Integrated Environmental Assessment and Management — Volume 1, Number 2 — pp.

95–108
Ó 2005 SETAC 95

Application of Multicriteria Decision Analysis in

Environmental Decision Making
Gregory A. Kiker,*
Todd S. Bridges,
Arun Varghese,` Thomas P. Seager,§ and Igor Linkovjj

U.S. Army Engineer Research and Development Center, Environmental Laboratory, 3909 Halls Ferry Road, Vicksburg, Mississippi 39180
`ICF Consulting, 33 Hayden Avenue, Lexington, Massachusetts 02421, USA
§School of Civil Engineering, Purdue University, West Lafayette, Indiana 47905, USA

Review
jjCambridge Environmental, 58 Charles Street, Cambridge, Massachusetts 02141, USA

ABSTRACT
Decision making in environmental projects can be complex and seemingly intractable, principally because of the inherent
trade-offs between sociopolitical, environmental, ecological, and economic factors. The selection of appropriate remedial
and abatement strategies for contaminated sites, land use planning, and regulatory processes often involves multiple
additional criteria such as the distribution of costs and benefits, environmental impacts for different populations, safety,
ecological risk, or human values. Some of these criteria cannot be easily condensed into a monetary value, partly because
environmental concerns often involve ethical and moral principles that may not be related to any economic use or value.
Furthermore, even if it were possible to aggregate multiple criteria rankings into a common unit, this approach would not
always be desirable because the ability to track conflicting stakeholder preferences may be lost in the process. Consequently,
selecting from among many different alternatives often involves making trade-offs that fail to satisfy 1 or more stakeholder
groups. Nevertheless, considerable research in the area of multicriteria decision analysis (MCDA) has made available practical
methods for applying scientific decision theoretical approaches to complex multicriteria problems. This paper presents a
review of the available literature and provides recommendations for applying MCDA techniques in environmental projects. A
generalized framework for decision analysis is proposed to highlight the fundamental ingredients for more structured and
tractable environmental decision making.

Keywords: Decision criteria Environmental risk assessment Site remediation

INTRODUCTION prematurely adopt a common perspective that excludes

Environmental decisions are often complex and multi- contrary information. This tendency has been termed by
faceted and involve many different stakeholders with different some as ‘‘groupthink’’ (McDaniels et al. 1999).
priorities or objectives—presenting exactly the type of For environmental management projects, decision makers
problem that behavioral decision research has shown humans often receive 4 generalized types of technical input: the
are poorly equipped to solve unaided. Most people, when results of modeling and monitoring studies, risk assessment,
confronted with such problems, will attempt to use intuitive cost or cost-benefit analysis, and stakeholder preferences
or heuristic approaches to simplify the complexity until the (Figure 1a). However, current decision processes typically
problem seems more manageable. In the process, important offer little guidance on how to integrate or judge the relative
information may be lost, opposing points of view may be importance of information from each source. Also, informa-
discarded, and elements of uncertainty may be ignored. In tion comes in different forms. While modeling and monitor-
short, there are many reasons to expect that, on their own, ing results are usually presented as quantitative estimates, risk
individuals (either lay or expert) will often experience assessment and cost-benefit analyses may incorporate a higher
difficulty making informed, thoughtful choices in a complex degree of qualitative judgment by the project team. Struc-
decision-making environment involving value trade offs and tured information about stakeholder preferences may not be
uncertainty (McDaniels et al. 1999). presented to the decision maker at all and may be handled in
Moreover, environmental decisions typically draw on an ad hoc or subjective manner that exacerbates the difficulty
multidisciplinary knowledge bases, incorporating natural, of defending the decision process as reliable and fair. More-
physical, and social sciences and medicine, politics, and over, where structured approaches are employed, they may be
perceived as lacking the flexibility to adapt to localized
ethics. This fact, and the tendency of environmental issues
concerns or faithfully represent minority viewpoints. A
to involve shared resources and broad constituencies, means
systematic methodology to combine both quantitative and
that group decision processes are often necessary. These may
qualitative inputs from scientific or engineering studies of
have some advantages over individual processes; in particular,
risk, cost, and benefit, as well as stakeholder views and values
more perspectives may be put forward for consideration, the
to rank project alternatives, has yet to be fully developed for
probability of benefiting from the presence of natural
environmental decision making. As a result, decision makers
systematic thinkers is higher, and groups often learn to rely
are prevented from identifying all plausible alternatives and
on more deliberative, well-informed members. However,
from making full use of all available and necessary informa-
groups are also susceptible to the tendency to establish
tion in choosing between identified project alternatives.
entrenched positions (defeating compromise initiatives) or to
In response to current decision-making challenges, this
paper reviews the efforts of several government agencies and
* To whom correspondence may be addressed scientists to implement new concepts in decision analysis and
[email protected]
operations research for complex, environmental projects.
96 Integr Environ Assess Manag 1, 2005—G.A. Kiker et al.

Figure 1. Current and evolving decision-making processes for contaminated sediment management.

Recent (published over the past 10 y) literature on environ- risks and cost while maximizing public acceptance. Three
mental applications of multicriteria decision theory and remediation alternatives (A, B, and C) are identified for
regulatory guidance developed by the United States and consideration by stakeholders and/or project team. Criteria are
international agencies is summarized. The general trends in established to aid decision makers in judging the relative
the field are reflected in Figure 1b. Multicriteria decision strengths of the alternatives. For ecological risk, 2 criteria are
analysis (MCDA) tools can be applied to assess value selected, specifically, the number of complete exposure
judgments of individual decision makers or multiple stake- pathways (i.e., distinct exposure route that leads to a receptor)
holders. For individuals, risk-based decision analysis quantifies and the maximum calculated hazard quotient from all the
value judgments, scores different project alternatives on the pathways. For human health risk, 2 similar criteria are selected,
criteria of interest, and facilitates selection of a preferred including the number of complete human exposure pathways
course of action. For group problems, the process of and the maximum cancer risk calculated from the complete
quantifying stakeholder preferences may be more intensive, pathways. Decision makers use dollars per cubic yard of
often incorporating aspects of group decision making. sediment to represent the cost criteria. The impacted area (i.e.,
One of the advantages of an MCDA approach in group amount of surface or subsurface area required to manage the
decisions is the capacity for calling attention to similarities or sediment) is used as a measure of public acceptance.
potential areas of conflict between stakeholders with different Quantitative estimates for these criteria are developed
views, which results in a more complete understanding of the through research, monitoring, and survey studies or through
values held by others. The evolving decision-analytic ap- expert judgment elicitation. The resulting data are used to
proaches are applicable to the full range of environmental parameterize the decision table or matrix depicted in Figure 2.
projects, but for purposes of focusing subsequent discussion, A decision matrix in a form similar to Figure 2 is usually the
more specific attention is given to decision making at final product of feasibility studies for Superfund projects in
contaminated sites. This paper has 4 objectives: (1) to the United States and similar investigations conducted else-
summarize the most common MCDA methods in practice, where. Decisions are typically based on an informal, ad hoc
(2) to discuss the use of MCDA methods within selected comparison of the considered alternatives. For example, the
regulatory agencies in the United States and the European decision matrix developed in Figure 2 clearly shows that
Union, (3) to review the use of MCDA methods within alternative C is most efficient from risk reduction point of
selected environmental challenges, and (4) to identify the view but is also most costly. Depending on the decision
components of an appropriate decision-analytic framework. maker’s sensitivity to the budget, a decision maker may select
alternative C or a cheaper alternative B, which may still result
APPROACHES TO MCDA in a significant risk reduction.
Figure 2 illustrates decision dilemmas for a contaminated The methods of MCDA evolved as a response to the
sediment management project summarized from Kane Dris- observed inability of people to effectively analyze multiple
coll et al. (2002). The decision makers must select a manage- streams of dissimilar information. There are many different
ment alternative that minimizes human health and ecological MCDA methods, and a detailed analysis of the theoretical
Multicriteria Decision Making—Integr Environ Assess Manag 1, 2005 97

Figure 2. Example decision criteria and matrix.

