L, & Hauschild, M. (2008). Methodologies for
Guitouni, A., & Martel, J. M. (1998). Tentative guidelines to help choosing an appro- social life cycle assessment. The International Journal of Life Cycle Assessment,
priate MCDA method. European Journal of Operational Research, 109(2), 501–521. 13(2), 96–103. doi:10.1065/lca2007.11.367.
Gutierrez, E., Lozano, S., Moreira, M. T., & Feijoo, G. (2010). Assessing relationships Jornada, D., & Leon, V. J. (2016). Biobjective robust optimization over the efficient set
among life-cycle environmental impacts with dimension reduction techniques. for Pareto set reduction. European Journal of Operational Research, 252(2), 573–
Journal of Environmental Management, 91(4), 1002–1011. doi:10.1016/j.jenvman. 586. doi:10.1016/j.ejor.2016.01.017.
20 09.12.0 09. Kadziński, M., Cinelli, M., Ciomek, K., Coles, S. R., Nadagouda, M. N., Varma, R. S.,
Hafizan, C., Noor, Z. Z., Abba, A. H., & Hussein, N. (2016). An alternative aggregation et al. (2018). Co-constructive development of a green chemistry-based model
method for a life cycle impact assessment using an analytical hierarchy process. for the assessment of nanoparticles synthesis. European Journal of Operational
Journal of Cleaner Production, 112, 3244–3255. doi:10.1016/j.jclepro.2015.09.140. Research, 264(2), 472–490. doi:10.1016/j.ejor.2016.10.019.
Halog, A. (2004). An approach to selection of sustainable product improvement al- Kao, C. (2014). Network data envelopment analysis: A review. European Journal of
ternatives with data uncertainty. Journal of Sustainable Product Design, 4(1-4), Operational Research, 239(1), 1–16. doi:10.1016/j.ejor.2014.02.039.
3–19. doi:10.1007/s10970- 006- 0002- y. Keeney, R. L., & Raiffa, H. (1976). Decisions with multiple objectives. Preferences and
Hanes, R. J., Cruze, N. B., Goel, P. K., & Bakshi, B. R. (2015). Allocation games: Ad- value tradeoffs. New York: Wiley.
dressing the Ill-posed nature of allocation in life-cycle inventories. Environmen- Keller, H. (2015). Integrated life cycle sustainability assessment - A practical ap-
tal Science and Technology, 49(13), 7996–8003. doi:10.1021/acs.est.5b01192. proach applied to biorefineries. Applied Energy, 154, 1072–1081. doi:10.1016/j.
Hanks, R. W., Weir, J. D., & Lunday, B. J. (2017). Robust goal programming using dif- apenergy.2015.01.095.
ferent robustness echelons via norm-based and ellipsoidal uncertainty sets. Eu- Kengpol, A., & Boonkanit, P. (2011). The decision support framework for developing
ropean Journal of Operational Research, 262(2), 636–646. doi:10.1016/j.ejor.2017. Ecodesign at conceptual phase based upon ISO/TR 14062. International Journal
03.072. of Production Economics, 131(1), 4–14. doi:10.1016/j.ijpe.2010.10.006.
Khan, F. I., Natrajan, B., & Revathi, P. (2001). GreenPro. A new methodology for Mangun, D., & Thurston, D. L. (2002). Incorporating component reuse, remanufac-
cleaner and greener process design. Journal of Loss Prevention in the Process In- ture, and recycle into product portfolio design. IEEE Transactions on Engineering
dustries, 14(4), 307–328. doi:10.1016/S0950-4230(0 0)0 0 047-4. Management, 49(4), 479–490. doi:10.1109/TEM.2002.807292.
Khan, F. I., Sadiq, R., & Husain, T. (2002). GreenPro-I: A risk-based life cycle Manzardo, A., Ren, J., Piantella, A., Mazzi, A., Fedele, A., & Scipioni, A. (2014). Inte-
assessment and decision-making methodology for process plant design. gration of water footprint accounting and costs for optimal chemical pulp sup-
Environmental Modelling and Software, 17(8), 669–692. doi:10.1016/ ply mix in paper industry. Journal of Cleaner Production, 72, 167–173. doi:10.
S1364-8152(02)0 0 028-2. 1016/j.jclepro.2014.03.014.
