Selecting sustainable waste-to-energy technologies for municipal solid waste treatment: a game theory approach for group decision-making

Accepted Manuscript Selecting sustainable waste-to-energy technologies for municipal solid waste treatment: a game theory approach for group decision-making Atousa Soltani, Rehan Sadiq, Kasun Hewage PII:

S0959-6526(15)01865-X

DOI:

10.1016/j.jclepro.2015.12.041

Reference:

JCLP 6513

To appear in:

Journal of Cleaner Production

Received Date: 14 April 2015 Revised Date:

30 November 2015

Accepted Date: 1 December 2015

Please cite this article as: Soltani A, Sadiq R, Hewage K, Selecting sustainable waste-to-energy technologies for municipal solid waste treatment: a game theory approach for group decision-making, Journal of Cleaner Production (2016), doi: 10.1016/j.jclepro.2015.12.041. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Total word count: 10230

1 2

Selecting sustainable waste-to-energy technologies for municipal solid waste

3

treatment: a game theory approach for group decision-making

5

Atousa Soltania,*, Rehan Sadiqa, Kasun Hewagea

6

a

7

1V7, Canada

8

* Corresponding author: [email protected]

School of Engineering, University of British Columbia, 1137 Alumni Avenue, Kelowna, BC V1V

SC

9 10 11

M AN U

12 13 14 15 16 17

22 23 24 25 26 27 28 29

EP

21

AC C

20

TE D

18 19

RI PT

4

30 31 32 33

1

ACCEPTED MANUSCRIPT

Abstract: An efficient waste treatment strategy should be cost-effective and minimize potential

2

impacts on various stakeholders and the environment. This study proposes a decision framework

3

that can model the stakeholder’s conflicting priorities over the sustainability criteria, when

4

selecting a municipal solid waste treatment option. The proposed framework compares life cycle

5

sustainability impacts of selected options and develops a weighing scheme for combining impacts

6

based on stakeholders’ preferences. It then uses game theory to help the stakeholders fairly share

7

the costs and benefits, and guides the stakeholders to reach an agreement on a mutually

8

sustainable and pragmatic solution. In this study, the application of the framework to select a

9

waste-to-energy technology for Vancouver, Canada is demonstrated. The case study discusses the

10

prospect of producing refuse-derived fuel by cement industry and the municipality. Results show

11

that the cement industry and the municipality may mutually benefit from the refuse-derived fuel,

12

if the industry pays a tipping fee of $0.077-0.96 per kg waste to access the required amount of

13

solid waste from the municipality. The outcome of the framework can help in the approval and

14

application of an overall sustainable option by both stakeholders and in making the negotiation

15

more efficient and timely.

M AN U

SC

RI PT

1

16

Keywords: Sustainability, Municipal solid waste management (MSWM), Game theory,

18

Analytical Hierarchy Process (AHP), group decision-making

EP AC C

19

TE D

17

2

ACCEPTED MANUSCRIPT

1

1. Introduction Municipal Solid Waste (MSW) is defined as “refuse that originates from residential,

3

commercial, institutional, demolition, land clearing or construction sources” in the Environmental

4

Management Act (The government of British Columbia, 2015). The generation of MSW has

5

doubled, globally, in the past decade, and it is anticipated to triple in the next decade (The World

6

Bank, 2012). In 2012, one in eight deaths worldwide were linked to air pollution, a consequence

7

of unsustainable policies in transport, energy, and waste management sectors (WHO, 2014). In

8

2008, Canada disposed1 of 777 kg of MSW per capita, the third highest amount in the world and

9

the highest amount among the developed countries (Hoornweg & Bhada-Tata, 2012) (Figure 1).

10

Canadians disposed of about 25 million tonnes of MSW (720 kg per capita) in 2012, with the

11

province of British Columbia (BC) making up to 10% of that share (Statistics Canada, 2015).

SC

RI PT

2

M AN U

12

900 800 700 600 500 400 300 200 100

EP

0

TE D

Kg MSW per capita per year

1000

Comoros Canada USA Kuwait Switzerland Denmark Cyprus Luxemburg Trinidad and Tobago

AC C

13 14 15

Bahrain

Figure 1 – Ten countries with the highest disposal of MSW per capita in 2008 (derived from D-waste (2015))

16 17

Municipal Solid Waste Management (MSWM) is a complex process that requires group

18

decision-making on selecting waste collection routes, transfer stations, and treatment locations

19

and strategies from various available options (Dewi et al., 2010; Soltani et al., 2015). The

1

The disposed waste refers to the amount of generated waste after diversion and recycling.

3

ACCEPTED MANUSCRIPT

selection of a treatment strategy is one of the most debated issues in the literature and is the core

2

of MSWM (Achillas et al., 2013). Waste treatment strategies often comprise landfilling and

3

waste-to-energy (WTE) technologies. An optimal waste treatment strategy is a result of prudent

4

and scientifically justifiable decision-making that minimizes the risks to the environment and

5

human health, and maximizes cost efficiency (Sadiq, 2001). A sustainability paradigm looks for a

6

waste treatment strategy that “meets the needs of the present without compromising the ability of

7

future generations to meet their own needs” (Brundtland, 1987). In the context of MSWM,

8

sustainability refers to the assessment of environmental, economic, and social impacts of

9

available waste treatment options. Sustainable waste management aims to reduce waste

10

generation, re-use and recycle waste materials, and recover energy to ultimately preserve

11

resources for the future. Recovering energy from disposed waste can generate power for

12

municipalities, offer fossil fuel substitutes to industries, and reduce greenhouse gases and other

13

hazardous pollutants (Metro Vancouver, 2014a).

M AN U

SC

RI PT

1

There are various sustainability assessment frameworks and tools available for calculating

15

environmental impacts (e.g. life cycle assessment (LCA) (the International Organization for

16

Standardization (ISO), 2006a), environmental risk analysis (SETAC, 2004), environmental

17

impact assessment (Canter, 1977)) and net economic costs (e.g., life cycle costing (LCC)

18

(Blanchard et al., 1990)). In addition, there are numerous frameworks for comparing the

19

performances in one criterion to another and finding a balance. A multi-criteria decision analysis

20

(MCDA) framework provides different methods that can help decision-makers to find a suitable

21

trade-off amongst these criteria and choose an option with the lowest overall impacts (Soltani et

22

al., 2015). While LCA and LCC are effective in accounting for the impacts of waste treatment

23

options throughout their life cycle, MCDA methods are required to aggregate their outcomes.

EP

TE D

14

Meanwhile, waste treatment for achieving sustainability goals is even more complex when

25

multiple stakeholders with conflicting interests are involved. In Canada, collection, diversion (i.e.

26

reuse, recycling, and composting) and disposal of waste are generally handled by the municipal

27

governments (Environment Canada, 2012). In addition to the municipal government, multiple

28

stakeholders such as NGOs, environmental experts, general public, and industries affect policies

29

and decisions related to MSWM. Soltani et al. (2015) provided a state-of-the-art review with

30

respect to multiple stakeholders’ involvement in MSWM, in which municipalities and experts

31

were found to be involved in decision-making more than other stakeholders.

AC C

24

32

Waste treatment often creates a situation where the municipality bears all the costs of waste

33

treatment while other stakeholders benefit from it without contributing to the costs; this situation

34

is known as the ‘free-rider’ problem. The municipality is responsible for providing the public

4

ACCEPTED MANUSCRIPT

goods and services, such as waste treatment, and ends up paying the associated costs. The free-

2

rider problem can discourage the municipality from choosing the more advanced and more

3

expensive technologies or can result in the over-using of the provided service. A solution to this

4

problem is to involve other stakeholders in the decision-making and execution process. Other

5

stakeholders will be encouraged to contribute towards the relevant costs of sustainable waste

6

treatments, if municipalities combine these services with other in-demand services (Carraro &

7

Marchiori, 2003). Using WTE technologies for waste treatment mitigates the free-rider problem

8

by offering cleaner energy solutions to industries and encouraging them to pay for the recovered

9

energy and materials; but to make this feasible, the municipality and industry should first

10

negotiate their fair share of costs and benefits to mutually agree on a WTE technology. Hence,

11

effective waste management should evaluate stakeholders’ dialogues in addition to technical

12

assessments (Achillas et al., 2013).

M AN U

SC

RI PT

1

This study proposes a decision framework to help multiple stakeholders reach an

14

agreement on a “sustainable” waste treatment option and share the associated costs and benefits

15

in a fair and mutually acceptable way. This framework is especially helpful for using WTE

16

technology, where often various stakeholders get involved. This research aims to fill the gap in

17

the literature on choosing the most sustainable and pragmatic waste treatment option when

18

stakeholders have conflicting priorities.

19

2. Background Information

20

TE D

13

In this section, components of the proposed framework including sustainability assessment

21

tools, waste treatment options, and decision analysis methods are discussed in more detail.

22

2.1.

EP

Sustainability assessment tools

A sustainability paradigm helps in choosing, building, or offering products and services

24

that conserve resources such as money, water, soil, air, and humans in an acceptable balance.

25

Sustainability assessment tools such as LCA and LCC can evaluate the environmental impacts

26

and economic costs and benefits of selected MSWM options.