foundations of these methods and their comparative strengths and criteria, a condition that is rarely characteristic of
and weaknesses is presented in Belton and Steward (2002). environmental projects.
The common purpose of MCDA methods is to evaluate and Multiattribute utility theory or multiattribute value theory
choose among alternatives based on multiple criteria using (MAUT/MAVT) and the analytical hierarchy process (AHP)
systematic analysis that overcomes the limitations of unstruc- are optimization methods. They employ numerical scores to
tured individual or group decision making. communicate the merit of 1 option in comparison to others
Almost all decision analysis methodologies share similar on a single scale. Scores are developed from the performance
steps of organization in the construction of the decision matrix. of alternatives with respect to an individual criterion and
Each MCDA methodology synthesizes the matrix information aggregated into an overall score. Individual scores may be
and ranks the alternatives by different means (Yoe 2002). simply summed or averaged, or a weighting mechanism can
be used to favor some criteria more heavily than others. The
Different methods require diverse types of value information
goal of MAUT/MAVT is to find a simple expression for
and follow various optimization algorithms. Some techniques
decision-maker preferences. Through the use of utility/value
rank options, some identify a single optimal alternative, some
functions, the MAUT method transforms the diverse criteria
provide an incomplete ranking, and others differentiate
in Figure 2 (such as cost, risks, and stakeholder acceptance)
between acceptable and unacceptable alternatives.
into 1 common dimensionless scale (typically 0–1) of utility
Within MCDA, elementary methods can be used to reduce or value. Utility functions for each criteria convert the criteria
complex problems to a singular basis for selection of a units into the 0-to-1 utility scale and are combined with
preferred alternative. Competing decision criteria may be weighting functions of the criteria within the overall decision
present, but intercriteria weightings are not required. For to form a decision score for each alternative. MAUT also relies
example, an elementary goal aspiration approach may rank on the assumptions that the decision maker is rational (e.g.,
the dredging alternatives in relation to the total number of more utility is preferred to less utility), preferences do not
performance thresholds met or exceeded. While the analysis change, and the decision maker has perfect knowledge and is
can, in most applications of elementary approaches, be consistent in his or her judgments. The goal of decision
executed without the help of computer software, these makers in this process is to maximize utility/value, which
methods are best suited for problems with few alternatives makes this a compensatory optimization approach.
98 Integr Environ Assess Manag 1, 2005—G.A. Kiker et al.

Table 1. Applications of decision support tools in environmental managementa

Application area Method Decision context Funding agency Citation

Prioritization of sites/ AHP þ GIS Land condition assess- U.S. Army Engineering Mendoza et al. (2002)
areas for industrial/ ment for allocation of Research and
military activity military training areas Development Center
AHP þ GIS Selection of boundaries International Institute Sharifi et al. (2002)
for national park for Geoinformation
Science and Earth
Observation, The
Netherlands
PROMETHEE Waste management Natural Sciences and Vaillancourt and Waaub
activities in Canada Engineering Research (2002)
Council of Canada
ELECTRE þ GIS Land management: Swiss National Found- Joerin and Musy (2000)
develop a land ation for Research
suitability map for (FNRS)
housing in Switzerland
AHP þ GIS Landfill sitting Siddiqui et al. (1996)
MAUT þ GIS Selection of park USDOE Keisler and Sundell
boundaries (1997)
Environmental/remedial SMART Choosing a remedial U.S. Army Corps of Wakeman (2003)
technology selection action alternative at Engineers
Superfund site
MAUT Selection of University of Missouri— Prato (2003)
management alternative Columbia, USA
Missouri River
MAUT þ AHP Regulation of water flow Academy of Finland Hamalainen et al. (2001)
in a lake–river system
MAUT Offsite emergency European Commission, Ehrhardt and Shersha-
management following Ukraine kov (1996); Hamalainen
a nuclear accident (such et al. (2000)
as the Chernobyl
accident)
Environmental impact Review Review of MCDA use for Vrije University, The Janssen (2001)
assessment EIAs in Netherlands Netherlands
AHP Socioeconomic impact Indira Gandhi Institute Ramanathan (2001)
assessment for a con- of Development
struction project in India Research, India
ELECTRE Highway environmental Dublin Institute of Rogers and Bruen
appraisal in Ireland Technology; University (1998)
College Dublin, Ireland
AHP and Environmental impact Finnish Environmental Marttunen and Hamalai-
MAUT/SMART assessment of 2 water Agency; Helsinki nen (1995)
development projects on University of
a Finnish river Technology
PROMETHEE Prioritization of EIAs in Staffordshire University, Al-Rashdan et al. (1999)
Jordan United Kingdom
Natural resource AHP Natural park USDA Forest Services Schmoldt et al. (1994);
management management Peterson et al. (1994);
Schmoldt and Peterson
(2001b)
AHP Management of small USDA Forest Services Rauscher et al. (2000)
forest in North Carolina,
USA
Multicriteria Decision Making—Integr Environ Assess Manag 1, 2005 99

Table 1. Continued

Application area Method Decision context Funding agency Citation

MAUT Management of spruce National Science and Levy et al. (2000)
budworm in Canadian Engineering Research
forests Council of Canada
AHP, MAUT, and Forestry planning in Finnish Academy of Kangas et al. (2001)
outranking Finland Sciences; Finnish Forest
Research Institute
MAUT Improvement of habitat Finnish Forest Research Store and Kangas (2001)
suitability measurements Institute
AHP Environmental USEPA /USDOE Tran et al. (2002)
vulnerability assessment
for mid-Atlantic region
Weighting Management of marine U.K. Department of Brown et al. (2001)
protected areas in International
Tobago Development
MAUT Fisheries management: Fisheries and Ocean, McDaniels (1995)
select among alternative Canada
commercial fishery
opening days
AHP, MAUT, and Fisheries management Mardle and Pascoe
outranking (1999)
a
PROMETHEE = Preference Ranking Organization METHod for Enrichment Evaluations; ELECTRE = Elimination Et Choix Traduisant la Realite;
AHP = analytical hierarchy process; GIS = geographic information system; MAUT = multiattribute utility theory; MCDA = multicriteria
decision analysis; EIA = environmental impact assessment; USDA = U.S. Department of Agriculture; USDOE = U.S. Department of Energy;
SMART = simple multiattribute rating technique.

Like MAUT, AHP is a compensatory optimization ap- easily aggregated, measurement scales vary over wide ranges,
proach. However, AHP uses a quantitative comparison and units are incommensurate or incomparable.
method that is based on pairwise comparisons of decision
criteria rather than utility and weighting functions. All REGULATORY USES OF MCDA
individual criteria must be paired against all others and the Decision process implementation is often based on the
results compiled in matrix form. In the contaminated sedi- results of physical modeling and engineering optimization
ment example, the AHP method would require the decision schemes. Even though federal agencies are required to
maker to answer questions such as, ‘‘With respect to the consider social and political factors, the typical decision
selection of a sediment alternative, which is more important, analysis process does not provide specifically for explicit
public acceptability or cost?’’ The user uses a numerical scale consideration of such issues. Comparatively little effort is
to compare the choices, and the AHP method moves applied to engaging and understanding stakeholder perspec-
systematically through all pairwise comparisons of criteria tives (including the general public as well as potentially
and alternatives. The AHP technique thus relies on the responsible parties and natural resource trustees) or to
supposition that humans are more capable of making relative provide for potential learning among stakeholders. A result
judgments than absolute judgments. Consequently, the of this weakness in current and common decision models is
rationality assumption in AHP is more relaxed than in MAUT. that the process tends to quickly become adversarial whereby
Unlike MAUT and AHP, outranking is based on the there is little incentive to understand multiple perspectives or
principle that one alternative may have a degree of dominance to share information. However, a review of regulatory and
over another (Kangas et al. 2001) rather than the supposition guidance documents reveals several programs in the United
that a single best alternative can be identified. Outranking States where regulatory agencies involved in environmental
models compare the performance of 2 (or more) alternatives issues are beginning to implement formal decision-analytic
at a time, initially in terms of each criterion, to identify the tools (such as MCDA) in their decision-making process.
extent to which a preference for one over the other can be
asserted without using a prescribed scale such as the AHP U.S. Army Corps of Engineers
method. In aggregating preference information across all Historically, the U.S. Army Corps of Engineers (USACE)
relevant criteria, the outranking model seeks to establish the has used essentially a single-measure approach to civil works,
strength of evidence favoring the selection of one alternative planning decisions through its Principles and Guidelines
over another, for example, by favoring the sediment alter- (P&G) framework (USACE 1983). The USACE has primarily
native that performs the best on the greatest number of used net national economic development (NED) benefits as
criteria. Therefore, outranking models are partially compen- the single measure to choose among different alternatives.
satory and most appropriate when criteria metrics are not The P&G method makes use of a complex analysis of each
100 Integr Environ Assess Manag 1, 2005—G.A. Kiker et al.

