water corporation mundaring water treatment plant site selection

Water Corporation

Mundaring Water TreatmentPlant Site Selection

MultiCriteria Analysis (MCA)Report

September 2007

61/16611/71624 Mundaring Water Treatment Plant Site SelectionMultiCriteria Analysis (MCA) Report

Contents

1. Introduction 1

2. Overview of MCA 3

3. Planning the MCA 4

4. Sustainability Criteria 5

5. Impact Assessment and Scoring 8

6. Weighting 12

7. Results of MCA 15

8. Conclusions 24

9. References 26

Figure IndexFigure 1: Location of Potential WTP Sites 2

Table IndexTable 1 Comparison of draft and final sustainability criteria 6Table 2 Comparison of Community and Water Corporation

sustainability criteria 6Table 3 Summary of Scoring Process 9Table 4 Summary of Scores – 1 (Best) to 6 (Worst) 10Table 5 Guidance for Allocating Weights 13Table 6 Comparison of Average Normalised Weights 14Table 7 MCA Ranking using Average Community Weights 16Table 8 MCA Ranking using Average Water Corporation

Weights 17Table 9 Summary of MCA Results using Individual

Community Weights 20Table 10 Summary of MCA Results using Individual Water

Corporation Weights 21


AppendicesA Making Good DecisionsB Attendees at Water Corporation Preliminary Workshop 8th May

2007C Summary of SitesD Community AmenityE Recreation and Tourism ValuesF Biodiversity and Potential for Land DegradationG Site Flexibility and OperabilityH Waterways and Water CyclesI Community Weighting Workshop MaterialJ Water Corporation Weighting Workshop AttendeesK Community and Water Corporation WeightsL Worked Examples of Normalisation and Concordance Analysis


1. Introduction

Through an extensive planning and community consultation process during which alarge number of sites was considered, Water Corporation has identified four possiblelocations for the proposed new Mundaring Water Treatment Plant (WTP). The sitesremaining under consideration are:

» Site 1: Below Mundaring Weir, in the Heritage and Tourist Precinct, south of theHelena River;

» DEC land: Land used by the Department of Environment and Conservation(formerly the Department of Conservation and Land Management) for their officesand depot;

» Pine Plantation: Site adjacent to DEC airstrip;

» O’Connor Site (added to the list after the Initial Selection process).

The location of the four sites is shown in Figure 1.

The key steps undertaken to identify these sites were:

» Initial Selection (April 2006) through which the number of sites was reduced from 21to 11 through the application of initial selection criteria;

» Community Forum 7 (November 2006), through which members of the communitywere invited to filter the remaining sites to a smaller number and then assess theseusing eight criteria to rank them in order of preference;

» Internal Water Corporation ranking process (February 2007) through which theremaining sites were evaluated against internal sustainability criteria.

At this point, GHD was engaged to facilitate a sustainability assessment process toassist Water Corporation in distinguishing between the four remaining sites andidentifying its preferred site. This decision is complex since Water Corporation iscommitted to taking into consideration a large number of issues, which reflect the fullspectrum of sustainability. For these reasons, a structured multicriteria analysis (MCA)was adopted as the most appropriate methodology for the sustainability assessment.

This report documents the MCA process facilitated by GHD and reports on its findings.


Figure 1: Location of Potential WTP Sites


2. Overview of MCA

MCA is ideally suited to complex decisionmaking, since it provides a structuredframework for managing and evaluating large amounts of information. One feature ofMCA that makes it different from the other methodologies applied in the site selectionprocess to date is that it includes a step of weighting the criteria, that is, making avaluebased decision as to how important a particular criterion is in relation to eachother criterion. This feature means that MCA is potentially a more powerful tool toselect between the remaining four options.

The basic elements of any MCA process are:

» Planning the MCA, including deciding which of a number of available analyticalmethods should be used and who should be involved;

» Identifying criteria (in this case sustainability criteria) against which each option is tobe assessed;

» Scoring each option according to each criterion, i.e. undertaking impactassessments to determine how well each option performs against each criterion;

» Weighting each criterion, that is, making a valuebased decision as to howimportant a particular criterion is in relation to each other criterion;

» Analysing the results by combining the scores and weights for each criterion togenerate a picture of the overall sustainability performance of the options and torank them in relation to each other, incorporating sensitivity analysis for thepurposes of assessing the significance of any data uncertainties and for refiningand enhancing the options.

It is important to note that MCA is a decisionaiding, not a decisionmaking tool, andwhile it can provide detailed information to Water Corporation, responsibility for thefinal selection of a preferred site rests with the Water Corporation Executive.Furthermore, while the MCA process provides a detailed picture of the respectivesustainability implications of the different site options, it does not indicate whether theresidual impacts are acceptable or not. Water Corporation must consider theseresidual impacts and other implications of its decision in the process of selecting thepreferred site.