27

2.1.1. Life Cycle Assessment

28

AC C

23

LCA is a popular method that helps experts estimate environmental burdens of products,

29

processes, and services throughout their life (USEPA, 2006). LCA identifies and quantifies inputs

30

and outputs to a system, including materials, energy, waste, and pollution (SETAC, 1993).

31

According to the ISO 14040 and 14044 standards, LCA consists of four general steps (ISO,

32

2006a; 2006b):

5

ACCEPTED MANUSCRIPT

1. Goal and scope encompass system boundary, functional unit, criteria under study,

2

available options, and involved stakeholders. The system boundary is the section of the life

3

of a product or service that is considered in the LCA. Waste treatment is often the end-of-

4

life process for products, but in waste management studies, the life of a disposed waste is

5

often from the collection point to the waste treatment plant, landfill, and even the re-using

6

destination. The functional unit is the homogenous unit of waste for which all impacts are

7

estimated. The criteria are directly connected to the goal of the study. In sustainable waste

8

management, these criteria include environmental, economic, and - when available - social

9

criteria. The available options are the alternatives that are being compared based on the

10

selected criteria, and the stakeholders are the agents or individuals that are impacting or

11

being impacted by the outcomes of these assessment.

12

2. Life cycle Inventory (LCI) is a collection of all materials, energy, and discharges

13

entering a system boundary or released to land, water, and air. To develop the LCI, a flow

14

chart of input and output materials and discharges from and to the system boundary is

15

initially displayed. SimaPro is a comprehensive tool for collecting data and analyzing the

16

impacts of various products and services, with access to various databases (e.g., ecoinvent,

17

Agri-Footprint, European reference Life Cycle Database, and U.S. Life Cycle Inventory

18

Database).

19

3. Life cycle Impact Assessment (LCIA) groups the inputs and outputs collected in LCI

20

under environmental impact categories based on their potential hazards to human health

21

and the environment. LCIA usually includes the following steps: selection of impact

22

categories, classification, characterization, normalization, grouping, weighting, and

23

reporting the results (USEPA, 2006). Based on the ISO standards 14040 and 14044 for Life

24

Cycle Impact Assessment and their technical report ISO/TR 14047, the most common

25

environmental impact categories are abiotic depletion (depletion of resources),

26

stratospheric ozone depletion, summer smog (photochemical oxidation or photo oxidant

28 29 30

SC

M AN U

TE D

EP

AC C

27

RI PT

1

formation), acidification, human toxicity, and ecotoxicity (terrestrial and aquatic) (ISO, 2006a; 2006b; 2012). SimaPro can present both mid-point and end-point impacts using different impact assessment methods (e.g., CML-IA, EDIP, ILCD, ReCiPe, BEES, and TRACI). SimaPro also follows different studies to develop the default characterization

31

factors (e.g., Intergovernmental Panel on Climate Change for climate change impact

32

category). A list of all characterization models can be found in PRe (2015).

33

4. Interpretation of results is the last step, which includes explanation of the outcomes by

34

experts.

6

ACCEPTED MANUSCRIPT

1

Since LCA does not consider economic or social impacts of a product or process, LCC is

2

often suggested to be used in parallel with LCA in a sustainability assessment (Gluch &

3

Baumann, 2004).

4

2.1.2. Life Cycle Costing LCC sums up the monetary values of costs and benefits from all stages of the life of a

6

product or service in a system boundary (Gluch & Baumann, 2004). Common costs include

7

investment, operation, and borrowing costs, while benefits include sale revenues (Carlsson Reich,

8

2005). Various methods are used to perform LCC in waste management studies, including Net

9

Present Value (NPV) (Carlsson Reich, 2005), equivalent annual cost (Tsilemou &

11 12 13 14 15 16

SC

Panagiotakopoulos, 2006), and internal rate of return (Caputo et al., 2003).

In this paper, NPV method is proposed to consider the time value of money. Future costs and benefits (recurrent or one-time) are discounted to present values using equation [1].
 = ∑
 (1 + ) 

M AN U

10

RI PT

5

[1]

where NPV is the present value of net economic cost,
 is the net future costs,  is the real interest rate,  is time, and  is the total number of periods.

It should be noted that decision-making solely based on the outcomes of LCC will end in

subjective decisions that overlook the environmental dimension (Gluch & Baumann, 2004).

18

Many social impacts can also be considered through economic assessment (e.g., employment,

19

neighborhood land prices, etc.). There are various quantitative and qualitative methods for

20

evaluating social impacts (Refer to a review by Chhipi-Shrestha et al., 2014). Although these

21

methods are not further discussed in this study, the following sub-criteria can be considered for

22

social assessment of waste treatment options:

TE D

17

- Proximity to residential area (e.g., Noise, Odour (den Boer et al., 2007))

24

- Workers’ and neighborhood’s safety (Sheppard & Meitner, 2005)

25

- Employment (den Boer et al., 2007)

27 28 29 30

AC C

26

EP

23

- Affordability

- Public acceptability

- Land use (den Boer et al., 2007)

2.2.

Waste-to-energy technologies

Among all MSWM stages, waste treatment is often the main path toward the protection of

31

the environment and human health, and growth of the economy (Soltani et al., 2015). In recent

32

years, waste treatment is facing new constraints as a result of limited space for landfills,

33

increasing opportunity cost of disposed waste, and strict environmental regulations (Reza et al.,

34

2013). The objective of sustainable assessment in the waste context is to reduce environmental

7

ACCEPTED MANUSCRIPT

1

impacts and economic costs of waste treatments, while being aware of their social effects, and

2

eventually create a balance between these outcomes. Recovering energy from waste is a more

3

advanced strategy that conceives of waste as an opportunity rather than a liability. WTE technologies recover energy in the form of heat, electricity, or steam, and retrieve

5

bottom ash and metal from disposed waste. Various WTE technologies are as follows (Metro

6

Vancouver, 2014a):

RI PT

4

7

1. Mass-burn incinerations are the most commonly used WTE technology. In mass-burn

8

incineration, waste is directly combusted after mild or moderate pre-processing to produce

9

electricity. In this study WTE options refers to mass-burn incineration.

10

SC

11

2. Gasification and pyrolysis convert waste to syngas or vapour to generate electricity and heat.

3. Co-combustion of Refuse-derived fuel (RDF) in a cement kiln is another use of WTE

13

technology in MSW treatment that substitutes fossil fuels with RDF in the production process of

14

cement. RDF is a solid fuel recovered from high-calorific value fraction of MSW (Genon &

15

Brizio, 2008; Reza et al., 2013). To produce RDF, MSW undergoes some or all of the following

16

stages: sorting or mechanical separation, shredding, screening, blending, and pelletizing

17

(Gendebien et al., 2003). The extent of environmental impacts and economic net costs of RDF

18

depends on MSW composition in the region, and recovery rate; but substitution of fossil fuels in

19

an energy-intensive industry such as cement with RDF will definitely reduce greenhouse gas

20

emissions and mitigate many environmental concerns (Reza et al., 2013).

21

2.3.

Decision analysis

TE D

M AN U

12

The first step in selecting a sustainable waste treatment strategy is to calculate the total

23

impact of all options on the stakeholders. Many frameworks and methods have been developed to

24

combine the results of environmental and economic assessments (e.g., energy and material

25

intensity metrics (Schwarz et al., 2002), sustainability accounting (Bebbington et al., 2007),

26

MCDA (Contreras et al., 2008), and monetized ecological footprints (Sutton et al., 2012)).

27

MCDA, the most popular among these frameworks, is a collection of various methods that

28

analyzes multiple criteria and helps decision-makers to rank or select acceptable options, or to

29

choose one optimal option. Finding a balance among the criteria is an important part of

30

sustainable development (Neugebauer et al., 2015). Since impact assessments are presented in

31

different units and the value of impacts in one criterion is different from the other ones, using

32

MCDA is necessary to present a unified sustainability index. In addition, for deciding on

33

investment opportunities in waste treatment, the concept of MCDM is more helpful than relying

34

solely on life cycle assessments (Spengler et al., 1998).

AC C

EP

22

8

ACCEPTED MANUSCRIPT

However, to use MCDA methods, stakeholders should first agree on the criteria of interest

2

and the importance of each criterion in comparing the available options (van den Hove, 2006). If

3

stakeholders have conflicting priorities over criteria, reaching an agreement is likely to be

4

challenging (De Feo & De Gisi, 2010). MCDA techniques aggregate the impacts on stakeholders,

5

but fall short on considering stakeholders’ conflicts and their influences on each other in reaching

6

a mutual decision. Game theory, on the other hand, is a natural choice for analyzing the trade-offs

7

between environment and economy, and considering stakeholders’ conflicts and dialogues

8

(Moretti, 2004).

9

2.3.1. Analytical hierarchy process (AHP)

RI PT

1

AHP is a common MCDA technique that offers a mathematical solution for presenting

11

preferences by using a pairwise comparison of criteria (Hossaini et al., 2014; Sadiq, 2001). AHP

12

can help experts assign weights to various criteria, aggregate impacts from those criteria, and

13

compare MSWM options accordingly. There are five steps in AHP (Saaty, 1980):

M AN U

SC

10

-

Present the problem in a hierarchy of goals, criteria, and alternatives

15

-

Collect data on available options and criteria of interest

16

-

Generate a weighting system for criteria through pair-wise comparison

17

-

Rank alternatives by aggregating scores and weights

18

-

Perform a sensitivity analysis to validate the data and mitigate uncertainty.