Table 2. Applications of decision support tools for stakeholder involvementa

Method Application format Decision context Funding agency Citation

MAUT Individual surveys Generic radiologically USDOE/NSF Arvai and Gregory (2003)
performed under contaminated site
supervision
MAUT Workshop and Consensus building for Social Sciences and McDaniels and Roessler
individual interviews water resource Humanities; Research (1998); Gregory et al.
management Council of Canada; NSF (2001)
MAUT Workshop Developing alternatives Social Sciences and Gregory and Keeney
for coal mine exploration Humanities Research (1994)
in Malaysia Council of Canada; NSF
MAUT Value elicitation in Elicitation of value Social Sciences and McDaniels and Roessler
workshop settings judgments for wilderness Humanities; Research (1998)
preservation Council of Canada; NSF
MAUT Workshop Developing alternatives NSF; USEPA Gregory and Wellman
for coal mine exploration (2001)
in Malaysia
MAVT Individual surveys Regional forest planning La Trobe University, Ananda and Herath
Australia (2003)
MAUT Individual surveys Air quality valuation in Korea University Kwak et al. (2001)
Korea
MAUT Individual surveys Risk attitudes by farmers European Union; Gomez-Limon et al.
in Spain Spanish government (2003)
AHP Workshop Forest fire management USDA Schmoldt and Peterson
(2001b)
MAUT Individual surveys Water use planning Gregory and Failing
(2002)
Mental Individual surveys, Watershed management USEPA Focht et al. (1999); Whi-
modeling workshop taker and Focht (2001)
Mental Individual surveys, Energy policy Gregory et al. (2003)
modeling workshop
a
MAUT = multiattribute utility theory; USDOE = U.S. Department of Energy; NSF = National Science Foundation; USEPA = U.S. Environ-
mental Protection Agency; USDA = U.S. Department of Agriculture; AHP = analytical hierarchy process.

alternative to determine the benefits and costs in terms of knowledge of the costs, benefits, impacts, and interactions is
dollars and other nondollar measures (e.g., environmental rarely precisely known. This single-number approach is
quality and safety); the alternative with the highest net NED limiting and may not always lead to an alternative or decision
benefit (i.e., with no environmental degradation) is usually process satisfactory to stakeholders.
selected. The USACE uses a variety of mechanistic and In response to a USACE request for a review of P&G
deterministic fate and transport models to provide informa- planning procedures, the National Research Council (NRC
tion in quantifying the various economic development and 1999) provided recommendations for streamlining planning
ecological restoration accounting requirements as dictated by processes, revising P&G guidelines, analyzing cost-sharing
P&G procedures. The level of complexity and scope
requirements, and estimating the effects of risk and un-
addressed by these models is determined at the project level
certainty integration in the planning process. As an integration
by a planning team. Issues such as uncertainty and risk are also
mechanism, the NRC (1999) recommended that further
addressed through formulation at the individual project
decision analysis tools be implemented to aid in the
management level.
While the P&G method is not specifically required for comparison and quantification of environmental benefits
planning efforts related to military installation operation and from restoration, flood damage reduction, and navigation
maintenance, regulatory actions, or operational or mainte- projects. In addition, new USACE initiatives, such as the
nance dredging, it presents a general decision approach that Environmental Operating Principles within USACE civil
influences many USACE decisions. The USACE planning works planning, dictate that projects adhere to a concept of
approach is essentially a monocriterion approach where a environmental sustainability that is defined as ‘‘a synergistic
decision is based on a comparison of alternatives using 1 or 2 process, whereby environmental and economic considerations
factors. Cost-benefit analysis, for example, is a monocriterion are effectively balanced through the life of project planning,
approach. The P&G approach has its challenges in that design, construction, operation and maintenance to improve
Multicriteria Decision Making—Integr Environ Assess Manag 1, 2005 101

the quality of life for present and future generations’’ determine, manage, and communicate risk, there is a great
(USACE 2003a, p. 5). need for comparative risk assessment tools, risk management
In addition, revised planning procedures have been decision trees, and risk communication tools that would allow
proposed to formulate more sustainable options through site managers to reach agreement with their regulators and
‘‘combined’’ economic development and ecosystem restora- other stakeholders while achieving mutual understanding of
tion plans (USACE 2003b). While still adhering to the overall the relationship between risk parameters, regulatory con-
P&G methodology, USACE (2003b) advises project delivery straints, and cleanup. Several USDOE models have been
teams to formulate acceptable, combined economic develop- developed specifically for dealing with radiologically con-
ment and ecosystem restoration alternatives using MCDA and taminated sites and sites with dual (chemical and radiological)
trade-off methods (Males 2002). Despite the existence of new contamination. Several of the current models are determin-
guidance and revisions on the application of MCDA istic, although probabilistic multimedia models have also been
techniques to environmental projects, there remains a need developed and used (USDOE 2003).
for a systematic strategy to implement these methods within Several USDOE guidance documents introduce decision-
specific USACE mission areas (e.g., navigation and restora- analytical tools and recommend their use. Technical guidance
tion) as well as linkage with existing risk analysis and adaptive developed for a wide range of USDOE decision needs (Baker et
management procedures. al. 2001) segregate the decision process into 8 sequential steps:
defining the problem, determining the requirements, estab-
U.S. Environmental Protection Agency lishing the goals of the project, identifying alternative methods
Stahl et al. (2002) and Stahl (2003) recently reviewed the and products, defining the criteria of concern, selecting an
decision analysis process in the U.S. Environmental Protection appropriate decision-making tool for the particular situation,
Agency (USEPA) and observed that although USEPA has a evaluating the alternatives against the criteria, and, finally,
mandate to make decisions in the public interest pertaining to validating solution(s) against the problem statement. Guid-
the protection of human health and welfare, there are barriers ance also focuses on how to select a decision-making tool from
in current USEPA decision processes that may discourage among 5 recommended evaluation methods. These methods
stakeholder participation, integration of perspectives, learning include pros-and-cons analysis, Kepner-Tregoe (K-T) decision
about new alternatives, and consensus building. Similar to the analysis, AHP, MAUT, and cost-benefit analysis.
USACE, the USEPA uses a variety of modeling tools to The USDOE has developed a standard paradigm for
support its current decision-making processes. The majority of selecting or developing a risk-based prioritization (RBP)
these tools are quantitative multimedia systems that assess system (USDOE 1998). The paradigm describes the issues
benefits and risks associated with each proposed alternative that should be considered when comparing, selecting, or
with the objective of selecting the best option (Stahl 2003). implementing RBP systems. It also identifies characteristics
Several USEPA guidance documents introduce decision- that should be used in evaluating the quality of a RBP system
analytic tools and recommend their use. Multicriteria inte- and its associated results. The USDOE (1998) recommends
grated resource assessment (MIRA) has been proposed as an the use of MAUT as an RBP model because it is a flexible,
alternative framework to existing decision analysis approaches quantitative decision analysis technique and management tool
at USEPA (Stahl et al. 2002; USEPA 2002; Stahl 2003). MIRA for clearly documenting the advantages and disadvantages of
policy choices in a structured framework. The MAUT merits
is a process that directs stakeholders to organize scientific data,
special consideration because it provides sound ways to
establishes links between the results produced by the research
combine quantitatively dissimilar measures of costs, risks,
community, and organizes applications in the regulatory
and benefits, along with decision-maker preferences, into
community. MIRA utilizes AHP-based trade-off analysis to
high-level, aggregated measures that can be used to evaluate
determine the relative importance of decision criteria.
alternatives. The MAUT allows full aggregation of perform-
Multiattribute product evaluation is inherent in the nature
ance measures into 1 single measure of value that can be used
of life-cycle assessment, which has rapidly emerged as a tool to
for ranking alternatives. However, USDOE (1998) cautions
analyze and assess the environmental impacts associated with a
that the results of MAUT analysis should not normally be used
product, process, or service (Miettinen and Hamalainen 1997;
as the principal basis for decision making because decision
Seppala et al. 2002). Further, the USEPA has developed the
making will generally require accounting for factors that
‘‘Framework for Responsible Environmental Decision-Mak-
cannot be readily quantified (e.g., equity). Furthermore,
ing’’ to assist the Office of Pollution Prevention and Toxics in
USDOE (1998) guidance states that no technique can
their development of guidelines for promoting the use of
eliminate the need to rely heavily on sound knowledge, data,
environmentally preferable products and services (USEPA
and judgments or the need for a critical appraisal of results.
2000). The ‘‘Framework for Responsible Environmental
The USDOE has used a multiattribute model as the core of
Decision-Making’’ decision-making method provides a foun- its Environmental Restoration Priority System for prioritizing
dation for linking life cycle indicator results with technical and restoration projects developed in the late 1980s (Jenni et al.
economic factors for decision makers when quantifying the 1995). Although the Environmental Restoration Priority
environmental performance of competing products. System was designed to operate with any specified set of
values and trade offs, its use was limited to values that were
U.S. Department of Energy
elicited from USDOE managers, including those based on risk
Similar to the USACE and USEPA, the U.S. Department of analysis. The USDOE has not applied the Environmental
Energy (USDOE) uses a variety of multimedia models to Restoration Priority System because of stakeholder opposition,
support its decision-making process. A recent review (Cor- although similar decision support systems have since been
porate Project 7 Team 2003) concluded that even though adopted for use at various USDOE sites (CRESP 1999). The
there are a significant number of guidance documents, USDOE has attempted to use simple weighting to aid program
systems, and processes in use within the USDOE to planning and budget formulation processes (CRESP 1999).
102 Integr Environ Assess Manag 1, 2005—G.A. Kiker et al.