Kim, J., Hwang, Y., & Park, K. (2009). An assessment of the recycling potential of Mareschal, B., & Brans, J. P. (1988). Geometrical representations for MCDA. Euro-
materials basedon environmental and economic factors; case study in South Ko- pean Journal of Operational Research, 34(1), 69–77. doi:10.1016/0377-2217(88)
rea. Journal of Cleaner Production, 17(14), 1264–1271. doi:10.1016/j.jclepro.2009. 90456-0.
03.023. Martín-Gamboa, M., Iribarren, D., Susmozas, A., & Dufour, J. (2016). Delving into sen-
Kim, S., & Moon, S. K. (2017). Sustainable platform identification for product family sible measures to enhance the environmental performance of biohydrogen: A
design. Journal of Cleaner Production, 143, 567–581. doi:10.1016/j.jclepro.2016.12. quantitative approach based on process simulation, life cycle assessment and
073. data envelopment analysis. Bioresource Technology, 214, 376–385. doi:10.1016/j.
Klein, S. J. W. (2013). Multi-criteria decision analysis of concentrated solar power biortech.2016.04.133.
with thermal energy storage and dry cooling. Environmental Science and Tech- Michaud, F., Castéera, P., Fernandez, C., & Ndiaye, A. (2009). Meta-heuristic methods
nology, 47(24), 13925–13933. doi:10.1021/es403553u. applied to the design of woodgˆ €” plastic composites, with some attention to
Kloepffer, W. (2008). Life cycle sustainability assessment of products. The environmental aspects. Journal of Composite Materials, 43(5), 533–548. doi:10.
International Journal of Life Cycle Assessment, 13(2), 89–95. doi:10.1065/lca2008. 1177/0021998308097681.
02.376. Miettinen, P., & Hämäläinen, R. P. (1997). How to benefit from decision analysis in
Komly, C.-E., Azzaro-Pantel, C., Hubert, A., Pibouleau, L., & Archambault, V. (2012). environmental life cycle assessment (LCA). European Journal of Operational Re-
Multiobjective waste management optimization strategy coupling life cycle as- search, 102(2), 279–294.
sessment and genetic algorithms: Application to PET bottles. Resources, Conser- Milani, A. S., Eskicioglu, C., Robles, K., Bujun, K., & Hosseini-Nasab, H. (2011).
vation and Recycling, 69, 66–81. doi:10.1016/j.resconrec.2012.08.008. Multiple criteria decision making with life cycle assessment for material se-
Korhonen, P. J., & Luptacik, M. (2004). Eco-efficiency analysis of power plants. An ex- lection of composites. Express Polymer Letters, 5(12), 1062–1074. doi:10.3144/
tension of data envelopment analysis. European Journal of Operational Research, expresspolymlett.2011.104.
154(2), 437–446. doi:10.1016/S0377-2217(03)00180-2. Moed, H. F., Bar-Ilan, J., & Halevi, G. (2016). A new methodology for comparing
Korpi, E., & Ala-Risku, T. (2008). Life cycle costing. A review of pub- google scholar and scopus. Journal of Informetrics, 10(2), 533–551. doi:10.1016/
lished case studies. Managerial Auditing Journal, 23(3), 240–261. doi:10.1108/ j.joi.2016.04.017.
02686900810857703. Munda, G. (2016). Multiple criteria decision analysis and sustainable development.
Kortelainen, M., & Kuosmanen, T. (2007). Eco-efficiency analysis of consumer In Salvatore Greco, Matthias Ehrgott, & Jos Rui Figueira (Eds.), Multiple criteria
durables using absolute shadow prices. Journal of Productivity Analysis, 28(1-2), decision analysis (pp. 1235–1267). New York, NY: Springer New York (233)..
57–69. doi:10.1007/s11123- 007- 0046- 6. Myllyviita, T., Leskinen, P., Lähtinen, K., Pasanen, K., Sironen, S., Kähkönen, T., et al.
Kravanja, Z., & Čuček, L. (2013). Multi-objective optimisation for generating sustain- (2013). Sustainability assessment of wood-based bioenergy - A methodological
able solutions considering total effects on the environment. Applied Energy, 101, framework and a case-study. Biomass and Bioenergy, 59, 293–299. doi:10.1016/j.