19

In AHP, the outcomes of pairwise comparisons are presented in a priority matrix and

20

developed into a set of “priority ratio scale” (Hossaini et al., 2014; Sadiq et al., 2003). In Matrix

21

A [Eq. 2], each entry aij shows on what scale criterion i is preferred to criterion j (Hossaini et al.,

22

2014). The advantages of the pairwise comparison technique in AHP include its ease of use and

23

understandability for non-expert users. AHP is also the most common method in waste treatment

24

studies with multiple stakeholders. Saaty’s 9-scale is often used to compare the criteria verbally

25

(Table 1) (Saaty, 1988). It is important to be consistent with the scale in pairwise comparisons.

27 28

EP

1  =  … 

…  1 … … 1

AC C

26

TE D

14

[2]

Table 1 – Saaty’s 9-scale (Saaty, 1988)

Scale

Verbal definition

1

Equal importance

3

Moderate importance

5

Strong importance

7

Very strong or demonstrated dominance

9

ACCEPTED MANUSCRIPT

9

Extreme importance or strongest affirmation

2,4,6,8

Intermediate values

1 Weights are often developed from a priority matrix using various mathematical approaches

3

such as eigenvector, geometric mean, and arithmetic mean, which are believed to present similar

4

results and not be significantly different (Hossaini et al., 2014). Assuming that matrix A shows

5

the relevant importance of each criterion against the other ones, the multiplication of A and W

6

(weight matrix) results in a scalar multiple of W (Saaty, 1994). The scalar value (Eigenvalue) and

7

W (Eigenvector) are calculated using equation [3]. Once weights are developed, their consistency

9

should be examined. The values in a matrix are consistent where each entry  =

SC

8

RI PT

2

weight of criterion or sub-criterion i (∑ " = 1)) (Tesfamariam & Sadiq, 2006).

 !

( " is the

12

other MCDA techniques in considering conflicts in priorities in a group decision-making process

13

(Tseng et al., 2009). A few studies in the literature have offered solutions for conflicts among

14

stakeholders (e.g. Munda, 2002; van den Hove, 2006), but ‘game theory’ is more suitable in

15

choosing among waste treatment options. Game theory can model different types of interactions

16

among stakeholders and predict the outcomes of negotiations.

17

2.3.2. Game theory

M AN U

11

W=$W

[3]

Although AHP is more convenient and effective, it suffers from the same shortcoming as

TE D

10

Game theory studies self-interested2 stakeholders when they interact in a series of games

19

(Leyton-Brown & Shoham, 2008). What makes game theory versatile is its use of mathematical

20

modeling to understand human interactions. Game theory is based on the fact that satisfaction of

21

each stakeholder can change in response to other stakeholders’ actions as well as their own.

22

Games are portrayals of actions that players are interested in and can do, while solutions present

23

the actions they do take (Osborne & Rubinstein, 1994).

AC C

EP

18

24

In a two-player game, game theory first gathers the information on stakeholders’ utilities

25

(or net benefits) from each pair of actions (e.g., how much will each stakeholder benefit if

26

stakeholder 1 chooses action a and stakeholder 2 chooses action b). Based on the type of

27

decision-making problem, this information is then portrayed in a decision tree or a table.

28

Although each stakeholder might choose an optimal option when deciding individually, game

29

theory looks for solutions that are stable in a mutual setting. In this setting, stakeholders answer a

2

Self-interested here means that stakeholders prefer some situations to other situations (or states) and they will act toward making those situations happen (Leyton-Brown & Shoham, 2008).

10

ACCEPTED MANUSCRIPT

1

series of “what if …” questions before finalizing their decision. These questions are often asking

2

whether they would change their decision, if the other stakeholder chooses any of the possible

3

options. When WTE technologies are proposed for a region, municipalities and industry need to

5

share costs and benefits in a way that satisfies both of them. In many studies (e.g., McGinty et al.,

6

2012; Weikard and Dellink, 2008; Finus, 2000; Kaitala and Pohjola, 1995), game theory offers

7

solutions for fair distribution of costs and benefits among stakeholders to stabilize environmental

8

decisions. Game theory is also an effective method for decisions that require stakeholders’

9

collaboration (Nagarajan & Sošić, 2008).

RI PT

4

There have been only a few studies that used game theory for waste management decision-

11

making. Cheng et al. (2002, 2003) applied a cooperative game theory approach to select a new

12

landfill site. Moretti (2004) proposed a cooperative game theory method to divide costs of waste

13

collection between municipalities. Jørgensen (2010) used game theory for a regional waste

14

disposal problem. Karmperis et al. (2013) proposed a framework called the waste management

15

bargaining game to help players negotiate over the surplus profit of various MSWM options.

16

More details on studies with game theory solutions are presented in Table 2. These studies do not

17

consider stakeholders’ impacts on each other’s decisions when interactions are unavoidable.

18

These studies also fail to guide stakeholders to reach a mutual agreement by changing their share

19

of costs and benefits.

20 21 22

TE D

M AN U

SC

10

Table 2 – MSWM studies with game theory solutions

Paper/Authors Cheng et al. (2002)

✓

Cheng et al. (2003)

✓

Jørgensen (2010)

✓

Cooperative game theory – MCDA Inexact linear programming – MCDA Dynamic cooperative game theory

11

Social

Agriculture

Economy

Criteria Environment

Waste industry

Public

Cost-benefit

Municipality

Stakeholders

Experts

Method

EP

Disposal / collection

Location

AC C

Cooperative

Game theory model

Topic

✓ ✓ ✓

✓ ✓ ✓ ✓

✓ ✓ ✓

✓ ✓ ✓ ✓

✓

✓

ACCEPTED MANUSCRIPT

✓

Competitive

Karmperis et al. (2013)* Davila et al. (2005)

✓

✓

Cooperative game theory – Shapley value

✓

✓

Lexicographic mini-max approach

✓

✓ ✓

Waste management bargaining game

✓

Grey integer programming – Zero-sum game

✓

RI PT

Erkut et al. (2008)

✓ ✓

✓

✓ ✓

✓

✓

✓

✓ ✓

SC

Moretti (2004)

* Partially competitive and cooperative

4

3. Proposed Framework

5

This study proposes a decision framework to address the challenges of choosing a mutually

6

sustainable waste treatment strategy when the stakeholders have conflicting preferences. This

7

decision framework aims to help the main stakeholders, the municipality, and industry to reach a

8

mutual agreement on a sustainable and pragmatic waste treatment option. In this framework,

9

game theory complements AHP, LCA, and LCC to model the dialogues among stakeholders and

10

guide them to reach a sustainable solution. Figure 2 presents a schematic of the proposed

11

framework for waste treatment options. Production and co-combustion of RDF in cement kilns in

12

Metro Vancouver, Canada is chosen as a case study to demonstrate the proposed framework.

13

3.1. Scope definition

EP

TE D

M AN U

1 2 3

The first step in the proposed framework is to define the scope of the study, namely the

15

objectives and scenarios. The objective of this study is to compare the available waste treatment

16

options in municipalities and then guide the interested stakeholders to mutually agree on a

17

sustainable and pragmatic option. The scenarios are often built according to available and

18

proposed waste treatment options, composition of disposed waste, and involved stakeholders.

19

AC C

14

12

1 2 3 4

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 2 – Decision framework 3.2. Sustainability assessment

Life cycle assessment and life cycle costing are used as sustainability assessment methods to

6

evaluate the impacts of selected waste treatment options on each stakeholder. The social impacts

7

are not separately assessed through social life cycle assessment in this study, but if available,

8

labour safety, neighborhood land pricing, and political stability due to the decision are among the

9

factors that can be considered (de la Fuente et al., 2015). Other factors such as pollution, changes

10

in employment, and impacts on welfare and market of natural resources can be considered

11

through environmental and economic impacts. Although in many investment decisions, social

12

factors are often ignored by industry, including the municipality as the main stakeholder with

13

veto power on the final decision helps include social impacts as part of the decision. This study

14

will focus on the environmental and economic impacts, but the framework has the ability to take

15

the outcomes of social impact assessments (grey area in Figure 2) into the consideration.

AC C

EP

TE D

5

16

In the LCA, the system boundary is defined to include all activities in the life of disposed

17

waste from the disposal of waste at treatment plants or landfill to mechanical treatment,

18

incineration, and recovery of materials (a cradle to gate approach). In the cradle to gate approach

19

all activities from the extraction of materials (disposal of waste) to the execution of the project

20

(recovery of material and energy or landfilling of the waste) are considered. The transportation of

13

ACCEPTED MANUSCRIPT

1

recovered material and energy to the destination point is not considered. The functional unit is 1

2

kg of MSW. The inflows and outflows of energy, mass, emissions, and discharges to and from the system

4

boundary are derived using the European Life Cycle Database (ELCD) in SimaPro 8.0 software.