European Union described in this study (e.g., MAUT, AHP, and outranking)
A detailed review of the regulatory background and use of have been used, in conjunction with geographic information
decision-analytic tools in the European Union (EU) was system, for selection of site boundaries and the identification of
recently conducted within the EU-sponsored Contaminated geographical areas for related uses (e.g., industrial or military).
Land Rehabilitation Network for Environmental Technologies Mendoza et al. (2002) used AHP for allocating areas for
project (Bardos et al. 2002). The review found that environ- military training exercises at Ford Hood, Texas, USA. Keisler
mental risk assessment, cost-benefit analysis, life cycle assess- and Sundell (1997) and Sharifi et al. (2003) proposed a
ment, and MCDA were the principal analytical tools used to framework that integrates MAUT and spatial analysis to
support environmental decision making for contaminated determine national park boundaries. Joerin and Musy (2000)
land management in 16 EU countries (Austria, Belgium, developed a generic method to integrate multiple consider-
Denmark, Finland, France, Germany, Greece, Ireland, Italy, ations, such as impacts, air quality, noise, accessibility, climate,
the Netherlands, Norway, Portugal, Spain, Sweden, Switzer- utility networks (e.g., water, electricity), and aesthetics
land, and the UK). Similar to the United States, quantitative related to land management. Vaillancourt and Waaub
methods such as risk assessment and cost-benefit analysis are (2002) used outranking and a geographic information system
presently the dominant decision support approaches, while framework to select a site for a new waste management
MCDA and explicit trade offs are used less frequently. facility in Montreal, Quebec, Canada.
Pereira and Quintana (2002) reviewed the evolution of Environmental/remedial technology selection—The selection
decision support systems for environmental applications of a feasible remedial action is usually the final stage of a
developed by the EU Joint Research Center. The concept of contaminated site investigation (e.g., as required under the
environmental decision support has evolved from highly U.S. Superfund program). This review identified several
technocratic systems aimed at improving understanding of instances in which MCDA methods were used to select the
technical issues by individual decision makers to a platform best technology or remedial method.
for helping all parties involved in a decision process engage in For example, a MAUT-based method was applied to
meaningful debate. Applications developed in the group compare current and alternative water control plans in the
include water resources management, siting of waste disposal Missouri River, USA (Prato 2003). A related problem of
plants, hazardous substance transportation, urban transporta- regulating water flow in a river–lake system was addressed by
tion, management, and groundwater management. Hamalainen et al. (2001) from the perspectives of group
decision theory and stakeholder consensus building. Wake-
MCDA APPLICATIONS FOR ENVIRONMENTAL man (2003) used the simple multiattribute rating technique
MANAGEMENT AND RELATED USES (SMART) (Edwards 1977) to decide which action alternative
The MCDA applications are relevant to environmental to implement in handling the contaminated river sediment at
management, stakeholder involvement, and the management Milltown Dam, Montana, USA. Factors considered by
of contaminated lands. Recent publications present more Edwards (1977) included availability of materials and services,
comprehensive reviews of studies relevant to management of ability to construct, and reliability. One of the most advanced
terrestrial sites (Linkov et al. 2004) and contaminated applications of MCDA techniques in this area was imple-
sediments (Linkov et al. 2005). The use of MCDA is more mented for nuclear accident emergency management as a part
strongly evident within the broad areas of environmental of the EU-RODOS project, which used a MAUT analysis for
management and stakeholder involvement. Fewer efforts have strategy selection for population protection after a nuclear
been made to apply MCDA to the management of accident (Ehrhardt and Shershakov 1996).
contaminated lands and risk analysis. It should be noted that Environmental impact assessment—Environmental impact
MCDA has also been applied in many other related policy assessments (EIA) are routinely conducted for all major
development areas, such as manufacturing and services; projects in the United States with the potential to affect the
medical, military, and public policy (Keefer et al. 2002a, environment. The assessment of site contamination is often an
2002b); climate change (Bell et al. 2003); industrial facility integral part of EIA. Janssen (2001) reviewed 21 EIAs
siting (Larichev and Olson 2001); energy policy (Hobbs and conducted in the Netherlands in the period 1992–2000. Most
Meier 2000; Keefer et al. 2002a, 2002b); agricultural resource of the EIAs reviewed by Janssen (2001) used weighted
management (Hayashi 2000); and life-cycle assessment summation methods, although a few projects used either the
(Seppala et al. 2002). AHP or a MAUT-based approach.
Marttunen and Hamalainen (1995) reviewed MAUT/
Application to general environmental management SMART and the AHP methods used for decision analysis in
The MCDA methods have been extensively applied to a EIAs for the assessment of environmental impacts of a water
range of environmental management challenges (Table 1). development project in Finland. SMART was chosen over
Each of the examples identified in the course of this review AHP because the AHP procedure proved to be too time
were classified into 1 of 5 application areas: (1) prioritization consuming for stakeholders (Marttunen and Hamalainen
of site/areas for industrial/military activity, (2) environ- 1995). Ramanathan (2001) recommended the use of AHP
mental/remedial technology selection, (3) environmental for considering multiple criteria and multiple stakeholders in
impact assessment, (4) stakeholder involvement, and (5) EIA as well as to assess the socioeconomic impact of a
natural resource planning. proposed liquefied petroleum gas recovery plant in an
Prioritization of sites/areas for industrial/military activity— industrial area in India. Rogers and Bruen (1998) used
Management of contaminated sites often requires site zoning Elimination Et Choix Traduisant la Realite (ELECTRE) III
for remediation, restoration, or other uses. Even though (outranking) methodology in evaluating thresholds for noise
applications of MCDA methods for contaminated site zoning impacts from a highway project in Ireland. Al-Rashdan et al.
could not be found in this review, the MCDA methods (1999) used Preference Ranking Organization METHod for
Multicriteria Decision Making—Integr Environ Assess Manag 1, 2005 103