As an additional contribution to the decisionmaking process, it was determined at theMundaring Site Selection Project Team meeting held on 8th May 2007 that, separatelyfrom the MCA process, members of the Community Forum would also be offered theopportunity to indicate their own preferred site based on the detailed information thatWater Corporation has provided to them. This process was outside the GHD Scope ofWork and is not discussed further in this report; however, its outcomes will provideWater Corporation with further important information to assist in the selection of thepreferred site.

The outcomes of each stage of the MCA process are discussed in the followingsections.


3. Planning the MCA

MCA is a generic term that encompasses a wide range of techniques, and theselection of the most appropriate MCA methodology depends to a large extent on thenature of the decision, the criteria selected and the data that is available. GHDengaged an experienced MCA practitioner from Murdoch University to help select themost appropriate methodology for the Mundaring WTP site selection, taking intoconsideration the history of the project and the nature of the data available.

Through this process it was determined that concordance analysis would be used inthis case. Concordance analysis is a technique in which all mathematical functions,apart from the comparison of scores, are conducted on the weighting data alone andnot on a combination of weights and scores. This was considered appropriate giventhe qualitative nature of much of the data. Further information on different MCAtechniques and their mathematical algorithms applied is available in Appendix A.

A Preliminary Workshop, convened and facilitated by GHD, was conducted at WaterCorporation offices on 8th May 2007 for the purpose of planning the MCA process, andseveral other decisions important decisions were made at this workshop. Theattendees at the Preliminary Workshop are listed in Appendix B.

Firstly it was confirmed that all four of the remaining sites would be included in the finalstage assessment because of community preferences, despite the known challengesassociated with some of the sites. In particular, since Water Corporation had beenadvised that the chlorine plant for Site 1 would have to be located remotely from thetreatment plant for safety reasons, and no other sites beyond the four listed above areto be considered at this stage, this meant that three variations of the Site 1 option wereincluded in the MCA:

» Site 1 + chlorine and drying beds at the Pine Plantation

» Site 1 + chlorine and drying beds at DEC land

» Site 1 + chlorine and drying beds at the O’Connor site

Secondly, it was determined that the community would be offered an opportunity tocontribute to the MCA process, primarily through the weighting step whereby therelative significance of each criterion is evaluated, and would also be invited to assistwith scoring appropriate criteria. Accordingly, GHD personnel presented an overview ofthe proposed MCA methodology to the Community Forum on 14th May 2007, basedupon the planning decisions made to date.


4. Sustainability Criteria

The first stage of the MCA process was the identification of appropriate criteria againstwhich the performance of each option could be assessed. Criteria should serve twopurposes: to allow differentiation between the site options, and to reflect significantimpacts. It is also important for the integrity of the MCA process that the set of criteriaused have the following characteristics (refer to Annandale and Lantzke (2000) inAppendix A):

» Complete, meaning that there should be no additional basis other than the definedcriteria for distinguishing between options;

» Operational, that is able to be practically applied;

» Decomposable, meaning that each criterion should be able to be analysedindependently of all others;

» Nonredundant, meaning that criteria should not overlap and therefore result in‘double accounting’ (to the extent possible);

» Minimal, that is the smallest number of criteria possible while still embodying theprevious characteristics.

A draft set of sustainability criteria for the MCA process was developed at thePreliminary Workshop on 8th May 2007, to reflect Water Corporation SustainabilityPrinciples, criteria used in earlier stages of the process (particularly those defined andused by the community); and issues associated with the four sites.

This list was subsequently refined by GHD to ensure the final set of criteria reflectedthe characteristics listed above. Two minor changes were made: it was decided that‘soil structure’ and the ‘introduction of diseases and weeds’ could be combined into thesingle criterion of ‘potential for land degradation’; and ‘site flexibility’ was extended to‘site flexibility and operability’ to ensure that the implications of the site selection for theoperational phase, as well as planning and design phases of the project, were takeninto consideration. The draft and the final sustainability criteria are presented in Table1.

Detailed descriptions of the final Water Corporation sustainability criteria wereprepared by GHD as part of the information packages mailed to approximately 200members of the Mundaring Community Forum by the Water Corporation in early July2007. This document is presented as Appendix C.