5

ELCD gathers data on average waste treatment plants and landfills with leachate control in

6

Europe. The inventories are derived for incineration and landfilling of 1 kg of waste components

7

such as paper and plastic in an average treatment plant or landfill in Europe. Inventories are then

8

adapted to waste composition of the region under study. To perform an impact assessment,

9

inventories are grouped for each impact category. RECIPE method for mid-point impacts is used

10

as a default in SimaPro 8.0 to aggregate the impacts under each mid-point impact category. These

11

values are then unified and presented for each treatment option.

SC

RI PT

3

In the LCC, it is important to define a scope coherent with the scope of LCA. A scheme of

13

all costs and benefits within the scope of study is formed and values per functional unit are

14

estimated in dollars. For this study, previous data on similar waste treatment plants, experts’

15

knowledge, and published literature are used to generate these estimates. Costs and benefits

16

considered in this framework are presented in Table 3. Opportunity cost is in fact the benefit that

17

is not achieved as a result of a decision. Carbon tax is the tax that some governments assign to

18

productions and services with high levels of greenhouse gas emission. Operation and

19

maintenance costs are the values that businesses pay monthly or annually to achieve a desired

20

level of productivity, and these include salaries, rent, and maintenance of buildings and

21

equipment. Depending on the project under study, the transportation cost can include the costs of

22

trucks and gas from the disposal station to the treatment plant. Building, equipment, and land

23

costs are paid at the beginning of the production process as an investment. Stakeholders earn

24

revenue by selling the recovered material and energy (as electricity, gas, etc.). If the recovered

25

energy substitutes fossil fuels, the price of the substituted fossil fuels can be considered as a

26

benefit. Sunk costs or costs that have already been paid are not considered. Net economic cost of

27

each treatment option is calculated for each stakeholder and then presented in dollar value.

29

TE D

EP

AC C

28

M AN U

12

Table 3 – Costs and benefits considered in proposed framework

Costs

Benefits

Opportunity cost

Fossil fuel saving

Carbon tax

Recovered materials revenue

Operation and maintenance cost

Energy revenue (e.g. Electricity sale)

14

ACCEPTED MANUSCRIPT

Transportation cost Land costs Building and equipment costs,

1

RI PT

23.3. Decision-making for sustainable waste treatment

Once the magnitude of impacts is calculated, the AHP method is used to develop weights

4

and then assign them to environmental and economic criteria in a two-step hierarchy. The first

5

priority matrix is developed from pairwise comparison of environmental impact categories. This

6

matrix shows the significance of different environmental impacts on humans and the

7

environment. Although stakeholders can perform these comparisons on their own, the framework

8

suggests a more consistent and transparent approach with scientific proof: Tool for the Reduction

9

and Assessment of Chemical and other environmental impacts (TRACI). The US EPA gathered a

10

well-mixed group of experts, industry, and municipalities to develop TRACI, create the priority

11

matrices, and evaluate weights (Bare, 2002). TRACI provides three weighting systems of

12

Environmentally Preferable Purchasing (EPP), US EPA Science Advisory Board, and Harvard

13

Kennedy School of Government (Table 4) (Gloria et al., 2007). The suggested weighting system

14

in the framework is EPP, as it asks experts to compare the impacts in short, medium, and long

15

term and then uses AHP to develop the weights.

16

Table 4 – Weighting systems for environmental impact categories (Gloria et al., 2007) Impact category Weights EPP

Science advisory

Harvard

29.3

16

11

EP

Global Warming Fossil Fuel Depletion

9.7

5

7

Criteria Air Pollutants

8.9

6

10

Water Intake

7.8

3

9

11

6

AC C

17

TE D

M AN U

SC

3

Human Health Cancerous

7.6

Human Health noncancerous

5.3

Ecological toxicity

7.5

11

6

Eutrophication

6.2

5

9

Habitat alteration

6.1

16

6

Smog

3.5

6

9

Indoor air quality

3.3

11

7

15

ACCEPTED MANUSCRIPT

Acidification

3.0

5

9

Ozone depletion

2.1

5

11

Total

100

100

100

1 In the next level of hierarchy, stakeholders are asked to compare environmental burdens

3

with economic costs to create a priority matrix (Figure 3). This matrix is designed to show each

4

stakeholder’s priority, when making a sustainable decision. AHP uses this priority matrix and the

5

eigenvalue method to calculate weights for the criteria. The overall weights of the impact

6

categories are now calculated from the weights developed for the environmental criterion and the

7

TRACI weights. To estimate the utility of the stakeholders or sustainability Indices, the overall

8

weights (significance of criteria and sub-criteria) are multiplied by impacts (magnitude of burdens

9

or costs) [5]. In this framework, a higher sustainability index means that the waste treatment

11 12 13 14

SC

M AN U

10

RI PT

2

option is less sustainable for the stakeholder.

%& "'%(ℎ *+ ',%*-'& %-./ /'(*0 % ("1 ) = " × "1 345%6%&%0 %7'8 (39) = :ℎ' "'%(ℎ *+ /%'%* ("- * "'% ) × '&', %-./ [5]

Level 2 of the hierarchy

TE D

Level 1 of the hierarchy Environmental impact categories

Criteria/Costs

Environmental 1

s1

s2

…

sn

Environmental

TRACI - EPP

w1

w2

…

wn

Economic

Final weights

we* w1

we* w2

…

we* wn

Weights

EP

Sub-criteria

Economic

1 we

wm

Figure 3 – Weights for each stakeholder in a two-level hierarchy Finally, the proposed framework uses game theory to model the dialogues and predict the

AC C

15 16 17

[4]

18

best option for each stakeholder. The normal model in the game theory is used for this framework

19

as it represents simultaneous decision-making when stakeholders have full information about

20

each other. The game is portrayed in a matrix form with stakeholder 1 in rows and stakeholder 2

21

in columns. Values in the matrix are sustainability indexes for available and proposed waste

22

treatment options (Table 5). If equilibrium exists, the outcome of the game will be a pure

23

strategy. Pure strategy selects the most sustainable waste treatment option for each stakeholder.

24

Decisions are stable in the equilibrium point, therefore stakeholders do not benefit from changing

25

their choices.

16

ACCEPTED MANUSCRIPT

1 2 3

Table 5 – Example of a two-player normal game theory model

6

Option b

Option 1

SI11, SI2a *

SI11, SI2b

Option 2

SI12, SI2b

SI12, SI2b

Option 3

SI13, SI2c

SI13, SI2b

Option c SI11, SI2c SI12, SI2c SI13, SI2c

* SIij is the sustainability index of option j for stakeholder i. i = {1,2} and j = {1,2,3,a,b,c}

SC

4 5

Option a

RI PT

Stakeholder 1

Stakeholder 2

3.4. Mutual agreement

At this point, if the framework predicts landfill for the municipality and a WTE technology

8

for the selected industry, it will provide no waste for the industry to recover energy from the

9

waste. Hence, the proposed framework takes it further to estimate fair shares of costs and benefits

10

(i.e. a tipping fee that one stakeholder should pay to the other one) to make the WTE technology

11

attractive to the municipality and assure mutual agreement on a sustainable option.

M AN U

7

Independent variables such as waste composition have uncertain values, which can change

13

sustainability indexes and ultimately stakeholders’ decisions. To take this into account, a general

14

regression equation is presented in equation [6] to estimate the relation between these variables

15

and sustainability indexes.

17 18 19

39 =  ; + < ;< + ⋯ +  ;

i=1,…,n j=1,…,n

where SIij is sustainability index of alternative i for stakeholder j, > is the coefficient of ;> , and ;> is the kth independent variable (e.g., cost estimates, weights, etc.).

The higher the > , the more dependent is the final decision to ;> . This regression equation

will show how changes in ;> will affect SI.

21

4. Case Study

AC C

20

22

[6]

EP

16

TE D

12

The goal of this case study is to implement the developed framework for WTE decision-

23

making to help Metro Vancouver and the cement industry in the region by estimating their fair

24

shares of costs and benefits as well as reaching a mutual agreement on RDF. Metro Vancouver is

25

a regional district in the province of BC, Canada that consists of 21 municipalities. In this study,

26

Metro Vancouver is referred to as a municipality.

27

4.1. Scope definition for MSWM in Metro Vancouver

17

ACCEPTED MANUSCRIPT

In 2008, Canada’s local governments spent $2.6 billion, while earning $1.8 billion on

2

waste management. Nova Scotia and BC spent the highest per person ($30) on waste treatment

3

and operation, about twice the national average value (Statistics Canada, 2011). Metro Vancouver

4

aims to reduce waste management costs and environmental impacts to generate earnings for the

5

municipality. In addition, they are aware of social impacts such as employment, material and

6

energy markets, and pollution and noise in the neighbourhood of waste treatment plants.

7

Therefore, additional steps are always taken to make sure these impacts are minimized. For

8

example, waste treatment plants are built in industrial neighbourhoods and it is considered that

9

new waste treatment plants follow increased employment in the waste section.

RI PT

1

Metro Vancouver’s MSWM plan includes recycling and re-using of waste materials,

11

producing energy from waste, and disposing of the rest in landfills. In 2010, around 1.4 million

12

tonnes of solid waste, equivalent to about 45% of the generated waste, were disposed of in

13

landfills or burned in the WTE plant (Metro Vancouver, 2010). Delta and Cache Creek landfills

14

receive waste from the Metro Vancouver area. The WTE facility in Burnaby currently carries out

15

mass burn thermal treatment3. Metro Vancouver plans to build a new WTE plant, and RDF is one

16

of the proposed options for this upcoming plant (Reza et al., 2013; Metro Vancouver, 2014c).