67–80. doi:10.1016/j.apenergy.2012.04.025. biombioe.2013.07.010.
Krieg, H., Albrecht, S., & Jäger, M. (2013). Systematic monetisation of environmental Myllyviita, T., Leskinen, P., & Seppälä, J. (2014). Impact of normalisation, elicitation
impacts. WIT Transactions on Ecology and the Environment, 173, 513–524. doi:10. technique and background information on panel weighting results in life cy-
2495/SDP130431. cle assessment. The International Journal of Life Cycle Assessment, 19(2), 377–386.
Kucukvar, M., Gumus, S., Egilmez, G., & Tatari, O. (2014). Ranking the sustainabil- doi:10.1007/s11367-013-0645-6.
ity performance of pavements. An intuitionistic fuzzy decision making method. Naidu, S., Sawhney, R., & Li, X. (2008). A methodology for evaluation and selection
Automation in Construction, 40, 33–43. doi:10.1016/j.autcon.2013.12.009. of nanoparticle manufacturing processes based on sustainability metrics. Envi-
Kuo, T.-C., Smith, S., Smith, G. C., & Huang, S. H. (2016). A predictive product at- ron. Sci. Technol, 42(17), 6697–6702. doi:10.1021/es703030r.
tribute driven eco-design process using depth-first search. Journal of Cleaner Ness, B., Urbel-Piirsalu, E., Anderberg, S., & Olsson, L. (2007). Categorising tools
Production, 112, 3201–3210. doi:10.1016/j.jclepro.2015.10.005. for sustainability assessment. Ecological Economics, 60(3), 498–508. doi:10.1016/
Kuo, T.-C., Wu, H.-H., & Shieh, J.-I. (2009). Integration of environmental considera- j.ecolecon.2006.07.023.
tions in quality function deployment by using fuzzy logic. Expert Systems with Ng, C. Y., & Chuah, K. B. (2012). Evaluation of eco design alternatives by integrating
Applications, 36(3 PART 2), 7148–7156. doi:10.1016/j.eswa.2008.08.029. ahp and topsis methodology under a fuzzy environment. International Journal
Le Téno, J. F., & Mareschal, B. (1998). An interval version of PROMETHEE for the of Management Science and Engineering Management, 7(1), 43–52. doi:10.1080/
comparison of building products’ design with ill-defined data on environmental 17509653.2012.10671206.
quality. European Journal of Operational Research, 109(2), 522–529. Olesen, O. B., & Petersen, N. C. (2016). Stochastic data envelopment analysis - A
Li, Y., Xie, J., Wang, M., & Liang, L. (2016). Super efficiency evaluation using a com- review. European Journal of Operational Research, 251(1), 2–21. doi:10.1016/j.ejor.
mon platform on a cooperative game. European Journal of Operational Research, 2015.07.058.
255(3), 884–892. doi:10.1016/j.ejor.2016.06.001. Onat, N., Kucukvar, M., Halog, A., & Cloutier, S. (2017). Systems thinking for life cycle
Limleamthong, P., Gonzalez-Miquel, M., Papadokonstantakis, S., Papadopoulos, A. I., sustainability assessment. A review of recent developments, applications, and
Seferlis, P., & Guillén-Gosálbez, G. (2016). Multi-criteria screening of chemicals future perspectives. Sustainability, 9(5), 706. doi:10.3390/su9050706.
considering thermodynamic and life cycle assessment metrics via data envelop- Onat, N. C., Gumus, S., Kucukvar, M., & Tatari, O. (2016). Application of the TOPSIS
ment analysis. Application to CO2 capture. Green Chemistry, 18(24), 6468–6481. and intuitionistic fuzzy set approaches for ranking the life cycle sustainability
doi:10.1039/C6GC01696K. performance of alternative vehicle technologies. Sustainable Production and Con-
Lipušček, I., Bohanec, M., Oblak, L., & Zadnik Stirn, L. (2010). A multi-criteria sumption, 6, 12–25. doi:10.1016/j.spc.2015.12.003.
decision-making model for classifying wood products with respect to their im- Ong, S., Koh, T., & Nee, A. (2001). Assessing the environmental impact of materials
pact on environment. International Journal of Life Cycle Assessment, 15(4), 359– processing techniques using an analytical hierarchy process method. Journal of