Table 1 Comparison of draft and final sustainability criteria

Draft sustainability criteria Final sustainability criteria

Environmental

Biodiversity

Embedded (embodied) energy of assets

Waterways and water cycles

Introduction of diseases and weeds

Soil structure

Environmental

Biodiversity



Potential for land degradation

Social

Hazardous chemical risks

Community amenity (visual, noise, light,traffic)

Recreational and tourism values

European heritage

Indigenous heritage

Social




European heritage

Indigenous heritage

Economic

Site flexibility

Costs

Economic

Site flexibility and operability

Costs

Particular efforts were made to ensure that the community criteria applied atCommunity Forum 7 were reflected within the Water Corporation’s criteria. Therelationship between the two sets of criteria is presented in Table 2 (noting that the firstsix community criteria were considered ‘essential’ while the last two were designated‘desirable’).

Table 2 Comparison of Community and Water Corporation sustainabilitycriteria

Community criteria Water Corporation final criteria

Safety for residents Hazardous chemical risks

No compulsory acquisition of privateproperty

Deleted – as not relevant to this stage ofthe process since no remaining sites areprivately owned

Least disruptive and best practiceinfrastructure


Least environmental impact Biodiversity


Community criteria Water Corporation final criteria




Limit amenity impact for residents Community amenity (visual, noise, light,traffic)

Social, heritage and cultural Recreational and tourism values

Indigenous heritage

European heritage

Economic and financial Site flexibility and operability

Cost

Status of sites (national parks, RPZ) Included under biodiversity


5. Impact Assessment and Scoring

In this step, the impacts with respect to each criterion were firstly assessed and thengiven a score reflecting the magnitude of the predicted impact. It was important for theintegrity of the MCA process that the impact assessment and scoring for each criterionbe conducted by appropriate specialists, that is, the people who have the necessaryexpertise and experience to accurately predict and evaluate the potential impacts. Theappropriate specialists in each case were identified by Water Corporation, inconsultation with GHD, and included Water Corporation personnel, specialistconsultants engaged by Water Corporation, members of affected groups andorganisations, and members of the Mundaring community.

The selection of concordance analysis as the appropriate MCA methodology in thiscase (as discussed in Section 3) meant that scoring could be undertaken using asimple ordinal scale; in other words, for each criterion the sites were ranked in order ofperformance from best to worst.

The scoring processes for four of the criteria were undertaken in group workshops,facilitated by GHD and conducted in Mundaring, in which all identified specialists foreach criterion participated:

» Community amenity 19th July 2007

» Recreational and tourism values 19th July 2007

» Biodiversity 23rd July 2007

» Potential for land degradation 23rd July 2007

Information packs were posted to workshop participants prior to these workshopsexplaining the MCA process, the relevant criterion, and what would be asked ofworkshop participants. Following the workshops, summary reports of each of theprocesses were prepared by GHD. These summary reports are presented inAppendices D (Community amenity), E (Recreational and tourism values), and F(Biodiversity and potential for land degradation).

Scoring for site flexibility and operability was undertaken in a workshop involving WaterCorporation engineers, conducted on 6th July 2007. This process involved identifyingeight subcriteria and then evaluating the sites against each subcriterion. A summaryreport describing this process and its outcomes is presented in Appendix G.

Scoring for waterways and water cycles was conducted by GHD together withrepresentatives from the Department of Water and the Conservation Council ofWestern Australia. A summary of this process is presented in Appendix H.

Specialist consultants engaged by Water Corporation conducted the impactassessments for hazardous chemical risks, indigenous heritage and Europeanheritage. At Water Corporation’s request, the consultants included scores (i.e. rankingof the sites in terms of performance against the relevant criteria) in their reports. It isimportant to note that at this stage in the process the indigenous heritage assessment


was based only on an archaeological survey, while ethnographic studies are to beconducted at a later stage when the preferred site has been selected.

Water Corporation engineers calculated and scored the costs and embedded(embodied) energy implications of each option.

Table 3 summarises who was involved in the scoring process for each criterion andprovides links to the impact assessment data upon which scoring was based. Asummary of impacts was included in the information package mailed to members ofthe Mundaring Community Forum, which is included as Appendix C.

Table 3 Summary of Scoring Process

Criterion Specialists involved inscoring

Further impact data

1. Biodiversity GHD (Ecology group) andrepresentatives ofenvironment groups

Appendix D

GHD (2007a)

2. Potential for landdegradation

GHD (Ecology group) andrepresentatives ofenvironment groups

Appendix D

GHD (2007a)

3. Embedded (embodied)energy of assets

Water Corporation Appendix C

4. Waterways and watercycles

GHD (Ecology group),Department of Water andConservation Council of WA

Appendix D

5. Hazardous chemicalrisks

GHD (GHD’s risk specialists) GHD (2007b)

6. Community amenity(visual, noise, light, traffic)

Community representatives Appendix E

7. Recreational and tourismvalues

Managers of local tourismand recreational facilities

Appendix F

8. European heritage Heritage and ConservationProfessionals

Heritage andConservationProfessionals (2007)

9. Indigenous heritage(archeological)

Australian InteractionConsultants

Australian InteractionConsultants (2007)

10. Site flexibility andoperability

Water Corporation Appendix G

11. Costs Water Corporation Appendix C

A summary of the final scores was provided to attendees at Community Forum 9 on30th July 2007, although the scoring for waterways and water cycles had not beencompleted at this point. The final scores are presented in Table 4.