17

About 58% of waste is recycled now, while the target is to recycle 80% by 2020. For the 700,000

18

tonnes of waste remaining after 80% diversion, the planned distribution is as follows (Metro

19

Vancouver, 2013):

TE D

M AN U

SC

10

20

- 50,000 tonnes to Delta landfill

21

- 280,000 tonnes to Burnaby WTE plant

22

- 370,000 tonnes to new WTE plant

In this case study, the two stakeholders of Metro Vancouver and the cement industry are

24

proposing the new waste treatment option of combusting RDF in cement kilns. Besides RDF,

25

other scenarios in this case study are landfilling and WTE (mass-burn). To fully understand the

26

scope of these scenarios, waste composition in Metro Vancouver is presented in Table 6.

AC C

EP

23

27 28

Table 6 – Metro Vancouver MSW composition in 2013 (Metro Vancouver, 2014c)

3

MSW components

Vancouver in 2013 (%)

Paper

13.6

Plastic

14.4

Compostable organics

36.2

In this case study, WTE option refers to mass burn treatment used in the Burnaby plant.

18

Textile

2.7

Rubber

2.7

Leather/multiple composite

2.7

Metals

3.2

Glass

1.6

Building material

8.4

Electronic waste

1.1

Household hazardous

0.9

Household hygiene

5.0

Bulky objects

4.1

Fines

0.6

Total

SC

2.7

M AN U

Wood*

RI PT

ACCEPTED MANUSCRIPT

100

1

* Wood, textile, rubber, and leather are reported in total under non-compostable

2

organics.

3 4

4.2. Sustainability assessment

Sustainability assessment in this case study is mainly built on the results from a study

6

conducted by Reza et al. (2013), where environmental and economic impacts of RDF were

7

compared with existing waste treatment options in Metro Vancouver. In 2013, around 3.4 million

8

tonnes of MSW were disposed of in Metro Vancouver. Three different scenarios were considered

9

for treatment of the disposed waste, in this study:

11 12

estimation of landfilling impacts. -

13 14 15 16

Landfilling of the disposed waste. The composition of mixed MSW in Vancouver is used for

EP

-

Mass-combustion of the disposed waste. The composition of mixed MSW in 2013 is used for

AC C

10

TE D

5

the calculation of impacts.

-

Co-combustion of RDF in cement kilns in Vancouver. Cement kilns are assumed to be close to potential cement manufacturers. In previous studies, RDF production recovered 20-50% of the disposed MSW (Nithikul, 2007; Rotter et al., 2004; Gendebien et al., 2003). In this study,

17

40% of the disposed MSW (e.g., paper, plastic, wood, textile, rubber, and leather) is

18

recovered for RDF production and co-combustion in cement kilns.

19

For LCA, inputs and outputs (e.g., materials, energy, and emissions) from landfilling and

20

incineration of each component of waste were collected from SimaPro 8.0 and then adapted to

21

Vancouver’s waste composition (Table 7). The values in Table 7 show the mid-point impacts of 1

19

ACCEPTED MANUSCRIPT

kg MSW in each impact category. For RDF co-combustion in cement kilns, the required

2

technological changes in the production line, additional indoor air emissions in the plant, and the

3

impacts on the quality of cement are not considered or discussed. Since RDF has not yet been

4

used in cement kilns in Canada, some additional impacts may exist that are not considered in this

5

study.

6 7

RI PT

1

Table 7 – Environmental impact assessment for Metro Vancouver case study Impact category

Unit

Landfill

WTE

Co-combustion

(Mass-combustion)

RDF in cement kilns

-4.49E-08

kg CFC-11 eq

2.26E-09

-3.77E-08

Global warming

kg CO2 eq

2.23E-01

-2.11E-02

-2.55E-02

Smog

kg O3 eq

5.60E-03

-8.64E-03

-1.44E-02

Acidification

mol H+ eq

1.38E-02

-9.97E-02

-1.21E-01

Eutrophication

kg N eq

5.09E-04

-1.97E-05

-3.01E-05

Carcinogenics

CTUh

2.88E-10

-1.70E-10

-2.23E-10

Non carcinogenics

CTUh

5.24E-10

-6.23E-09

-8.70E-09

Respiratory effects

kg PM10 eq

2.90E-04

-4.05E-04

-4.46E-04

Ecotoxicity

CTUe

2.26E-09

-1.52E-02

-1.46E-02

M AN U

SC

Ozone depletion

8

of

To perform LCC for this case study, landfilling costs were based on the current tipping fee

10

of $0.108 per kg waste (City of Vancouver, 2013). Cost of landfilling for the industry is the

11

opportunity cost of not choosing WTE or RDF, which is the cost of fossil fuels in their current

12

practice. Price of coal was considered as $0.07 per kg coal (78$ per ton). The energy of 1 kg coal

13

is 24.5 MJ (Reza et al., 2013) equivalent to 6.80 kWh, while the output energy for 1 kg of MSW

14

in the WTE plant in Burnaby was 15 MJ or 4.2 kWh in 2007 (TRI Environmental Consulting Inc,

15

2008). In addition, when landfilling is chosen as the waste treatment option, industry should pay

16

carbon tax at the rate of $30 per tonne of CO2 equivalent emissions (BC Ministry of Finance,

17

2014).

EP

AC C

18

TE D

9

WTE profits for municipality include metal recovery and electricity sale. If WTE is chosen,

19

industries save on fossil fuel and carbon tax. Metro Vancouver earns on average about $6 million

20

per year from electricity produced in the WTE facility (120000 MWh) and $1.4 million per year

21

from recovered metal (Metro Vancouver, 2014a). One (1) kg of waste generates $0.005 revenue

22

from 0.03 kg scrap metal and $0.022 revenue from electricity. Construction, operation, and land

23

costs of the existing WTE facility were considered as sunk costs.

20

ACCEPTED MANUSCRIPT

Landfill reduction of 60% (derived from Reza et al. 2013) was considered as a benefit for

2

the municipality, and tax saving, fuel saving, and metal recovery were considered as benefits for

3

industry. Since stakeholders impact each other, LCC is presented to show the result of these

4

impacts (Table 8). Metro Vancouver is considering 10 potential treatment plans for the future,

5

among which two options are planning to use RDF (Metro Vancouver, 2014d). Therefore, it is

6

assumed that Metro Vancouver can still produce RDF without the cement industry but the cost

7

will be slightly higher. In addition, the cement industry would follow its current practice if Metro

8

Vancouver chooses landfill or WTE. Table 8 shows the LCC results of stakeholders selecting a

9

pair of actions at the same time. For example, when stakeholder 2 (Metro Vancouver) decides to

10

landfill waste and stakeholder 1 (cement industry) prefers RDF, the economic impacts on

11

stakeholder 2 and 1 are $0.63 and $0.108 per kg MSW, respectively.

SC

RI PT

1

13

M AN U

12

Table 8 – Economic net costs for case study in Metro Vancouver Stakeholder 1’s choice –

Stakeholder 1 – Industry Stakeholder 2 – Municipality

Stakeholder 2’s choice

($ per kg MSW)

($ per kg MSW)

Landfill – Landfill

0.63

0.108

Landfill – WTE

0.63

Landfill – RDF

0.63

WTE – Landfill

0.63

0.108

0.63

-0.026

0.63

0.05

0.63

0.108

0.63

-0.026

-0.31

0.043

WTE – RDF

RDF – WTE RDF – RDF

EP

RDF – Landfill

0.05

TE D

WTE – WTE

-0.026*

* Negative value indicates earnings.

16

4.3. Decision-making for multiple stakeholders

17

AC C

14 15

The developed framework creates two levels of weights for environmental and economic

18

criteria and their sub-criteria. TRACI provides the weights of the environmental impact

19

categories. These weights are similar for both stakeholders. According to the specifics of each

20

case study and the availability of impact data, the collection of impact categories can vary. Once

21

the impact categories are selected, their weights can be calculated (out of 100%) from the original

22

EPP weights (Table 9).

23

21

ACCEPTED MANUSCRIPT

1

Table 9 – The first-level weights for Metro Vancouver case study Sum

C5. Ozone depletion

C6. Human toxicity

C7. Fresh aquatic ecotoxicity

40

3

10

11

Si : Stakeholders , Ci : Sub-criteria

C9. Photochemical oxidation

C4. Global warming 100

9

10

5

RI PT

C3. Eutrophication

4

C8. Terrestrial ecotoxicity

C2. Acidification

8

100

SC

2 3

C1. Abiotic depletion

Weights of the selected environmental sub-criteria (%)

4

In addition, the authors have made reasonable assumptions based on their expert

6

judgements for stakeholders’ priorities in comparing environmental and economic criteria. They

7

have made assumptions about the stakeholder’s priorities (i.e. whether they prefer environmental

8

or economic criterion) and the degree of those priorities (i.e. how much they prefer one criterion

9

over another). These judgements arrive from the authors’ involvement in many individual and

10

group discussions with Metro Vancouver and the cement industry, and participation in relevant

11

conferences by Metro Vancouver. Metro Vancouver has continuously explored sustainable waste

12

management plans to prioritize the environment and human health, while it is a safe assumption

13

that the cement industry would only invest in an option when it is financially sound. Table 10

14

presents the subsequent priority matrix. The stakeholders express their preferences through

15

pairwise comparisons of criteria, using Saaty’s 9-scale. Weights of environmental impacts and net

16

economic costs are estimated for each stakeholder using AHP.