367. doi:10.1007/s11367-010-0157-6. Materials Processing Technology, 113(1-3), 424–431. doi:10.1016/S0924-0136(01)
Liu, S., Leat, M., & Smith, M. H. (2011). State-of-the-art sustainability analy- 00618-5.
sis methodologies for efficient decision support in green production opera- Ordouei, M. H., Elkamel, A., Dusseault, M. B., & Al-Hajri, I. (2015). New sustainabil-
tions. International Journal of Sustainable Engineering, 4(3), 236–250. doi:10.1080/ ity indices for product design employing environmental impact and risk reduc-
19397038.2011.574744. tion: Case study on gasoline blends. Journal of Cleaner Production, 108, 312–320.
Lozano, S., Adenso-Díaz, B., & Barba-Gutiérrez, Y. (2011). Russell non-radial eco- doi:10.1016/j.jclepro.2015.06.126.
efficiency measure and scale elasticity of a sample of electric/electronic prod- Parajuli, R., Knudsen, M. T., & Dalgaard, T. (2015). Multi-criteria assessment of
ucts. Journal of the Franklin Institute, 348(7), 1605–1614. doi:10.1016/j.jfranklin. yellow, green, and woody biomasses: Pre-screening of potential biomasses as
2011.02.005. feedstocks for biorefineries. Biofuels, Bioproducts and Biorefining, 9(5), 545–566.
Lozano, S., Iribarren, D., Moreira, M. T., & Feijoo, G. (2009). The link between op- doi:10.1002/bbb.1567.
erational efficiency and environmental impacts. A joint application of life cycle Park, P.-J., Tahara, K., Jeong, I.-T., & Lee, K.-M. (2006). Comparison of four methods
assessment and data envelopment analysis. The Science of the total environment, for integrating environmental and economic aspects in the end-of-life stage of
407(5), 1744–1754. doi:10.1016/j.scitotenv.2008.10.062. a washing machine. Resources, Conservation and Recycling, 48(1), 71–85. doi:10.
Lye, S. W., Lee, S. G., & Khoo, M. K. (2002). ECoDE - An environmental component 1016/j.resconrec.20 06.01.0 01.
design evaluation tool. Engineering with Computers, 18(1), 14–23. doi:10.1007/ Pask, F., Lake, P., Yang, A., Tokos, H., & Sadhukhan, J. (2017). Sustainability indicators
s0 0366020 0 0 01. for industrial ovens and assessment using Fuzzy set theory and Monte Carlo
Ma, J., & Okudan Kremer, G. E. (2015). A fuzzy logic-based approach to determine simulation. Journal of Cleaner Production, 140, 1217–1225. doi:10.1016/j.jclepro.
product component end-of-life option from the views of sustainability and de- 2016.10.038.
signer’s perception. Journal of Cleaner Production, 108, 289–300. doi:10.1016/j. Petti, L., Serreli, M., & Di Cesare, S. (2016). Systematic literature review in social life
jclepro.2015.08.029. cycle assessment. The International Journal of Life Cycle Assessment. doi:10.1007/
Mahdiloo, M., Jafarzadeh, A. H., Saen, R. F., Tatham, P., & Fisher, R. (2016). A mul- s11367- 016- 1135- 4.
tiple criteria approach to two-stage data envelopment analysis. Transportation Picazo-Tadeo, A. J., Beltrán-Esteve, M., & Gómez-Limón, J. A. (2012). Assessing eco-
Research Part D: Transport and Environment, 46, 317–327. doi:10.1016/j.trd.2016. efficiency with directional distance functions. European Journal of Operational
04.008. Research, 220(3), 798–809. doi:10.1016/j.ejor.2012.02.025.
Pineda-Henson, R., & Culaba, A. B. (2004). A diagnostic model for green produc- Scott, R. P., Cullen, A. C., Fox-Lent, C., & Linkov, I. (2016). Can carbon nanomate-
tivity assessment of manufacturing processes. International Journal of Life Cycle rials improve CZTS photovoltaic devices? Evaluation of performance and im-
Assessment, 9(6), 379–386. doi:10.1065/lca2004.09.180.7. pacts using integrated life-cycle assessment and decision analysis. Risk Analysis.
Pineda-Henson, R., Culaba, A. B., & Mendoza, G. A. (2002). Evaluating environmen- doi:10.1111/risa.12539.