Table 4 Summary of Scores – 1 (Best) to 6 (Worst)

Criterion DECland

PinePlantation

O’Connorsite

Site 1+DECland

Site 1+PinePlantation

Site 1+O’Connorsite

1a. Biodiversity (4) 1 3 6 2 4 5

1b. Biodiversity (2) 1 2 6 3 4 5

1c. Biodiversity (1) 2 3 6 1 4 5

2a. Potential for landdegradation (6)

1 4 6 2 3 5

2b. Potential for landdegradation (1)

2 5 6 1 3 4

3. Embedded energy ofassets

3 5 6 2 4 1

4. Waterways andwater cycles

1 5 2 3 6 4

5. Hazardous chemicalrisks

3 1 2 3 1 2

6a. Community amenity(7)

5 1 6 3 2 4

6b. Community amenity(3)

3 2 1 6 4 5

6c. Community amenity(1)

6 1 5 4 2 3

7a. Recreational andtourism values

3 2 1 4 4 4

7b. Recreational andtourism values

3 1 2 4 4 4

8. European heritage 3 2 1 4 4 4

9. Indigenous heritage(archeological)

3 2 4 1 1 1

10. Site flexibility andoperability

1 3 1 2 2 2

11. Costs 4 6 5 2 3 1

Some aspects of the summary of scores require further explanation, as follows:

» In most cases the scores distinguish between six possible locations/configurations,reflecting the three possible locations for the chlorination plant for Site 1, and


therefore scores run from 1 (best) to 6 (worst). In a few cases, however, therelevant specialists did not deem it necessary to distinguish between the differentconfigurations of Site 1 and therefore scores run from 1 (best) to 4 (worst);

» In some cases where scoring was conducted in a workshop, consensus among thegroup could not be achieved and therefore several different sets of scores havebeen recorded. The numbers in brackets indicate the number of people supportingeach variation;

» In the case of recreational and tourism values, two different traffic access routes tothe Pine Plantation site were assessed (see Appendix E). The scoring identified as7a in Table 4 assumed the currently proposed access via Allen Rd, whereas thescoring identified as 7b was undertaken assuming that an alternative, morenortherly route could be found.

» Although the preliminary risk assessment (GHD, 2007b) included an assessment ofthe risks associated with locating the chlorination plant at Site 1, the scores given inTable 4 reflect the Water Corporation’s decision that should Site 1 be selected thechlorination plant would be located remotely at one of the other three sites.Therefore, the scores for each site is the same as the score for the combination ofSite 1 and that site.


6. Weighting

Weighting is the process by which the relative significance of each criterion isevaluated relevant to each other criterion. This process of deriving weights isfundamental to the effectiveness of a multicriteria analysis process as it involvesindividual perspectives and values that vary from person to person.

Water Corporation personnel, many of whom had not been directly involved in theMundaring WTP site selection process, were invited to participate in a weightingworkshop on 24th July 2007. The attendees at this workshop are listed in Appendix J.The community weighting process was conducted at Community Forum 9 on 30th July2007.

It is important for the overall integrity of the MCA process that participants in theweighting workshops understand both the weighting process to be followed and thenature of the criteria they are weighting. Considerable time was therefore spent at bothworkshops explaining these two aspects and responding to questions fromparticipants. It was also emphasised that the weighting should be conducted in thecontext of the purpose of the process, namely to select the best site for the MundaringWTP.

In order to generate ratioscaled data, the weighting technique of “modified pairwisecomparison” was used, in which each criterion is compared with each other criterion.Specifically, participants were asked to:

» Identify any criteria they considered not at all important and allocate these a weightof zero;

» Identify the least important criterion of those remaining and allocate this a weightingof 1;

» Weight each remaining criterion in relation to on a scale of 110 according to howmany times more important it is than the criterion/criteria with a score of 1, to giveratioscaled data. For example, this means that a criterion with a weight of 2 isconsidered to be twice as important as a criterion with a weight of 1, and half asimportant as a criterion with a weight of 4.

The table reproduced below as Table 5 was provided as guidance to participants.