EP

TE D

M AN U

5

17

Table 10 – The second priority matrix and weighting system for Metro Vancouver case study

AC C

18 19

Stakeholder 1 - Industry

Stakeholder 2 - Municipality

Criteria

Environmental

Economic

Environmental

Economic

Environmental

1.00

0.25

1.00

7.00

Economic

4.00

1.00

0.14

1.00

Weights (%)

20

80

87

13

20 21

Table 10 shows that the economic net cost is a much greater concern for industry as

22

opposed to the municipality. Final weights are calculated by multiplying the weights of the

22

ACCEPTED MANUSCRIPT

1

environmental impact categories and the weight of environmental criterion itself. Since economic

2

criterion is not branched out into other sub-criteria, its initial and final weights are the same. In the next step, the sustainability index of each waste treatment option is calculated from

4

the impacts and final weights of the criteria. Game theory presents these sustainability indices in a

5

table format with stakeholder 1 (the cement industry) in the rows and stakeholder 2 (Metro

6

Vancouver) in the columns. In this study, game theory searches for pure strategy solutions (a

7

definitive solution rather than a mixture of solutions). The solution shows that RDF is a dominant

8

strategy for the cement industry as RDF always has a lower or equal sustainability index, in spite

9

of Metro Vancouver choosing any other option. WTE is also a strict dominant strategy for Metro

10

Vancouver. As a result, game theory analysis suggests that WTE and RDF are the best options for

11

Metro Vancouver and for the cement industry, respectively, considering the current assumptions

12

(Table 11). The industry should pay a tipping fee to convince the municipality to select the RDF

13

option.

M AN U

SC

RI PT

3

14 15

Table 11 - Game theory results for Metro Vancouver case study Stakeholder 2 Metro Vancouver

18

TE D

Co-combustion

combustion)

of RDF in cement kilns

Landfilling

0.50

0.09

0.50

-0.02

0.47

-0.01

WTE

0.50

0.09

0.47

-0.02

0.47

-0.01

RDF

0.05

0.09

0.47

-0.02

-0.24

-0.01

EP

17

WTE (Mass-

4.4. Mutual agreement on RDF option

AC C

16

Cement industry

Stakeholder 1

Landfilling

The tipping fee should be an amount that will convince the municipality to prefer RDF to

19

WTE, while keeping the industry interested in RDF. The developed framework suggests that if

20

the cement industry pays a tipping fee of $0.077-0.96 per kg waste to Metro Vancouver, both

21

stakeholders will benefit from a proposed new RDF plant in Vancouver, BC [7, 8]. This value is

22

calculated based on previous assumptions.

23

?$0.24 + 0.75 8 G $0.47 → 8 G $0.96 K7L

?$0.01 ? 0.13 8 G ?$0.02 → 8 N $0.077 K8L

where 8 is a positive tipping fee per kg MSW.

23

ACCEPTED MANUSCRIPT

For Metro Vancouver to efficiently plan future MSWM strategies, the relationship between

2

RDF’s sustainability index and independent variables was presented in a regression equation.

3

This regression will show how important the impact of uncertainties is on the outcome of the

4

framework. More detailed uncertainty assessment is required in the future to include more

5

sources of uncertainty. Waste composition and economic net cost carry parameter uncertainty due

6

to unavailability of complete and precise data. Therefore, waste composition components and

7

economic net cost versus final sustainability index for RDF was graphed. Figure 4 shows that in

8

producing RDF, economic net cost has the highest impact on the sustainability index of the

9

cement industry, while plastic has the highest positive impact on sustainability index of Metro

10

Vancouver. At all times, the sustainability index of the cement industry from RDF is higher than

11

that of Metro Vancouver.

5%

4%

3%

2%

M AN U

1%

-1 % Cu rre nt da ta

-2 %

-3 %

-4 %

-5 %

0

SC

RI PT

1

-0.01

-0.02

-0.03

-0.04

TE D

Sustainability Index

RDF for cement industry

Plastic Plastic Compostable Compostableorganics organics, Metals, Econ 1 Glass, Plastic Paper Compostable organics Economic net cost Econ 2

-0.05

12 13 14 15 16

Changes in the variable

AC C

-0.06

EP

RDF for Metro Vancouver

Figure 4 – Parameter uncertainty of RDF in Metro Vancouver case study

5. Summary and Conclusions MSWM is complex and can have high costs. Waste treatment strategy is the core of

17

MSWM, due to the significance of its environmental and economic impacts. To choose and apply

18

a sustainable waste treatment option, municipalities generally develop partnerships with other

19

stakeholders such as industries. In case of selecting a WTE treatment technology, this partnership

20

has a side of competition as the valuable product of energy enters the system. Therefore, reaching

21

a mutual agreement among multiple stakeholders with conflicting priorities is often a complicated

24

ACCEPTED MANUSCRIPT

1

affair. If a mutual agreement is not reached, the disposed waste is not directed into the most

2

mutually sustainable path. This study proposes a decision framework to assist two stakeholders with conflicting

4

priorities in evaluating environmental and economic – and whenever available, social – impacts

5

of selected waste treatment options and choosing the most mutually sustainable one. Sustainable

6

options are different from each stakeholder’s point of view, which results in sustainable options

7

to often not get selected by both stakeholders. This will result in a perfectly reasonable treatment

8

option to never get approved. This framework estimates stakeholders’ fair shares of costs and

9

benefits and helps them reach a mutual agreement on a single, optimal option that is

10

environmentally superior and economically feasible. The developed framework uses LCA, LCC,

11

and AHP to help stakeholders to compare various options, and then applies game theory to model

12

stakeholders’ actions, conflicts, and dialogues. The results of this framework will help

13

municipalities to explore more advanced and sustainable treatment options such as WTE

14

technologies and avoid the free-rider problem.

M AN U

SC

RI PT

3

The application of the developed framework was demonstrated with a case study of RDF in

16

Metro Vancouver. Two stakeholders of Metro Vancouver and the cement industry compared

17

sustainability impacts of landfilling, using mass-burn WTE, and co-combusting RDF in cement

18

kilns to investigate the possibility of collaborating on a new RDF treatment plant. As a result,

19

mass-burn WTE and RDF were selected as the most sustainable options for Metro Vancouver and

20

the cement industry, respectively. The developed framework estimated that a tipping fee of $0.77-

21

$0.96 per kg MSW should be paid by the cement industry to Metro Vancouver, so that RDF can

22

be applied as Metro Vancouver’s new waste treatment option and the cement industry’s solution

23

to high-intensity fossil fuel consumption. This value is only based on assumptions and factors

24

considered in this paper. Without the proposed framework, the collaboration might not take place,

25

or negotiations might be inefficient.

EP

There are various types of uncertainties (parameter, model, and scenario) in this framework

AC C

26

TE D

15

27

that will affect the robustness of the results. Parameter uncertainty refers to vague, incorrect, and

28

imprecise values such as composition of MSW in Metro Vancouver, input and output emission,

29

data derived from other databases and tools, as well as assumptions made for costs and benefits.

30

Model uncertainty discusses inaccurate hypotheses about the general model or relationship of

31

variables such as impact assessment outcomes and the game theory model. Scenario uncertainty

32

is concerned with the environment of assessments such as differences in the regions of collected

33

data and case study. These uncertainties should be explored in more detail in a future study. To

34

show the sensitivity of the outcomes of the framework to the uncertainties, a sensitivity analysis

25

ACCEPTED MANUSCRIPT

1

is performed. The assessment of the uncertainties in waste composition and economic net cost

2

shows that RDF is preferred by industry at all times. Also, sustainability indexes of industry and

3

the municipality are more impacted by economic net cost and share of plastic in waste

4

composition, respectively. Social impacts are difficult to assess quantitatively and carry the subjectivity of decision-

6

makers into the decision-making process. The developed framework has the ability to combine

7

the outcomes of social life cycle assessment methodologies with environmental and economic

8

assessment results according to the stakeholders’ priorities, but it does not discuss methodologies

9

for quantifying these impacts. In addition, there is no other case of co-combustion of RDF in

10

cement kilns in Canada, which makes the evaluation of social impacts prone to subjectivity, hard

11

to evaluate, and out of the scope of the case study. While social impacts are not quantified for the

12

discussed case study, they are still believed to impact municipality’s actions and priorities.

M AN U

SC

RI PT

5

The most challenging part of this framework was providing accurate data and real

14

scenarios that represented the waste treatment in a region and interactions among stakeholders.

15

Other limitations of this framework and potentials for future studies include consideration of

16

interest rate in LCC, consideration of additional costs of RDF in LCC, using other software

17

packages for LCA, using real stakeholders for pair-wise comparisons, expansion of two-player

18

game into n-player game, and consideration of other game theory models (e.g. uncertainty about

19

other stakeholders’ actions) in the framework.