Enrichment Evaluations (PROMETHEE) (outranking) meth- Arvai and Gregory (2003) was the only study identified
odology to rank environmental impact assessments related to that addressed the application of decision-analytic tools to
wastewater projects in Jordan; the methodology was found to include stakeholder involvement at contaminated sites (Table
be very useful in solving problems with conflicting criteria. 2). Arvai and Gregory (2003) compared 2 approaches for
Natural resource management—The management of natural involving stakeholders in identifying radioactive waste clean-
resources has involved the application of MCDA. Steiguer et up priorities at USDOE sites, (1) a traditional approach that
al. (2003) developed an annotated bibliography that includes involved communication of scientific information that is
124 examples of the application of MCDA to projects ranging currently in use in many USDOE, USEPA, and other U.S.
from theoretical studies to real-world forest and natural federal programs and (2) a values-oriented communication
resource management situations. Steiguer et al. (2003) approach that helped stakeholders make difficult trade-offs
indicate that MCDA constitutes a newer and, perhaps, more across technical and social concerns. The 2nd approach has
acceptable method for quantifying and evaluating public strong affinity to the MAUT-based trade-offs discussed earlier
preferences. Nevertheless, few studies included empirical in this paper. Arvai and Gregory (2003) concluded that the
testing of MCDA utility or its feasibility and, in most of the incorporation of value-based trade-offs information leads
studies, researchers have used hypothetical data or, at best, stakeholders to making more informed choices.
simplified decision situations; few studies were designed to Table 2 summarizes several other representative stake-
implement an MCDA-generated management strategy. holder involvement studies in the areas related to manage-
The AHP approach within MCDA has received the most ment of natural resources and technology selection. This
attention in natural resource management applications (Stei- review addresses studies that involve local communities at
guer et al. 2003). The application of AHP in natural resource action-specific levels rather than broad-based public involve-
planning is summarized in Schmoldt et al. (2001) with Table 1 ment efforts.
listing some representative publications. Schmoldt et al. Several studies propose the use of MCDA tools for
(1994) and Schmoldt and Peterson (2001b) used AHP to consensus building and advocate the utility of this application
address different aspects of natural park management, includ- or illustrate the value-oriented approaches that are based on
ing developing inventory and monitoring programs, as well as MAUT. In general, applications may include individual
strategic management plans. Pavlikakis and Tsihrintzis (2003) surveys and workshops designed to elicit value judgment
evaluated the utility of MAUT and AHP in selecting a and construct decision alternatives. Specific applications
technically suitable and socially acceptable management plan include water resource management (McDaniels et al. 1999;
for a national park in eastern Macedonia and Thrace in Greece. Gregory et al. 2001), mining (Gregory and Keeny 1994),
Methods of MCDA have been extensively applied to a wide wilderness preservation (McDaniels and Roessler 1998), and
range of projects in forest management; AHP was applied for estuary management (Gregory and Wellman 2001). The
a project-scale forest management problem by Rauscher et al. McDaniels et al. (1999) study concludes that value-based
(2000), and MAUT analysis was applied to identify policy approaches result in a higher level of comfort for participants
alternatives to manage a budworm outbreak in a local site in and are useful in developing consensus-based management
Canada (Levy et al. 2000). Kangas et al. (2001) tested the decisions. The MAUT-based applications appear to be used in
application of several MAUT and outranking methods for stakeholder value elicitation for regional forest planning
large-scale forest policy planning in Finland. Store and Kangas (Ananda and Herath 2003), air quality valuation (Kwak et
(2001) used MAUT-based methods to conduct a habitat al. 2001), and agricultural applications (Gomez-Limon et al.
suitability evaluation over large forested areas. Finally, Tran et 2003). In addition, Schmoldt and Peterson (2001a) advocated
al. (2002) used AHP to assess the environmental vulnerability the use of AHP as a decision support tool in workshop settings
of forests across the mid-Atlantic region in the United States. for forest resource management.
The MCDA has also been applied to manage aquatic The examples presented previously used MCDA to
resources. Simon and Pascoe (1999) reviewed applications of facilitate consensus building. An alternative application of
MCDA in fisheries management. Brown et al. (2001) used MCDA is in the organization of diverse interests instead of
weighting-based trade-off analysis to select a management seeking consensus-based middle ground. Gregory and Failing
option for Buccoo Reef Marine Park in Tobago; criteria (2002) argue that a clear expression of difference facilitates
evaluated included ecological, social, and economic factors. development and acceptance of management plans. Another
McDaniels (1995) used a MAUT approach to select among approach to ranking risk involves soliciting the views of
alternative commercial fishery openings involving conflicting participants both as individuals and in a group setting
long-term objectives for salmon management. (Morgan et al. 2000; Florig et al. 2001). In this manner,
decision makers can obtain information on the rankings of
Application to stakeholder involvement options that involve multiple objectives by weighing the
Most of the examples presented here attempt to represent attributes identified by individuals and groups developed
the value judgments of a single decision maker and from the 2 methods.
incorporate these value judgments into the overall decision- Mental modeling (Morgan et al. 2002) may be a promising
making process. Stakeholder values are often considered as 1 tool for assessing individual judgments. It involves individual,
attribute, along with others, such as costs or risk reduction. 1-on-1 interviews, leading participants through a jointly
The MCDA can also be used as a framework that permits determined agenda of topics. The method allows free
stakeholders to structure their views about the pros and cons expression and encourages elaboration on topics in order to
of different environmental and remedial management options. reveal individual perspectives at considerable depth. When
Applications of MCDA for group decision making in other effectively done, analysts can identify what people believe and
areas have been reviewed by Bose et al. (1997) and why they believe it. They are also able to compare analyses
Matsatsinis and Samaras (2001). over time and provide insights into why beliefs change.
104 Integr Environ Assess Manag 1, 2005—G.A. Kiker et al.

Table 3. Applications of decision support tools for contaminated sitesa

Funding
Method Site type Decision context Criteria agency Citation
AHP þ Hazardous chemical Selection of remedial Programmatic, life cycle USDOE Accorsi (1999a, 1999b);
MAUT þ waste landfill action alternative cost, socioeconomic im- Bonano et al. (2000);
fuzzy located at a U.S. pact, environmental, hu- Apostolakis (2001)
national laboratory man health and safety
AHP þ Savannah River Site Optimal budget/ Cost of remediation, USDOE Deschaine et al. (1998)
linear (nuclear and resource allocation mortgage cost reduc-
programming chemical industries) for remediation tion, technical feasibility,
annual funding con-
straints
MAUT INEEL subsurface Selection of remedial Implementability, short- USDOE Grelk (1997);
disposal area action alternative term effectiveness, long- Parnell et al. (2001)
term effectiveness, re-
duction of toxicity/mobi-
lity/volume, cost
MAUT INEEL landfill Selection of remedial Cost, time, and risk USDOE Ralston et al. (1996)
technology
MAUT INEEL landfill Selection of remedial Risk of successful devel- USDOE Timmerman et al.
technology opment and risk of suc- (1996)
cessful field implemen-
tation
AHP Contaminated Selection of Resource requirements, USDOD Hartman and Goltz
USDOD installations brownfield data quality, method (2001)
in Korea management limitations, compliance
technologies with policy, input re-
quirements, and output
a
AHP = analytical hierarchy process; USDOE = U. S. Department of Energy; MAUT = multiattribute utility theory; INEEL = Idaho National
Environmental Engineering Laboratory; USDOD = U. S. Department of Defense.

Environmental applications of mental modeling include the lowest level of failure or development risks. Deschaine et
management of the Illinois River basin in eastern Oklahoma, al. (1998) used a MCDA simulation model based on AHP to
USA, (Focht et al. 1999; Whitaker and Focht 2001) and in select the most promising remediation projects from a 114
energy policy development (Gregory et al. 2003). radiological site remediation portfolio at the USDOE
Savannah River Site. Accorsi et al. (1999a, 1999b), Bonano
Application to management of contaminated lands et al. (2000), and Apostolakis (2001) developed a method-
Review of the recent literature reveals few studies that use ology that uses AHP, influence diagrams, MAUT, and risk
MCDA techniques to facilitate decision making for the assessment techniques to integrate the results of advanced
management of contaminated sites (Table 3). The absence impact evaluation techniques with stakeholder preferences.
of more examples of the application of MCDA at contami- Most of the studies presented in Table 3 focus on evaluation
nated sites emphasizes the need to provide formal training on of technical risk and comparison of alternative technologies;
complex decision-making analysis. environmental risk assessment and stakeholder opinions were
Most applications of MCDA have been conducted by not usually quantified.
USDOE to develop decision models to evaluate specific
criteria for the selection of remediation technologies. Grelk SYNTHESIS OF DECISION-MAKING CONCEPTS
(1997), Grelk et al. (1998), and Parnell et al. (2001) have Successful environmental decision making in complex
developed a decision analysis value model that is based on the settings will depend on the extent to which 3 key components
process required by the legislation in the United States are integrated within the process: people, process, and tools.
supporting the Superfund program. The USDOE has also Based on this review of MCDA concepts and applications, a
sponsored a series of studies designed to develop decision systematic decision framework is proposed in Figure 3. This
models used to perform analysis of remedial alternatives for a framework is intended to give a generalized road map to the
mixed-waste subsurface disposal site at Idaho National environmental decision process.
Environmental Engineering Laboratory, USA. Ralston et al. The correct combination of people is the 1st essential
(1996) developed a generic model that incorporates life cycle element to the overall decision process shown in Figure 3.
cost and technological risk assessment for landfill waste site The involvement of 3 groups of stakeholders (i.e., decision
remediation. Timmerman et al. (1996) proposed the use of makers, scientists and engineers, and the general public) are
MAUT by USDOE for selecting technology judged to pose symbolized by dark lines for direct involvement and dotted
Multicriteria Decision Making—Integr Environ Assess Manag 1, 2005 105

Figure 3. Synthesis of decision-making ingredients.