Table 5 Guidance for Allocating Weights

Noimportancewhatsoever

Lowestimportance

A little moreimportant

Moderatelymore

important

Stronglymore

important

Verystrongly

moreimportant

Extremelymore

important

0 1 2 3 4 5 6 7 8 9 10

Participants in both workshops were provided with a handout describing the criteriaand the weighting process. This document is included in Appendix I.

The weighting sheets provided to participants in the community weighting processwere numbered to provide an audit trail should individuals wish to check that theirweights were correctly transferred. A total of 38 members of the community and 10Water Corporation representatives completed weighting worksheets. Appendix Kanonymously presents the weights that individual attendees attributed to the variouscriteria, and also documents the participants in the Water Corporation weightingworkshop.

It was also explained to participants that weights would be normalised to ensure thateach person’s weights have the same influence on the final results, so that it would notmatter whether participants use the full range of 110 or a smaller range such as 15 inallocating weights. Normalisation involves adding up the total weights allocated by anindividual and then expressing the weight for each criterion as a fraction of the totalweight. This means that the weights for all participants are expressed as a fraction of 1(where 1 = 100%). A worked example of the normalisation process is given inAppendix L.

The average normalised weights allocated by the community and Water Corporationare compared in Table 6. Several observations can be made about the data in Table 6:

» Biodiversity was weighted as the most important criterion by both the communityand Water Corporation, and the normalised weight was similar in both cases (0.13rounded to two decimal places);

» Potential for land degradation and waterways and water cycles were also ratedhighly and very similarly by both the community and Water Corporation;

» The community and Water Corporation weighted both European and indigenousheritage as being of moderate significance;

» Water Corporation attached higher significance to engineeringrelated criteria suchas cost, site flexibility and operability and embedded (embodied) energy than thecommunity;

» The community attached the higher significance to community amenity, hazardouschemical risk and recreational and tourism values than Water Corporation.

These differences reflect the different perspectives of the two groups. It should benoted, however, that the spread of weights (0.5 to 0.13) is not great, indicating that


both groups generally considered all criteria to have some degree of importance, withthe most important criterion approximately two and a half times as important as theleast important criterion.

Table 6 Comparison of Average Normalised Weights

Criterion CommunityWeights

Water CorpWeights

1. Biodiversity 0.13 0.13

2. Potential for land degradation 0.11 0.11

3. Embedded energy of assets 0.05 0.10

4. Waterways and water cycles 0.10 0.11

5. Hazardous chemical risks 0.10 0.06

6. Community amenity 0.12 0.08

7. Recreational and tourism values 0.08 0.05

8. European heritage 0.07 0.06

9. Indigenous heritage 0.07 0.06

10. Site flexibility and operability 0.07 0.11

11. Costs 0.09 0.11

Total 1.00 1.00

(Note: The values in this table have been rounded to two decimal places for displaypurposes. In the actual calculations a higher precision was used.)

The community weighting data in Appendix K show that several participants appearedto have misunderstood the instructions, as they did not allocate the least importantcriterion (or criteria) a score of 1. However, following a careful examination of thesesets of weights, it was decided that the relative importance of each criterion was clearin all cases (specifically, no one had simply ordered the criteria for importance), so theywere included in the analysis and normalised along with all other weights.

The data also shows a distinct spread in the weights allocated by individuals. Thishighlights that averaging the weights of a group does not necessarily reflect thecommon views. For example if half the group allocated a criterion a weight in the rangeof 24, and the other half a range of 810 then the average may be calculated as 6,which would actually reflect the views of noone. To counter this, the analysis wasconduced using both average weights and individual weights, as discussed in moredetail in the next section.


7. Results of MCA

The scores and weights were combined using concordance analysis to compare theoverall performance of the four sites. Concordance analysis is explained in detail inAppendix A, but very simply, it involves comparing the criterion score for each optionwith the equivalent criterion score for another option and allocating the weight for thatcriterion to the option with the better score. This process is repeated till all options arecompared with all other options for each of the criteria, to ultimately produce a rankingof the options. In this report the ranking of the options has been produced from theperspective of both each individual involved in the weighting workshops and the twostakeholder groups, from the perspective of each individual and each stakeholdergroup, ie. the community (as represented by attendees at Community Forum 9) andWater Corporation. A worked example of concordance analysis is provided inAppendix L.

As indicated in Table 4, two or three different scores were recorded in the case of fourof the criteria: biodiversity, potential for land degradation, community amenity, andrecreation and tourism values. In the first three cases, this was because consensus onthe scores could not be reached, and in the latter case two different scenarios ofaccess to the Pine Plantation site were scored. This situation complicated the analysissignificantly, since it meant that there were 36 possible combinations of scores (3 x 2 x3 x 2). It was decided to analyse each of these possible combinations separately.