20

TE D

13

Acknowledgements

22

We would like to thank NSERC for the financial support from second and third authors’ NSERC-

23

DG grants. We also appreciate the help of Ms. Sarah Wellman at Metro Vancouver, Ms. Yihting

24

Lim at TRI Environmental Consulting, and Mr. Navid Hossaini and Mr. Fasihur Rahman at UBC.

26

AC C

25

EP

21

26

ACCEPTED MANUSCRIPT

TE D

M AN U

SC

RI PT

Achillas, C., Moussiopoulos, N., Karagiannidis, A., Banias, G., & Perkoulidis, G. (2013). The use of multi-criteria decision analysis to tackle waste management problems: a literature review. Waste Management & Research, 31(2), 115–29. doi:10.1177/0734242X12470203 Bare, J. (2002). Developing a consistent decision-making framework by using the US EPA’s TRACI. American Institute of Chemical Engineers Symposium. Retrieved from http://www.epa.gov/nrmrl/std/traci/aiche2002paper.pdf BC Ministry of Finance. (2014). How the Carbon Tax Works. Retrieved January 15, 2015, from http://www.fin.gov.bc.ca/tbs/tp/climate/A4.htm Bebbington, J., Brown, J., & Frame, B. (2007). Accounting technologies and sustainability assessment models. Ecological Economics, 61(2-3), 224–236. doi:10.1016/j.ecolecon.2006.10.021 Blanchard, Benjamin S., Wolter J. Fabrycky, and W. J. F. (1990). Systems engineering and analysis (Volume 4.). Englewood Cliffs, New Jersey: Prentice Hall. Brundtland, G. H. (1987). Our common future. In World Commission on Environment and Development. Tokyo, Japan. Canter, L. W. (1977). Environmental impact assessment. New York: McGraw-Hill, Inc. Caputo, A. C., Scacchia, F., & Pelagagge, P. M. (2003). Disposal of by-products in olive oil industry: waste-to-energy solutions. Applied Thermal Engineering, 23(2), 197– 214. doi:10.1016/S1359-4311(02)00173-4 Carlsson Reich, M. (2005). Economic assessment of municipal waste management systems—case studies using a combination of life cycle assessment (LCA) and life cycle costing (LCC). Journal of Cleaner Production, 13(3), 253–263. doi:10.1016/j.jclepro.2004.02.015 Carraro, C., & Marchiori, C. (2003). Endogenous Strategic Issue Linkage in International Negotiations. NOTA DI LAVORO, 40(April). Retrieved from http://www.feem.it/userfiles/attach/Publication/NDL2003/NDL2003-040.pdf Cheng, S., Chan, C. W., & Huang, G. H. (2003). An integrated multi-criteria decision analysis and inexact mixed integer linear programming approach for solid waste management. Engineering Applications of Artificial Intelligence, 16, 543–554. doi:10.1016/S0952-1976(03)00069-1 Cheng, Steven, Chan, C. W., & Huang, G. H. (2002). USING MULTIPLE CRITERIA DECISION ANALYSIS FOR SUPPORTING DECISIONS OF SOLID WASTE MANAGEMENT. Environmental Science and Health , Part A : Toxic / Hazardous Substances and Environmental Engineering, A37(6), 37–41. Chhipi-Shrestha, G. K., Hewage, K., & Sadiq, R. (2014). “Socializing” sustainability: a critical review on current development status of social life cycle impact assessment method. Clean Technologies and Environmental Policy. doi:10.1007/s10098-0140841-5 City of Vancouver. (2013). Fees and charges at the Transfer Station and Landfill. Retrieved November 15, 2014, from http://vancouver.ca/home-propertydevelopment/landfill-fees-and-charges.aspx Contreras, F., Hanaki, K., Aramaki, T., & Connors, S. (2008). Application of analytical hierarchy process to analyze stakeholders preferences for municipal solid waste

EP

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

References

AC C

1

27

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

management plans, Boston, USA. Resources, Conservation and Recycling, 52(7), 979–991. doi:10.1016/j.resconrec.2008.03.003 D-waste. (2015). Waste Atlas. Retrieved June 18, 2015, from http://www.atlas.dwaste.com Davila, E., Chang, N.-B., & Diwakaruni, S. (2005). Landfill space consumption dynamics in the Lower Rio Grande Valley by grey integer programming-based games. Journal of environmental management, 75(4), 353–65. doi:10.1016/j.jenvman.2004.11.015 De Feo, G., & De Gisi, S. (2010). Using an innovative criteria weighting tool for stakeholders involvement to rank MSW facility sites with the AHP. Waste management (New York, N.Y.), 30(11), 2370–82. doi:10.1016/j.wasman.2010.04.010 De la Fuente, A., Pons, O., Josa, A., & Aguado, A. (2015). Multicriteria–decision making in the sustainability assessment of sewerage pipe systems. The journal of Cleaner Production, In press. doi:10.1016/j.jclepro.2015.07.002. Den Boer, J., den Boer, E., & Jager, J. (2007). LCA-IWM: a decision support tool for sustainability assessment of waste management systems. Waste management (New York, N.Y.), 27(8), 1032–45. doi:10.1016/j.wasman.2007.02.022 Dewi, O., Koerner, I., & Harjoko, T. (2010). A Review on Decision Support Models for Regional Sustainable Waste Management. In the International Solid Waste Association World Conference. Retrieved from http://www.iswa.org/uploads/tx_iswaknowledgebase/Candra_Dewi.pdf Environment Canada. (2012). Municipal Solid Waste. Retrieved June 11, 2013, from http://www.ec.gc.ca/gdd-mw/default.asp?lang=En&n=EF0FC6A9-1 Erkut, E., Karagiannidis, A., Perkoulidis, G., & Tjandra, S. a. (2008). A multicriteria facility location model for municipal solid waste management in North Greece. European Journal of Operational Research, 187(3), 1402–1421. doi:10.1016/j.ejor.2006.09.021 Finus, M. (2000). Game Theory and International Environmental Cooperation : Any Practical Application ? In Controlling Global Warming: Perspectives from Economics, Game Theory and Public Choice (pp. 9–104). Gendebien, A., Leavens, A., Blackmore, K., Godley, A., Lewin, K., Whiting, K. J., & Davis, R. (2003). Refused derived fuel, current practice and perspectives: final report. Current Practice. Genon, G., & Brizio, E. (2008). Perspectives and limits for cement kilns as a destination for RDF. Waste management (New York, N.Y.), 28(11), 2375–85. doi:10.1016/j.wasman.2007.10.022 Gloria, T. P., Lippiatt, B. C., & Cooper, J. (2007). Life cycle impact assessment weights to support environmentally preferable purchasing in the United States. Environmental science & technology, 41(21), 7551–7. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/18044540 Gluch, P., & Baumann, H. (2004). The life cycle costing (LCC) approach: a conceptual discussion of its usefulness for environmental decision-making. Building and Environment, 39(5), 571–580. doi:10.1016/j.buildenv.2003.10.008 Hoornweg, D., & Bhada-Tata, P. (2012). WHAT A WASTE A Global Review of Solid Waste Management. Urban Development Series (p. 116). Retrieved from http://go.worldbank.org/BCQEP0TMO0

AC C

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

28

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

Hossaini, N., Reza, B., Akhtar, S., Sadiq, R., & Hewage, K. (2014). AHP based life cycle sustainability assessment (LCSA) framework: a case study of six storey wood frame and concrete frame buildings in Vancouver. Journal of Environmental Planning and Management, (August 2014), 1–25. doi:10.1080/09640568.2014.920704 the International Organization for Standardization (ISO). (2006a). 14040 Environmental management-- Life cycle assessment -- Principles and framework. Geneva: International Organization for Standardization. Retrieved from http://www.iso.org/iso/catalogue_detail?csnumber=37456 the International Organization for Standardization (ISO). (2006b). 14044 Environmental management -- Life cycle assessment -- Requirements and guidelines. Retrieved from http://www.iso.org/iso/catalogue_detail?csnumber=38498 the International Organization for Standardization (ISO). (2012). TR 14047 Environmental management -- Life cycle assessment -- Illustrative examples on how to apply ISO 14044 to impact assessment situations. Jørgensen, S. (2010). A dynamic game of waste management. Journal of Economic Dynamics and Control, 34(2), 258–265. doi:10.1016/j.jedc.2009.09.005 Kaitala, V., & Pohjola, M. (1995). Sustainable International Agreement on Greenhouse Warming- A Game Theory Study. In C. et Al. (Ed.), Control and Game-Theoretic Models of the Environment. Boston. Karmperis, A. C., Aravossis, K., Tatsiopoulos, I. P., & Sotirchos, A. (2013). Decision support models for solid waste management: Review and game-theoretic approaches. Waste management (New York, N.Y.). doi:10.1016/j.wasman.2013.01.017 Leyton-Brown, K., & Shoham, Y. (2008). Essentials of game theory. (R. J. Brachman & T. Dietterich, Eds.)Political Science (Vol. 2, p. 104). Morgan & Claypool Publishers. doi:10.2200/S00108ED1V01Y200802AIM003 McGinty, M., Milam, G., & Gelves, A. (2012). Coalition Stability in Public Goods Provision: Testing an Optimal Allocation Rule. Environmental and Resource Economics, 52(3), 327–345. doi:10.1007/s10640-011-9530-6 Metro Vancouver. (2010). Recycling and Solid Waste Management: 2010 Report. Retrieved from http://www.metrovancouver.org/about/publications/Publications/2010_Solid_Waste _Management_Annual_Summary.pdf Metro Vancouver. (2013). ZERO WASTE COMMITTEE REGULAR MEETING. Burnaby, BC. Retrieved from http://www.metrovancouver.org/boards/ZeroWaste/Zero_Waste_CommitteeJune_6_2013-Agenda-Revised_2.pdf Metro Vancouver. (2014a). Waste to Energy: Just one part of a forward looking approach to waste management. Retrieved November 19, 2014, from http://www.metrovancouver.org/services/solid-waste/about/wte/pages/index.aspx Metro Vancouver. (2014b). Frequently asked questions. Metro Vancouver. (2014c). 2013 Waste Composition Monitoring Program. Burnaby, BC. Retrieved from http://www.metrovancouver.org/about/publications/Publications/2013_Waste_Comp osition_Report.pdf