tal performance of pulp and paper manufacturing using the analytic hierar- Seppälä, J., Basson, L., & Norris, G. A. (2001). Decision analysis frameworks for life-
chy process and life-cycle assessment. Journal of Industrial Ecology, 6(1), 15–28. cycle impact assessment. Journal of Industrial Ecology, 5(4), 45–68. doi:10.1162/
doi:10.1162/108819802320971614. 10881980160084033.
Pires, A., Martinho, G., Ribeiro, R., Mota, M., & Teixeira, L. (2015). Extended Seuring, S. (2013). A review of modeling approaches for sustainable supply chain
producer responsibility: A differential fee model for promoting sustainable management. Decision Support Systems, 54(4), 1513–1520. doi:10.1016/j.dss.2012.
packaging. Journal of Cleaner Production, 108, 343–353. doi:10.1016/j.jclepro. 05.053.
2015.07.084. Shao, J., Taisch, M., & Ortega Mier, M. (2016). A study on a configuration model for
Pizzol, M., Weidema, B., Brandão, M., & Osset, P. (2015). Monetary valuation in life facilitating sustainable consumption. A case involving the automobile industry
cycle assessment: A review. Journal of Cleaner Production, 86, 170–179. doi:10. in Italy. Journal of Cleaner Production, 137, 507–515. doi:10.1016/j.jclepro.2016.07.
1016/j.jclepro.2014.08.007. 130.
Prado-Lopez, V., Seager, T. P., Chester, M., Laurin, L., Bernardo, M., & Tylock, S. (2014). Su, J. C. P., Chu, C.-H., & Wang, Y.-T. (2012). A decision support system to es-
Stochastic multi-attribute analysis (SMAA) as an interpretation method for com- timate the carbon emission and cost of product designs. International Jour-
parative life-cycle assessment (LCA). The International Journal of Life Cycle Assess- nal of Precision Engineering and Manufacturing, 13(7), 1037–1045. doi:10.1007/
ment, 19(2), 405–416. doi:10.1007/s11367-013-0641-x. s12541-012-0135-y.
Prado-Lopez, V., Wender, B. A., Seager, T. P., Laurin, L., Chester, M., & Ar- Subramanian, V., Semenzin, E., Zabeo, A., Saling, P., Ligthart, T., & van Harme-
slan, E. (2016). Tradeoff evaluation improves comparative life cycle assess- len, T. (2017). Assessing the social impacts of nano-enabled products through
ment. A photovoltaic case study. Journal of Industrial Ecology, 20(4), 710–718. the life cycle. The case of nano-enabled biocidal paint. The International Journal
doi:10.1111/jiec.12292. of Life Cycle Assessment, 15, 156. doi:10.1007/s11367- 017- 1324- 9.
Przybylski, A., & Gandibleux, X. (2017). Multi-objective branch and bound. European Tambouratzis, T., Karalekas, D., & Moustakas, N. (2014). A methodological study for
Journal of Operational Research, 260(3), 856–872. doi:10.1016/j.ejor.2017.01.032. optimizing material selection in sustainable product design. Journal of Industrial
Rajagopalan, N., Venditti, R., Kelley, S., & Daystar, J. (2017). Multi-attribute uncer- Ecology, 18(4), 508–516. doi:10.1111/jiec.12035.
tainty analysis of the life cycle of lignocellulosic feedstock for biofuel produc- Tan, R. R., Culaba, A. B., & Aviso, K. B. (2008). A fuzzy linear programming exten-
tion. Biofuels, Bioproducts and Biorefining, 11(2), 269–280. doi:10.1002/bbb.1737. sion of the general matrix-based life cycle model. Journal of Cleaner Production,
Ramadhan, N. J., Wan, Y. K., Ng, R., Ng, D., Hassim, M. H., Aviso, K. B, et al. (2014). 16(13), 1358–1367. doi:10.1016/j.jclepro.2007.06.020.
Life cycle optimisation (LCO) of product systems with consideration of occu- Tang, C. S., & Zhou, S. (2012). Research advances in environmentally and socially
pational fatalities. Process Safety and Environmental Protection, 92(5), 390–405. sustainable operations. European Journal of Operational Research, 223(3), 585–
doi:10.1016/j.psep.2014.04.003. 594. doi:10.1016/j.ejor.2012.07.030.