Table 7 presents the MCA ranking of the sites using the average community weightsapplied to each set of scores, where 1 is the best and six is the worst. Table 8 presentsthe results using the average Water Corporation weights.

In some cases several sites performed equally, and therefore have been allocatedequal ranks. The numbering system for the score scenarios refers back to the sets ofscores listed in Table 4. The asterisks (*) indicate how many of the scores for eachscenario were supported by only 1 person, as was the case for 1c biodiversity, 2bpotential for land degradation and 6c community amenity (refer to Table 4).


Table 7 MCA Ranking using Average Community Weights

Score scenarios(criteria 3, 4, 5, 8, 9,10, 11 fixed)

DECLand

PinePlantation

O’ConnorSite

Site 1 +DEC

Site 1 +Pine

Plantation

Site 1 +O’Connor

SiteC_1a2a6a7a 1 3 6 2 4 5C_1a2a6a7b 1 3 6 2 4 5C_1a2a6b7a 1 2 5 2 2 5C_1a2a6b7b 1 2 5 2 2 5C_1a2a6c7a * 1 2 6 2 4 5C_1a2a6c7b * 1 2 6 2 4 5C_1a2b6a7a * 2 3 6 1 3 5C_1a2b6a7b * 2 3 6 1 3 5C_1a2b6b7a * 1 4 5 2 2 5C_1a2b6b7b * 1 4 5 2 2 5C_1a2b6c7a ** 1 3 6 2 5 3C_1a2b6c7b ** 1 3 6 2 5 3C_1b2a6a7a 1 2 6 2 4 5C_1b2a6a7b 1 2 6 2 4 5C_1b2a6b7a 1 2 5 4 2 5C_1b2a6b7b 1 2 5 4 2 5C_1b2a6c7a * 1 2 6 3 3 5C_1b2a6c7b * 1 2 6 3 3 5C_1b2b6a7a * 2 3 6 1 3 3C_1b2b6a7b * 2 3 6 1 3 3C_1b2b6b7a * 1 3 5 3 2 5C_1b2b6b7b * 1 3 5 3 2 5C_1b2b6c7a ** 1 2 6 2 5 4C_1b2b6c7b ** 1 2 6 2 5 4C_1c2a6a7a * 2 3 6 1 4 5C_1c2a6a7b * 2 3 6 1 4 5C_1c2a6b7a * 1 2 5 2 2 5C_1c2a6b7b * 1 2 5 2 2 5C_1c2a6c7a ** 2 3 6 1 4 4C_1c2a6c7b ** 2 3 6 1 4 4C_1c2b6a7a ** 2 4 6 1 2 4C_1c2b6a7b ** 2 4 6 1 2 4C_1c2b6b7a ** 1 5 2 2 5 2C_1c2b6b7b ** 1 2 5 2 5 2C_1c2b6c7a *** 2 3 6 1 5 3C_1c2b6c7b *** 2 3 6 1 5 3


Table 8 MCA Ranking using Average Water Corporation Weights

Score scenarios(criteria 3, 4, 5, 8,9, 10, 11 fixed)

DECLand

PinePlantation

O’ConnorSite

Site 1 +DEC

Site 1 +Pine

Plantation

Site 1 +O’Connor

SiteW_1a2a6a7a 1 5 5 2 3 3W_1a2a6a7b 1 3 6 2 3 3W_1a2a6b7a 1 3 3 2 3 3W_1a2a6b7b 1 3 3 2 3 3W_1a2a6c7a * 1 5 5 2 3 3W_1a2a6c7b * 1 3 6 2 3 3W_1a2b6a7a * 2 5 6 1 3 4W_1a2b6a7b * 2 5 6 1 3 4W_1a2b6b7a * 1 4 4 2 4 3W_1a2b6b7b * 1 4 4 2 4 3W_1a2b6c7a ** 1 5 6 2 3 3W_1a2b6c7b ** 1 5 6 2 3 3W_1b2a6a7a 1 5 5 2 3 3W_1b2a6a7b 1 3 6 2 3 3W_1b2a6b7a 1 3 3 2 3 3W_1b2a6b7b 1 3 3 2 3 3W_1b2a6c7a * 1 5 5 2 3 3W_1b2a6c7b * 1 3 6 2 3 3W_1b2b6a7a * 2 5 6 1 3 4W_1b2b6a7b * 2 5 6 1 3 4W_1b2b6b7a * 1 4 4 2 4 3W_1b2b6b7b * 1 4 4 2 4 3W_1b2b6c7a ** 1 5 6 2 3 3W_1b2b6c7b ** 1 5 6 2 3 3W_1c2a6a7a * 2 4 6 1 3 4W_1c2a6a7b * 2 4 6 1 3 4W_1c2a6b7a * 1 4 4 2 3 4W_1c2a6b7b * 1 4 4 2 3 4W_1c2a6c7a ** 2 5 6 1 3 3W_1c2a6c7b ** 2 5 6 1 3 3W_1c2b6a7a ** 2 5 5 1 4 3W_1c2b6a7b ** 2 4 6 1 4 3W_1c2b6b7a ** 2 6 5 1 3 3W_1c2b6b7b ** 2 5 6 1 3 3W_1c2b6c7a *** 2 5 6 1 3 3W_1c2b6c7b *** 2 5 6 1 3 3