lines for less direct involvement. While the actual member- should be given to designing an adaptable structure such that
ship and the function of these 3 base groups may intersect or participants can modify aspects of the project to suit local
vary, the roles of each group are essential to collecting the concerns while still producing a structure that provides the
largest amount of information to support the decision process. required outputs. The process depicted in Figure 3 follows 2
Each group has its own way of viewing the world, its own basic themes: (1) generating management alternatives, success
method of envisioning solutions, and its own societal criteria, and value judgments and (2) ranking the alternatives
responsibility. Policymakers and decision makers spend most by applying the criteria levels and value weights. The 1st part
of their effort defining the problem context and the overall of the process generates and defines choices, success levels,
constraints to the decision. In addition, they may have and preferences. The latter section methodically prunes
responsibility for the selection of the final decision and its nonfeasible alternatives by 1st applying screening mechanisms
implementation. Stakeholders may provide input to defining (e.g., overall cost, technical feasibility, and general societal
the problem but have the highest degree of interaction in acceptance) followed by more detailed ranking of the
helping formulate success criteria and contributing value remaining options using decision analysis techniques (e.g.,
judgments for weighting the various success criteria. Depend- AHP, MAUT, and outranking) that adopt the various criteria
ing on the problem and regulatory context, stakeholders may levels generated by environmental tools, monitoring, or
have some responsibility in ranking and selecting the final stakeholder surveys.
option. Scientists and engineers have the most focused role As shown in Figure 3, the tools used within group decision
because they provide the measurements or estimations of the making and scientific research are essential elements of the
desired criteria that determine the success of various overall decision process. Similar to people, the applicability of
alternatives. While scientists and engineers may take a the various tools is symbolized by solid lines (representing
secondary role as stakeholders or decision makers, their direct, or high, utility) and dotted lines (representing indirect,
primary role is to provide the technical details as requested or lower, utility). Decision analysis tools help generate and
by the decision process. guide the preferences of stakeholder groups, as well as
In Figure 3, the framework places process in the center of individual value judgments, into organized structures that can
the overall decision process. While it is reasonable to expect be linked with the other technical tools from risk analysis,
that the decision-making process may vary in specific details modeling/monitoring, and cost estimations. The decision
among regulatory programs and project types, emphasis analysis software also provides useful graphical techniques
106 Integr Environ Assess Manag 1, 2005—G.A. Kiker et al.

and visualization methods to express the gathered information way that facilitates a decision-making process involving risk,
in understandable formats. When changes occur in the multiple criteria, and conflicting interests. The MCDA
requirements or decision process, decision analysis tools can visualizes trade offs among multiple conflicting criteria and
respond efficiently to reprocess and iterate with the new inputs. quantifies the uncertainties necessary for comparison of
The framework depicted in Figure 3 provides a focused role available remedial and abatement alternatives. This process
for the detailed scientific and engineering efforts invested in helps technical project personnel, as well as decision makers
experimentation, environmental monitoring, and modeling, and stakeholders, to systematically consider and apply value
which, together, provide the rigorous and defendable details judgments to derive a most favorable management alternative.
for evaluating criteria performance under various alternatives. The MCDA also provides methods for participatory decision
This symbiotic relationship between decision and scientific or making in which stakeholder values are elicited and explicitly
engineering tools allows each of the 3 components to have a incorporated into the decision process.
unique and valued role in the decision process. Different MCDA methods have strengths and limitations.
As with most other decision processes, it is assumed that No matter which analytical decision tool is selected, imple-
the framework presented in Figure 3 is iterative at each phase mentation requires complex, often impossible trade offs. This
throughout the course of complex decision making. A 1st- complexity is probably one of the main reasons why MCDA is
pass effort may efficiently point out challenges that may still not widely used in practical applications. However, explicit
occur, key stakeholders to be included, or modeling and and structured approaches will often result in a more efficient
analysis studies that should be initiated. As the challenges and effective decision process as compared with the often
become more apparent, the process repeats itself again to intuition- and bias-driven decision processes that regulatory
explore and adapt to more subtle aspects of possible decisions agencies are often accused of using in decision making.
and their outcomes. Formal applications of MCDA in the management of
contaminated sites are rare at present. Applications in related
SUMMARY AND PATH FORWARD areas are more numerous; however, to date, MCDA has
Effective environmental decision making requires an remained largely an academic exercise with some exceptions
explicit structure for coordinating joint consideration of the involving the use of AHP-based methods in natural resource
environmental, ecological, technological, economic, and planning. Nevertheless, the positive results reported in the
sociopolitical factors relevant to evaluating and selecting studies reviewed in this paper, as well as the availability of
among management alternatives. Each of these factors recently developed software tools such as Expert Choice
includes multiple subcriteria, making the process inherently (www.expertchoice.com), Criterim Decision Plus (www.
multiobjective. Integrating this heterogeneous information infoharvest.com), and Decision Lab (www.visualdecision.
with respect to human aspirations and technical applications com), provide more than an adequate basis for recommending
demands a systematic and understandable framework to the use of MCDA in contaminated site management.
organize the people, processes, and tools for making a Environmental decision making also involves complex trade
structured and defensible decision. offs between divergent criteria. The traditional approach to
Stakeholder involvement is increasingly recognized as an environmental decision making involves valuing multiple
essential element of successful environmental decision mak- criteria based on a common unit, usually monetary, and
ing. The challenge of capturing and organizing that involve- thereafter performing standard mathematical optimization
ment as structured inputs to decision making alongside the procedures. Extensive scientific research in the area of
results of scientific and engineering studies and cost analyses decision analysis has exposed several weaknesses in this
can be met through application of the tools reviewed in this approach (Belton and Steward 2002). At the same time,
paper. The current environmental decision-making context new methods that facilitate a more rigorous analysis of
limits stakeholder participation within the decide-and-defend multiple criteria have been developed. These methods,
paradigm that positions stakeholders as constraints to be collectively known as MCDA methods, are increasingly being
tested rather than the source of core values that should drive adopted in environmental decision making.
the decision-making process. Consequently, potentially con- In summary, this paper reviewed MCDA methods cur-
troversial alternatives are eliminated early, and little effort is rently in use and cited numerous environmental applications
devoted to maximizing stakeholder satisfaction with either of these methods. While MCDA offers demonstrable
the decision process or the outcome. Instead, the final advantages, choosing among MCDA methods is a complex
decision may be something to which no one objects too task. Each method has strengths and weaknesses; while some
strenuously. Ultimately, this process does little to serve the methods are better grounded in mathematical theory, others
needs or interests of stakeholders who must live with the may be easier to implement. The availability to evaluate and/
consequences of an environmental decision. or support a decision also may act as a constraint on method
The increasing volume of complex and, often, controversial of decision analysis. It is unavoidable that the decision maker
information generated to support environmental decisions will have to choose, on a case-by-case basis, the most suitable
and the limited capacity of any 1 individual decision maker to MCDA technique applicable for different situations.
integrate and process that information emphasizes the need Finally, this paper identified the components necessary to
for developing tractable methods for aggregating the infor- develop a decision analysis framework, to facilitate the
mation in a manner consistent with the values of the decision selection of an MCDA process, and thereafter to provide
maker. The field of MCDA includes methods that can help guidance on the implementation of the selected MCDA
develop a decision-analytic framework useful for environ- method within the larger context of the people, processes,
mental management, including the management of contami- and tools used in decision making. The extensive growth over
nated sites. The purpose of MCDA is not always to single out the past 30 y in the amount and diversity of information
the correct decision but to help improve understanding in a required for environmental decision making has exceeded the
Multicriteria Decision Making—Integr Environ Assess Manag 1, 2005 107