Ramanujan, D., Bernstein, W. Z., Choi, J.-K., Koho, M., Zhao, F., & Ramani, K. (2014). Tang, D., Wang, Q., & Ullah, I. (2016). Optimisation of product configuration in con-
Prioritizing design for environment strategies using a stochastic analytic hier- sideration of customer satisfaction and low carbon. International Journal of Pro-
archy process. Journal of Mechanical Design, Transactions of the ASME, 136(7). duction Research, 1–25. doi:10.1080/00207543.2016.1231430.
doi:10.1115/1.4025701. Tatari, O., & Kucukvar, M. (2012). Eco-efficiency of construction materials: Data en-
Ramzan, N., Degenkolbe, S., & Witt, W. (2008). Evaluating and improving environ- velopment analysis. Journal of Construction Engineering and Management, 138(6),
mental performance of HC’s recovery system: A case study of distillation unit. 733–741. doi:10.1061/(ASCE)CO.1943-7862.0 0 0 0484.
Chemical Engineering Journal, 140(1-3), 201–213. doi:10.1016/j.cej.2007.09.042. Tervonen, T., Linkov, I., Figueira, J. R., Steevens, J., Chappell, M., & Merad, M. (2009).
Reap, J., Roman, F., Duncan, S., & Bras, B. (2008). A survey of unresolved problems Risk-based classification system of nanomaterials. The Journal of Nanoparticle Re-
in life cycle assessment. The International Journal of Life Cycle Assessment, 13(5), search, 11(4), 757–766. doi:10.1007/s11051- 008- 9546- 1.
374–388. doi:10.10 07/s11367-0 08-0 0 09-9. Thurston, D. L., & de la Torre, J. P. (2007). Leasing and extended producer responsi-
Rebolledo-Leiva, R., Angulo-Meza, L., Iriarte, A., & González-Araya, M. C. (2017). Joint bility for personal computer component reuse. International Journal of Environ-
carbon footprint assessment and data envelopment analysis for the reduction ment and Pollution, 29(1-3), 104–126.
of greenhouse gas emissions in agriculture production. The Science of the Total Tian, Z.-P., Wang, J., Wang, J.-Q., & Zhang, H.-Y. (2016). Simplified neutrosophic lin-
Environment, 593-594, 36–46. doi:10.1016/j.scitotenv.2017.03.147. guistic multi-criteria group decision-making approach to green product devel-
Rochat, D., Binder, C. R., Diaz, J., & Jolliet, O. (2013). Combining material flow analy- opment. Group Decision and Negotiation. doi:10.1007/s10726- 016- 9479- 5.
sis, life cycle assessment, and multiattribute utility theory: Assessment of end- Tsang, M. P., Bates, M. E., Madison, M., & Linkov, I. (2014). Benefits and risks of
of-life scenarios for polyethylene terephthalate in Tunja, Colombia Rochat et al. emerging technologies: Integrating life cycle assessment and decision analysis
Combining MFA, LCA, and MAUT. Journal of Industrial Ecology, 17(5), 642–655. to assess lumber treatment alternatives. Environmental Science and Technology,
doi:10.1111/jiec.12025. 48(19), 11543–11550. doi:10.1021/es501996s.
Rogers, K., & Seager, T. P. (2009). Environmental decision-making using life cycle UNEP/SETAC. (2011). Towards a life cycle sustainability assessment. Making informed
impact assessment and stochastic multiattribute decision analysis. A case study choices on products. United Nations Environment Programme (UNEP)/Society of
on alternative transportation fuels. Environmental Science & Technology, 43(6), Environmental Toxicology and Chemistry (SETAC) accessed on 2017-02-06.
1718–1723. doi:10.1021/es801123h. Van Mierlo, K., Rohmer, S., & Gerdessen, J. C. (2017). A model for composing meat
Roth, S., Hirschberg, S., Bauer, C., Burgherr, P., Dones, R., Heck, T., & Schen- replacers: Reducing the environmental impact of our food consumption pattern
ler, W. (2009). Sustainability of electricity supply technology portfolio. Annals while retaining its nutritional value. Journal of Cleaner Production, 165, 930–950.
of Nuclear Energy, 36, 409–416. doi:10.1016/j.anucene.2008.11.029. doi:10.1016/j.jclepro.2017.07.098.