The following conclusions can be drawn from analysis and comparison of Tables 7 and8:

» DEC land is ranked as the best option in 24 out of 36 cases in Table 7 (Community)and 22 out of 36 cases in Table 8 (Water Corporation), with Site 1 + DEC Landranked as second best or equal second best in 18 cases (Community) and 22cases (Water Corporation).

» In all remaining cases, where DEC land didn't rank as the best option, Site 1 + DECland ranked as the best option. Even then, DEC land ranked as the second best;

» If score scenarios in which one or more score was supported by only one person(i.e. those marked with at least one asterisk) are disregarded, the DEC land ranksbest in all cases;

» The only score scenarios in which the Community and the Water Corporationweight generated a different best option were 1c2b6b7a and 1c2b6b7b, where thecommunity analysis preferred DEC land and the Water Corporation analysispreferred Site 1 + DEC land. This result was probably due to the fact that this scorescenario ranked Site 1 + DEC land poorly for community amenity, which wasweighted highly by the community in comparison with Water Corporation;

» O’Connor Site ranked as either 5 or 6 except under one score scenario usingcommunity weights and generally low using Water Corporation weights, and can beconsidered the least preferred option. This reflects its poor performance against thecriteria of biodiversity and potential for land degradation, which were weightedhighly by both the Community and Water Corporation. Site 1 + O’Connor Siteperformed slightly better, but was still generally unfavoured by both the Communityand Water Corporation;

» The Pine Plantation site ranked second under 15 score scenarios using theCommunity weights but did not rank above third using the Water Corporationweights. This reflects its relatively good performance against the criteria ofcommunity amenity and recreational and tourism values, which were weightedrelatively higher by the Community than the Water Corporation;

» The option of Site 1 + Pine Plantation was always ranked 3 or 4 by WaterCorporation, whereas the Community results showed greater variation, ranking it 2on twelve occasions, but 5 on eight occasions;

» The distinction between the two possible access routes to the Pine Plantation,which affected the scoring for recreational and tourism values (7a and 7b) made nodifference to results using Community weights, and a minor difference in 6 casesout of 18 possibilities cases using the Water Corporation weights;

» With respect to the biodiversity, scoring Site 1 + DEC land higher than DEC landalone (score 1c, supported by only one person), resulted in Site 1 + DEC beingranked higher overall than DEC land alone in most, though not all cases. Mostscore scenarios where DEC land was scored higher than Site 1 + DEC land (1a and1b) resulted in an overall preference for DEC land. This result also reflects thestrong influence over the final results of the biodiversity scores;


» Analysis of the three different sets of scores for community amenity (6a, 6b and 6c)suggests that even when DEC land and Site 1 + DEC land scored poorly forcommunity amenity, they still performed strongly overall. Under score scenariosincluding 6c, which ranked DEC land as 6 (worst), DEC land still ranked 1 undermost scenarios. Although 6a scored DEC land as 5 (second worst) DEC land stillranked 1 overall under most score scenarios, and 2 in all other cases. Similarly 6b,which scored Site 1 + DEC land as 6 (worst) resulted in an overall ranking of 1 forDEC land under all but two scenarios with Site 1 + DEC land ranked 2. The scorepreferred by most participants (6a), which scored Site 1 + DEC land higher thanDEC land alone, resulted in Site 1 + DEC being ranked 1 overall with DEC land 2;

» The different sets of scores for potential for land degradation (2a and 2b) made littledifference in the final rankings.

As discussed in Section 6, it is not always appropriate to use average weights as thiscan mask the views of different members of a group. Therefore, analysis was alsoconducted using each individual set of weights applied to each of the 36 sets of scores.This is a form of sensitivity analysis because a comparison of the different resultsshows the degree to which changes to either the scores for a particular criterion or theweights used would affect the final ranking of the sites. This approach provides WaterCorporation with a very comprehensive set of data in which the views of all thestakeholders involved in the MCA process are transparently represented.