AC C

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

29

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

Metro Vancouver. (2014d). Metro Vancouver’s New Waste-to-Energy Capacity Project. Retrieved from http://www.metrovancouver.org/services/solidwaste/planning/Recover/WTE/WTEP roponents/LehighCement.pdf Moretti, S. (2004). A model for cooperative inter-municipal waste collection: cost evaluation toward fair cost allocation. In C. Carraro & V. Fragnelli (Eds.), Game Practice and the Environment. Edward Elgar. Munda, G. (2002). Social multi-criteria evaluation. In the 4th UFZ Summer Symposium “New Strategies for Solving Environmental Conflicts: Potentials for Combining Participation and Multicriteria Analysis.” Leipzig. Nagarajan, M., & Sošić, G. (2008). Game-theoretic analysis of cooperation among supply chain agents: Review and extensions. European Journal of Operational Research, 187(3), 719–745. doi:10.1016/j.ejor.2006.05.045 Neugebauer, S., Martinez-Blanco, J., Scheumann, R., & Finkbeiner, M. (2015). Enhancing the practical implementation of life cycle sustainability assessment e proposal of a Tiered approach. The journal of Cleaner Production, 102, 165–176. Nithikul, J. (2007). POTENTIAL OF REFUSE DERIVED FUEL PRODUCTION FROM BANGKOK MUNICIPAL SOLID WASTE. Asian Institute of Technology. Osborne, M. J., & Rubinstein, A. (1994). A course in Game theory. 1994. Cambridge, MA: MIT Press. PRe. (2015). SimaPro Database Manual: Methods library. Retrieved from http://www.pre-sustainability.com/download/DatabaseManualMethods.pdf Reza, B., Soltani, A., Ruparathna, R., Sadiq, R., & Hewage, K. (2013). Environmental and economic aspects of production and utilization of RDF as alternative fuel in cement plants: A case study of Metro Vancouver Waste Management. Resources, Conservation and Recycling, 81, 105–114. doi:10.1016/j.resconrec.2013.10.009 Rotter, V. S., Kost, T., Winkler, J., & Bilitewski, B. (2004). Material flow analysis of RDF-production processes. Waste management (New York, N.Y.), 24(10), 1005–21. doi:10.1016/j.wasman.2004.07.015 Saaty, T. (1980). The Analytic Hierarchy Process: Planning, Priority Setting, Resource Allocation (Decision Making Series). New York: McGraw Hill. Saaty, T. L. (1988). What is the analytic hierarchy process? In G. Mitra (Ed.), Mathematical Models for Decision Support: Proceedings of the NATO Advanced Study Institute on Mathematical Models for Decision Support (Volume 48.). Berlin Heidelberg: Springer-Verlag Berlin Heidelberg. Saaty, T. L. (1994). How to make a decision: the analytic hierarchy process. Interfaces, 24(6), 19–43. Sadiq, R. (2001). Drilling waste discharges in the marine environment: a risk based decision methodology. Memorial University of Newfoundland. Retrieved from http://research.library.mun.ca/1250/ Sadiq, Rehan, Husain, T., Veitch, B., & Bose, N. (2003). Evaluation of Generic Types of Drilling Fluid Using a Risk-Based Analytic Hierarchy Process. Environmental Management, 32(6), 778–787. Schwarz, J., Beloff, B., & Beaver, E. (2002). Use Sustainability metric to guid decision making. Chemical Engineering …, (July), 58–63. Retrieved from http://people.clarkson.edu/~wwilcox/Design/sustain.pdf

AC C

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

30

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

Sheppard, S. R. J., & Meitner, M. (2005). Using multi-criteria analysis and visualisation for sustainable forest management planning with stakeholder groups. Forest Ecology and Management, 207(1-2), 171–187. doi:10.1016/j.foreco.2004.10.032 Society of Environmental Toxicology and Chemistry (SETAC). (1993). Guidelines for Life-cycle Assessment: A Code of Practice. Portugal. doi:First edition Society of Environmental Toxicology and Chemistry (SETAC). (2004). Environmental Risk Assessment: SETAC Technical Issue Paper. Pensacola, FL. doi:Third Printing Soltani, A., Hewage, K., Reza, B., & Sadiq, R. (2015). Multiple stakeholders in multicriteria decision-making in the context of Municipal Solid Waste Management: A review. Waste management (New York, N.Y.), 35, 318–328. doi:10.1016/j.wasman.2014.09.010 Spengler, T., Geldermann, J., Hähre, S., Sieverdingbeck, A., & Rentz, O. (1998). Development of a multiple criteria based decision support system for environmental assessment of recycling measures in the iron and steel making industry. Journal of Cleaner Production, 6(1), 37–52. doi:http://dx.doi.org/10.1016/S09596526(97)00048-6 Statistics Canada. (2011). Spending on Waste Management. Retrieved November 19, 2014, from http://www.statcan.gc.ca/pub/11-402-x/2011000/chap/gov-gouv/govgouv01-eng.htm Statistics Canada. (2015). Disposal and diversion of waste, by province and territory (Total waste disposal). Sutton, P. C., Anderson, S. J., Tuttle, B. T., & Morse, L. (2012). The real wealth of nations: Mapping and monetizing the human ecological footprint. Ecological Indicators, 16, 11–22. doi:10.1016/j.ecolind.2011.03.008 Tesfamariam, S., & Sadiq, R. (2006). Risk-based environmental decision-making using fuzzy analytic hierarchy process (F-AHP). Stochastic Environmental Research and Risk Assessment, Springer, 21(1), 35–50. The government of British Columbia. (2015). Part 3 — Municipal Waste Management. The Environmental Management Act. Retrieved July 08, 2015, from http://www.bclaws.ca/civix/document/LOC/complete/statreg/-- E --/Environmental Management Act [SBC 2003] c. 53/00_Act/03053_03.xml The World Bank. (2012). What a Waste: A Global Review of Solid Waste Management. Retrieved from http://go.worldbank.org/BCQEP0TMO0 TRI Environmental Consulting Inc. (2008). SOLID W ASTE COMPOSITION S TUDY for Metro Vancouver. Retrieved from http://www.metrovancouver.org/about/publications/Publications/SolidWasteCompos itionStudyFinal-2007.pdf Tseng, M.-L., Lin, Y.-H., & Chiu, A. S. F. (2009). Fuzzy AHP-based study of cleaner production implementation in Taiwan PWB manufacturer. The journal of Cleaner Production, 17, 1249–1256. Tsilemou, K., & Panagiotakopoulos, D. (2006). Approximate cost functions for solid waste treatment facilities. Waste management & research, 24(4), 310–322. United states Environmental Protection Agency (USEPA). (2006). LIFE CYCLE ASSESSMENT: PRINCIPLES AND PRACTICE. Cincinnati, OH. doi:EPA/600/R06/060

AC C

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

31

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

Van den Hove, S. (2006). Between consensus and compromise: acknowledging the negotiation dimension in participatory approaches. Land Use Policy, 23(1), 10–17. doi:10.1016/j.landusepol.2004.09.001 Weikard, H.-P., & Dellink, R. B. (2008). Sticks and carrots for the design of international climate agreements with renegotiations. Annals of Operation Research, 1–20. World Health Organization (WHO). (2014). 7 million premature deaths annually linked to air pollution. Retrieved January 27, 2015, from http://www.who.int/mediacentre/news/releases/2014/air-pollution/en/

AC C

1 2 3 4 5 6 7 8 9

32

ACCEPTED MANUSCRIPT

Highlights Journal of Cleaner Production

Atousa Soltani*, Rehan Sadiq, and Kasun Hewage Selecting a sustainable and pragmatic waste-to-energy technology is complicated.

•

Stakeholders have conflicting priorities over economic and environmental criteria.

•

Proposed framework guides stakeholders to reach a mutually sustainable decision.

•

A Sustainable decision is reached when costs are fairly shared among

RI PT

•

•

SC

stakeholders.

Production of Refuse-derived fuel in Vancouver can be sustainable and

AC C

EP

TE D

M AN U

pragmatic.