capacity of common, unstructured decision models. Focused 2001. A deliberative method for ranking risks (I): Overview and test bed
effort directed at integrating MCDA principles and tools with development. Risk Analysis 21:913–922.
Focht W, DeShong T, Wood J, Whitaker K. 1999. A protocol for the elicitation of
existing approaches, including the use of risk and cost-benefit
stakeholders’ concerns and preferences for incorporation into policy dialogue.
analysis, will lead to more effective, efficient, and credible
In: Proceedings of the third workshop in the environmental policy and economics
decision making. workshop series: Economic research and policy concerning water use and
Acknowledgement—The authors are grateful to Cooke, watershed management. Washington DC: Environmental Law Institute. p 1–24.
Small, Valverde, Gloria, Yoe, Florig, Sullivan, Dortch, Dyer, Gomez-Limon JA, Arriaza M, Riesgo L. 2003. An MCDM analysis of agricultural risk
and Gardner for their review and suggestions. Special thanks aversion. European Journal of Operational Research 151:569–585.
to Stahl for making her dissertation available. Support for this Gregory R, Failing L. 2002. Using decision analysis to encourage sound
study was provided by the USACE Hazard/Risk Focus Area deliberation: Water use planning in British Columbia, Canada. Journal of
of the Environmental Quality Technology Program and the Policy Analysis and Management 21:492–499.
Center for Contaminated Sediments. Additional support was Gregory R, Fischhoff B, Thorne S, Butte G. 2003. A multi-channel stakeholder
consultation process for transmission deregulation. Energy Policy 31:1291–1299.
provided by the National Oceanic and Atmospheric Admin-
Gregory R, Keeney RL. 1994. Creating policy alternatives using stakeholder values.
istration through the Cooperative Institute for Coastal and
Management Science 40:1035–1048.
Estuarine Environmental Technology. Permission was granted Gregory R, McDaniels T, Fields D. 2001. Decision aiding, not dispute resolution:
by the USACE Chief of Engineers to publish this material. Creating insights through structured environmental decisions. Journal of
Policy Analysis and Management 20:415–432.
REFERENCES Gregory R, Wellman K. 2001. Bringing stakeholder values into environmental
Accorsi R, Apostolakis GE, Zio E. 1999a. Prioritizing stakeholder concerns in policy choices: A community-based estuary case study. Ecological Economics
environmental risk management. Journal of Risk Research 2:11–29. 39:37–52.
Accorsi R, Zio E, Apostolakis GE. 1999b. Developing utility functions for Grelk BJ. 1997. A CERCLA-based decision support system for environmental
environmental decision-making. Prog Nucl Energy 34:387–411. remediation strategy selection [thesis]. Colorado Springs (CO): U.S. Depart-
Al-Rashdan D, Al-Kloub B, Dean A, Al-Shemmeri T. 1999. Theory and method- ment of the Air Force, Air University, Air Force Institute of Technology.
ology environmental impact assessment and ranking the environmental Grelk B, Kloeber JM, Jackson JA, Deckro RF, Parnell GS. 1998. Quantifying CERCLA
projects in Jordan. European Journal of Operational Research 118:30–45. using site decision maker values. Remediation 8:87–105.
Ananda J, Herath G. 2003. Incorporating stakeholder values into regional forest Hamalainen RP, Kettunen E, Ehtamo H. 2001. Evaluating a framework for multi-
planning: A value function approach. Ecological Economics 45:75–90. stakeholder decision support in water resources management. Group Decision
Apostolakis GE. 2001. Assessment and management of environmental risks. In: and Negotiation 10:331–353.
Linkov I, Palma-Oliveira J, editors. Assessment and management of environ- Hamalainen RP, Lindstedt M, Sinkko K. 2000. Multi-attribute risk analysis in nuclear
mental risks. Boston (MA), USA: Kluwer. p 211–220. emergency management. Risk Analysis 20:455–468.
Arvai J, Gregory R. 2003. Testing alternative decision approaches for identifying Hartman DH, Goltz MN. 2001. Application of the analytic hierarchy process to select
cleanup priorities at contaminated sites. Environ Sci Technol 37:1469–1476. characterization and risk-based decision-making and management methods
Baker D, Bridges D, Hunter R, Johnson G, Krupa J, Murphy J, Sorenson K. 2001. for hazardous waste sites. Environmental Engineering and Policy 3:1–7.
Guidebook to decision-making methods. Washington DC: U.S. Department of Hayashi K. 2000. Multi-criteria analysis for agricultural resource management: A
Energy. WSRC-IM-2002-00002. critical survey and future perspectives. European Journal of Operational
Bardos P, Lewis A, Nortcliff S, Matiotti C, Marot F, Sullivan T. 2002. CLARINET Research 122:486–500.
report: Review of decision support tools for contaminated land management, Hobbs BF, Meier P. 2000. Energy decisions and the environment: A guide to the
and their use in Europe. Vienna: Austrian Federal Environment Agency. use of multi-criteria methods. Boston (MA), USA: Kluwer.
Bell M, Hobbs BF, Ellis H. 2003. The use of multi-criteria decision-making methods Janssen R. 2001. On the use of multi-criteria analysis in environmental impact
in the integrated assessment of climate change: Implications for IA assessment in the Netherlands. Journal of Multi-Criteria Decision Analysis
practitioners. Socio-Economic Planning Sciences 37:289–316. 10:101–109.
Belton V, Steward T. 2002. Multiple criteria decision analysis: An integrated Jenni KE, Merkhofer MW, Williams C. 1995. The rise and fall of a risk-based priority
approach. Boston (MA), USA: Kluwer. system: Lessons from DOE’s environmental restoration priority system. Risk
Bonano EJ, Apostolakis GE, Salter PF, Ghassemi A, Jennings S. 2000. Application of Analysis 15:397–410.
risk assessment and decision analysis to the evaluation, ranking and selection Joerin F, Musy A. 2000. Land management with GIS and multi-criteria analysis.
of environmental remediation alternatives. J Hazard Mater 71:35–57. International Transactions in Operational Research 7:67–78.
Bose U, Davey AM, Olson DL. 1997. Multi-attribute utility methods in group Kane Driscoll SB, Wickwire WT, Cura JJ, Vorhees DJ, Butler CL, Moore DW, Bridges
decision-making: Past applications and potential for inclusion in GDSS. TS. 2002. A comparative screening-level ecological and human health risk
International Journal of Management Sciences 25:691–706. assessment for dredged material management alternatives in New York /New
Brown B, Neil-Adger W, Tompkins E, Bacon P, Shim D, Young K. 2001. Trade-off Jersey Harbor. International Journal of Human and Ecological Risk Assessment
analysis for marine protected area management. Ecological Economics 8:603–626.
37:417–434. Kangas J, Kangas A, Leskinen P, Pykalainen J. 2001. MCDM methods in strategic
Corporate Project 7 Team. 2003. Assessment report. Corporate project 7: A planning of forestry on state-owned lands in Finland: Applications and
cleanup program driven by risk-based end states. Washington DC: U.S. experiences. Journal of Multi-Criteria Decision Analysis 10:257–271.
Department of Energy. Keefer D, Kirkwood CW, Corner JL. 2002a. Perspective on decision analysis
[CRESP] Consortium for Risk Evaluation with Stakeholder Participation. 1999. Peer applications, 1990–2001. Decision Analysis 1: (online supplement).
review of the U.S. Department of Energy use of risk in its prioritization process. Keefer D, Kirkwood CW, Corner JL. 2002b. Summary of decision analysis applications
New Brunswick (NJ), USA: CRESP. in the operations research literature, 1990–2001. Tucson (AZ), USA: Technical
Deschaine LM, Breslau B, Ades MJ, Selg RA, Saaty TL. 1998. Decision support Report Department of Supply Chain Management, Arizona State University.
software to optimize resource allocation: Theory and case history. Society for Keisler JM, Sundell RC. 1997. Combining multi-attribute utility and geographic
Computer Simulation, Simulators International XV. Boston (MA), USA: Society information for boundary decisions: An application to park planning. Journal
for Modeling Simulations. p 139–144. of Geographic Information and Decision Analysis 1:101–118.
Edwards W. 1977. How to use multi-attribute utility measurement for social Kwak SJ, Yoo SH, Kim TY. 2001. A constructive approach to air-quality valuation in
decision-making. IEEE (Inst Electr Electron Eng) Trans Biomed Eng 7:326–340. Korea. Ecological Economics 38:327–344.
Ehrhardt J, Shershakov VM. 1996. Real-time on-line decision support systems Larichev OI, Olson DI. 2001. Multiple criteria analysis in strategic siting problems.
(RODOS) for off-site emergency management following a nuclear accident. Boston (MA), USA: Kluwer.
Final Report. Luxemburg: European Commission. Levy J, Hipel K, Kilgour DM. 2000. Using environmental indicators to quantify the
Florig HK, Morgan MG, Morgan KM, Jenni KE, Fischoff B, Fischbeck PS, DeKay ML. robustness of policy alternatives to uncertainty. Ecol Model 130:79–86.
108 Integr Environ Assess Manag 1, 2005—G.A. Kiker et al.