Rubem, A. P. d. S., Soares de Mello, J. C. C., & Angulo Meza, L. (2017). A goal pro- Vázquez-Rowe, I., Villanueva-Rey, P., Iribarren, D., Teresa Moreira, M., & Fei-
gramming approach to solve the multiple criteria DEA model. European Journal joo, G. (2012). Joint life cycle assessment and data envelopment analysis of
of Operational Research, 260(1), 134–139. doi:10.1016/j.ejor.2016.11.049. grape production for vinification in the Rías Baixas appellation (NW Spain).
Saaty, T. L. (1980). The analytic hierarchy process. Planning, priority setting, resource Journal of Cleaner Production, 27, 92–102. doi:10.1016/j.jclepro.2011.12.039.
allocation. New York: McGraw-Hill. Vogtländer, J. G., Brezet, H. C., & Hendriks, C. F. (2001). The virtual eco-costs ‘99 A
Saaty, T. L. (1996). Decision making with dependence and feedback: The analytic net- single LCA-based indicator for sustainability and the eco-costs-value ratio (EVR)
work process. Pittsburgh: RWS Publications. model for economic allocation. The International Journal of Life Cycle Assessment,
Sabaghi, M., Mascle, C., Baptiste, P., & Rostamzadeh, R. (2016). Sustainability as- 6(3), 157–166. doi:10.1007/BF02978734.
sessment using fuzzy-inference technique (SAFT): A methodology toward green Volkart, K., Bauer, C., Burgherr, P., Hirschberg, S., Schenler, W., & Spada, M. (2016).
products. Expert Systems with Applications, 56, 69–79. doi:10.1016/j.eswa.2016. Interdisciplinary assessment of renewable, nuclear and fossil power generation
02.038. with and without carbon capture and storage in view of the new Swiss energy
Sadiq, R., & Khan, F. I. (2006). An integrated approach for risk-based life cycle as- policy. International Journal of Greenhouse Gas Control, 54, 1–14. doi:10.1016/j.
sessment and multi-criteria decision-making: Selection, design and evaluation ijggc.2016.08.023.
of cleaner and greener processes. Business Process Management Journal, 12(6), Vukelic, D., Budak, I., Tadic, B., Simunovic, G., Kljajic, V., & Agarski, B. (2017). Multi-
770–792. doi:10.1108/14637150610710927. criteria decision-making and life cycle assessment model for optimal product
Samani, P., Mendes, A., Leal, V., Miranda Guedes, J., & Correia, N. (2015). A sustain- selection. Case study of knee support. International Journal of Environmental Sci-
ability assessment of advanced materials for novel housing solutions. Building ence and Technology, 14(2), 353–364. doi:10.1007/s13762- 016- 1151- 8.
and Environment, 92, 182–191. doi:10.1016/j.buildenv.2015.04.012. Wang, X., & Chan, H. K. (2013). An integrated fuzzy approach for evaluating reman-
Sanjuan, N., Ribal, J., Clemente, G., & Fenollosa, M. L. (2011). Measuring and improv- ufacturing alternatives of a product design. Journal of Remanufacturing, 3(1), 10.
ing eco-efficiency using data envelopment analysis: A case study of Mahón- doi:10.1186/2210- 4690- 3- 10.
Menorca cheese. Journal of Industrial Ecology, 15(4), 614–628. doi:10.1111/j. Wang, X., Chan, H. K., Lee, C., & Li, D. (2015a). A hierarchical model for eco-design of
1530-9290.2011.00347.x. consumer electronic products. Technological and Economic Development of Econ-
Santhanam, G. R., & Gopalakrishnan, K. (2013). Pavement life-cycle sustainability omy, 21(1), 48–64. doi:10.3846/20294913.2013.876685.
assessment and interpretation using a novel qualitative decision procedure. Wang, X., Chan, H. K., & Li, D. (2015b). A case study of an integrated fuzzy method-
Journal of Computing in Civil Engineering, 27(5), 544–554. doi:10.1061/(ASCE)CP. ology for green product development. European Journal of Operational Research,
1943-5487.0 0 0 0228. 241(1), 212–223. doi:10.1016/j.ejor.2014.08.007.