Linkov I, Sahay S, Seager TP, Kiker G, Bridges T. 2005. Multi-criteria decision and environmental decision-making. Amsterdam, The Netherlands: Kluwer. p
analysis: Framework for applications in remedial planning for contaminated 97–114.
sediments. In: Proth JM, Levner E, Linkov I, editors. Strategic management of Schmoldt D, Peterson D. 2001b. Strategic and tactical planning for managing
marine ecosystems. Amsterdam, The Netherlands: Kluwer. national park resources. In: Name, editors. The analytical hierarchy process in
Linkov I, Varghese A, Jamil S, Seager TP, Kiker GA, Bridges TS. 2004. Multi-criteria natural resource and environmental decision-making. Amsterdam, The
decision analysis: Framework for applications in remedial planning for Netherlands: Kluwer. p 67–79.
contaminated sites. In: Linkov I, Ramadan A, editors. Comparative risk Schmoldt D, Peterson D, Silsbee D. 1994. Developing inventory and monitoring
assessment and environmental decision making. Amsterdam, The Nether- programs based on multiple objectives. Environ Manag 18:707–727.
lands: Kluwer. p 15–54. Seppala J, Basson L, Norris GA. 2002. Decision analysis frameworks for life-cycle
Males RM. 2002. Beyond expected value: Making decisions under risk and impact assessment. J Ind Ecol 5:45–68.
uncertainty. RMM Technical Services, under contract to Planning and Sharifi MA, van den Toorn W, Rico A, Emmanuel M. 2003. Application of GIS and
Management Consultants Ltd. Prepared for U.S. Army Corps of Engineers, multicriteria evaluation in locating sustainable boundary between Tunari
Institute for Water Resources. IWR Report 02-R-4. www.iwr.usace.army.mil/ National Park and Cochabamba City (Bolivia). Journal of Multi-Criteria Decision
iwr/pdf/02r4bey_exp_val.pdf. Accessed 1 August 2003. Analysis 11:151–164.
Marttunen M, Hamalainen RP. 1995. Decision analysis interviews in environmental Siddiqui M, Everett J, Vieux B. 1996. Landfill siting using geographic information
impact assessment. European Journal of Operational Research 87:551–563. systems: A demonstration. Journal of Environmental Engineering 122:515–523.
Matsatsinis NF, Samaras AP. 2001. MCDA and preference disaggregation in group Simon M, Pascoe S. 1999. A review of applications of multiple criteria decision-
decision support systems. European Journal of Operational Research making techniques to fisheries. Marine Resource Economics 14:41–63.
130:414–429. Stahl CH. 2003. Multi-criteria integrated resource assessment (MIRA): A new
McDaniels TL. 1995. Using judgment in resource management: A multiple decision analytic approach to inform environmental policy analysis [thesis].
objective analysis of a fisheries management decision. Operations Research Wilmington (DE), USA: University of Delaware.
43:415–426. Stahl CH, Cimorelli AJ, Chow AH. 2002. A new approach to environmental
McDaniels TL, Roessler C. 1998. Multi-attribute elicitation of wilderness decision analysis: Multi-criteria integrated resource assessment (MIRA).
preservation benefits: A constructive approach. Ecological Economics Bulletin of Science, Technology and Society 22:443–459.
27:299–312. Steiguer JE, Liberti L, Schuler A, Hansen B. 2003. Multi-criteria decision models for
McDaniels TL, Gregory RS, Fields D. 1999. Democratizing risk management: forestry and natural resources management: An annotated bibliography.
Successful public involvement in local water management decisions. Risk Washington, DC: U.S. Department of Agriculture, Forest Service. General
Analysis 19:497–510. technical report NE-307.
Miettinen P, Hamalainen RP. 1997. How to benefit from decision analysis in Store R, Kangas J. 2001. Integrating spatial multi-criteria evaluation and expert
environmental life cycle assessment (LCA). European Journal of Operational knowledge for GIS-based habitat suitability modeling. Landsc Urban Plann
Research 102:279–294. 55:79–93.
Mendoza GA, Anderson AB, Gertner GZ. 2002. Integrating multi-criteria analysis Timmerman TJ, Kloeber JM Jr, Jackson JA, Deckro RF. 1996. Selecting remediation
and GIS for land condition assessment: Part 2—Allocation of military training technologies through a technical risk index: An application of multi-attribute
areas. Journal of Geographic Information and Decision Analysis 6:17–30. utility theory. Wright Patterson Air Force Base, USA: Center for Modeling,
Morgan MG, Florig HK, DeKay ML, Fischbeck PS. 2000. Categorizing risks for risk Simulation, and Analysis. Technical report 96–01.
ranking. Risk Analysis 20:49–58. Tran L, Knight CG, O’Neill R, Smith E, Ritters K, Wickham J. 2002. Environmental
Morgan MG, Fischhoff B, Bostrom A, Atman C J. 2002. Risk communication. assessment fuzzy decision analysis for integrated environmental vulnerability
Boston (MA), USA: Cambridge University Press. assessment of the mid-atlantic region. Environ Manag 29:845–859.
[NRC] National Research Council. 1999. New directions in water resources [USACE] U.S. Army Corps of Engineers. 1983. The economic and environmental
planning for the U.S. Army Corps of Engineers. Washington DC: National principles and guidelines for water and related land resources implementa-
Academy Press. tion. Engineering Regulation (ER) 105-2-100.
Parnell GS, Frimpon M, Barnes J, Kloeber JM Jr, Deckro RF, Jackson JA. 2001. [USACE] U.S. Army Corps of Engineers. 2003a. Environmental operating principles
Safety risk analysis of an innovative environmental technology. Risk Analysis and implementation guidance. www.hq.usace.army.mil/CEPA/7%20Environ%
21:143–155. 20Prin%20web%20site/Page1.html. Accessed 13 January 2005.
Pavlikakis GE, Tsihrintzis VA. 2003. A quantitative method for accounting human [USACE] U.S. Army Corps of Engineers. 2003b. Planning civil works projects under
opinion, preferences, and perceptions in ecosystem management. J Environ the environmental operating principles. Circular 1105-2-404. www.usace.
Manag 68:193–205. army.mil/inet/usace-docs/eng-circulars/ec1105-2-404/entire.pdf. Accessed 13
Pereira AG, Quintana SC. 2002. From technocratic to participatory decision January 2005.
support systems: Responding to the new governance initiatives. Journal of [USDOE] U.S. Department of Energy. 1998. Guidelines for risk-based prioritization
Geographic Information and Decision Analysis 6:95–107. of DOE Activities. Washington, DC: USDOE. DOE-DP-STD-3023–98.
Peterson D, Silsbee D, Schmoldt D. 1994. A case study of resources management [USDOE] U.S. Department of Energy. 2003. Washington, DC: USDOE-RESRAD
planning with multiple objectives and projects. Environ Manag 18:729–742. Environmental Assessment Division. http://web.ead.anl.gov/resrad/home2.
Prato T. 2003. Multiple-attribute evaluation of ecosystem management for the Accessed 13 January 2005.
Missouri River system. Ecological Economics 45:297–309. [USEPA] U.S. Environmental Protection Agency. 2000. Framework for responsible
Ralston BE, Jackson JA, Kloeber JM Jr, Deckro RF. 1996. Development of a decision environmental decision-making (FRED): Using life cycle assessment to evaluate
support system for the Department of Energy selection of waste remediation preferability of products. Washington, DC: USEPA. EPA/600/R-00/095.
technologies. Wright Patterson Air Force Base, USA: Center for Modeling, [USEPA] U.S. Environmental Protection Agency. 2002. Consistency and trans-
Simulation, and Analysis. Technical report 96-02:1–123. parency in determination of USEPA anticipated ozone designations. Special
Ramanathan R. 2001. A note on the use of the analytical hierarchy process for Review. Washington, DC: USEPA, Office of Inspector General. Report 2002-S-
environmental impact assessment. J Environ Manag 63:27–35. 00016.
Rauscher HM, Lloyd FT, Loftis DL, Twery MJ. 2000. A practical decision-analysis Vaillancourt K, Waaub JP. 2002. Environmental site evaluation of waste manage-
process for forest ecosystem management. Comput Electron Agric 27:195–226. ment facilities embedded into EUGENE model: A multi-criteria approach.
Rogers M, Bruen M. 1998. Choosing realistic values of indifference, preference European Journal of Operational Research 139:436–448.
and veto thresholds for use with environmental criteria within ELECTRE. Wakeman JS. 2003. Milltown reservoir sediment/Clark Fork River superfund site-
European Journal of Operational Research 107:542–551. focused feasibility study. www.epa.gov/region8/superfund/sites/mt/
Schmoldt DL, Kangas J, Mendoza GA, Pesonen M. 2001. The analytic hierarchy milltowncfr/home.html. Accessed 13 January 2005.
process in natural resource and environmental decision making. Amsterdam, Whitaker K, Focht W. 2001. Expert modeling of environmental impacts. OPS
The Netherlands: Kluwer. Special Issue: Environmental Policy in Oklahoma 10:179–186.
Schmoldt DL, Peterson DL. 2001a. Efficient group decision making in workshop Yoe C. 2002. Trade-off analysis planning and procedures guidebook. http://
settings. In: Name, editors. The analytical hierarchy process in natural resource www.iwr.usace.army.mil/iwr/pdf/tradeoff.pdf. Accessed 13 January 2005.