Schmitt, E., Galli, F., Menozzi, D., Maye, D., Touzard, J.-M., & Marescotti, A. (2017). Wang, X., Chan, H. K., & White, L. (2014). A comprehensive decision support model
Comparing the sustainability of local and global food products in Europe. Jour- for the evaluation of eco-designs. Journal of the Operational Research Society,
nal of Cleaner Production, 165, 346–359. doi:10.1016/j.jclepro.2017.07.039. 65(6), 917–934. doi:10.1057/jors.2013.23.
Weidema, B. P., & Wesnaes, M. S. (1996). Data quality management for life cycle Zadeh, L. (1965). Fuzzy sets. Information and Control, 8, 338–353. doi:10.1016/
inventories - an example of using data quality indicators. Journal of Cleaner Pro- S0019- 9958(65)90241- X.
duction, 4(3-4), 167–174. Zamagni, A. Buttol, P. Masoni, P. Guinee, J. B. Huppes, G.&, Heijungs, R. (2009).
Wier, M., Christoffersen, L. B., Jensen, T. S., Pedersen, O. G., Keiding, H., & Munks- D20 blue paper on life cycle sustainability analysis. CALCAS project. https:
gaard, J. (2005). Evaluating sustainability of household consumption - Using //www.leidenuniv.nl/cml/ssp/publications/calcas_report_d20.pdf, Accessed on 21
DEA to assess environmental performance. Economic Systems Research, 17(4), November 2017.
425–447. doi:10.1080/09535310500284276. Zanghelini, G. M., Cherubini, E., & Soares, S. R. (2018). How multi-criteria decision
Wulf, C., Zapp, P., Schreiber, A., Marx, J., & Schlör, H. (2017). Lessons learned from a analysis (MCDA) is aiding life cycle assessment (LCA) in results interpretation.
life cycle sustainability assessment of rare earth permanent magnets. Journal of Journal of Cleaner Production, 172, 609–622. doi:10.1016/j.jclepro.2017.10.230.
Industrial Ecology, 21(6), 1578–1590. doi:10.1111/jiec.12575. Zhou, C.-C., Yin, G.-F., & Hu, X.-B. (2009). Multi-objective optimization of material
Yang, Y., Lu, G.-H., Guo, X., & Yamamoto, R. (2003). Greenness assessment of prod- selection for sustainable products: Artificial neural networks and genetic algo-
ucts in PLCA by DEA approach. Materials Transactions, 44(4), 645–648. rithm approach. Materials and Design, 30(4), 1209–1215. doi:10.1016/j.matdes.
Yu, Q., Zhixian, H., & Zhiguo, Y. (2007). Integrated assessment of environmental 20 08.06.0 06.
and economic performance of chemical products using analytic hierarchy pro- Zhu, Z., Wang, K., & Zhang, B. (2014). Applying a network data envelopment analysis
cess approach. Chinese Journal of Chemical Engineering, 15(1), 81–87. doi:10.1016/ model to quantify the eco-efficiency of products: A case study of pesticides.
S10 04-9541(07)60 037-8. Journal of Cleaner Production, 69, 67–73. doi:10.1016/j.jclepro.2014.01.064.

Operations Research For Sustainability PDF

Uploaded by

Copyright:

Available Formats

Operations Research For Sustainability PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Operations Research For Sustainability PDF

Uploaded by

Copyright:

Available Formats

European Journal of Operational Research 274 (2019) 1–21

Contents lists available at ScienceDirect

European Journal of Operational Research

Operations research for sustainability assessment of products: A

1. Introduction eration. Consequently, much attention is directed on methods to

(2009) discuss the methodological developments in LCA, and

Fig. 2. Overview of search strategies and results.

Fig. 3. Distribution of articles by publication year.

Fig. 4. Distribution of articles by contributing journals.

Fig. 9. Distribution of articles by level of spatial differentiation.

indicators. Furthermore, there are 13 articles that apply a variety

Wang, Chan, Lee, and Li (2015a) Consumer electronic products x – – 5 4 –

Ameli et al. (2016) Waste cell phones x x – 2 a posteriori Stochastic optimization:

Bovea and Wang (2003) Oﬃce table x – – 1 Fuzzy QFD

You might also like