The results of the individual analysis are summarised in Tables 9 (Community) and 10(Water Corporation). These tables show the number of individuals in each case whoseweights resulted in a particular site being ranked as the best. Options ranked as bestusing the weights of most individuals are presented in bold. Note that in cases wherean individual’s weights resulted in two options being ranked 1 then the total number fora score scenario will be greater than the number of individuals participating (i.e.individuals are counted twice if two option 1s are generated).


Table 9 Summary of MCA Results using Individual Community Weights


DEC Land PinePlantation

O’ConnorSite

Site 1 +DEC

Site 1 +Pine

Plantation

Site 1 +O’Connor

SiteC_1a2a6a7a 25 5 2 5 0 1C_1a2a6a7b 25 7 1 5 0 1C_1a2a6b7a 27 3 10 0 0 1C_1a2a6b7b 28 5 5 0 0 1C_1a2a6c7a * 27 6 3 0 2 1C_1a2a6c7b * 28 8 2 0 2 1C_1a2b6a7a * 15 4 1 19 0 1C_1a2b6a7b * 15 5 1 19 0 1C_1a2b6b7a * 27 3 10 0 0 1C_1a2b6b7b * 28 5 5 0 0 1C_1a2b6c7a ** 16 6 3 11 2 3C_1a2b6c7b ** 17 8 2 12 2 3C_1b2a6a7a 23 8 2 5 0 1C_1b2a6a7b 22 10 1 5 0 1C_1b2a6b7a 26 4 8 0 0 1C_1b2a6b7b 26 7 5 0 0 1C_1b2a6c7a * 25 10 3 0 0 1C_1b2a6c7b * 25 12 1 0 0 1C_1b2b6a7a * 15 8 1 15 0 1C_1b2b6a7b * 14 10 1 14 0 1C_1b2b6b7a * 26 4 8 0 0 1C_1b2b6b7b * 26 7 5 0 0 1C_1b2b6c7a ** 15 10 3 9 0 3C_1b2b6c7b ** 15 12 1 9 0 3C_1c2a6a7a * 14 4 1 22 0 1C_1c2a6a7b * 14 5 1 21 0 1C_1c2a6b7a * 27 3 10 1 0 1C_1c2a6b7b * 28 5 5 1 0 1C_1c2a6c7a ** 15 6 3 13 2 3C_1c2a6c7b ** 15 8 2 13 2 3C_1c2b6a7a ** 3 4 1 30 0 1C_1c2b6a7b ** 3 4 1 30 0 1C_1c2b6b7a ** 13 3 10 12 2 1C_1c2b6b7b ** 14 5 5 12 2 1C_1c2b6c7a *** 2 7 3 25 2 6C_1c2b6c7b *** 3 7 1 26 2 5


Table 10 Summary of MCA Results using Individual Water CorporationWeights


DECLand

PinePlantation

O’ConnorSite

Site 1 +DEC

Site 1 + PinePlantation

Site 1 +O’Connor

SiteW_1a2a6a7a 8 0 0 1 0 1W_1a2a6a7b 8 0 0 1 0 1W_1a2a6b7a 10 0 0 0 0 0W_1a2a6b7b 10 0 0 0 0 0W_1a2a6c7a * 8 0 0 0 0 2W_1a2a6c7b * 8 0 0 0 0 2W_1a2b6a7a * 5 0 0 5 0 1W_1a2b6a7b * 5 0 0 5 0 1W_1a2b6b7a * 10 0 0 0 0 0W_1a2b6b7b * 10 0 0 0 0 0W_1a2b6c7a ** 7 0 0 2 0 2W_1a2b6c7b ** 7 0 0 2 0 2W_1b2a6a7a 8 0 0 1 0 1W_1b2a6a7b 8 0 0 1 0 1W_1b2a6b7a 10 0 0 0 0 0W_1b2a6b7b 10 0 0 0 0 0W_1b2a6c7a * 8 0 0 0 0 2W_1b2a6c7b * 8 0 0 0 0 2W_1b2b6a7a * 5 0 0 5 0 1W_1b2b6a7b * 5 0 0 5 0 1W_1b2b6b7a * 10 0 0 0 0 0W_1b2b6b7b * 10 0 0 0 0 0W_1b2b6c7a ** 7 0 0 2 0 2W_1b2b6c7b ** 7 0 0 2 0 2W_1c2a6a7a * 4 0 0 6 0 1W_1c2a6a7b * 4 0 0 6 0 1W_1c2a6b7a * 10 0 0 0 0 0W_1c2a6b7b * 10 0 0 0 0 0W_1c2a6c7a ** 4 0 0 5 0 2W_1c2a6c7b ** 4 0 0 5 0 2W_1c2b6a7a ** 0 0 0 9 0 1W_1c2b6a7b ** 0 0 0 9 0 1W_1c2b6b7a ** 2 0 0 7 0 1W_1c2b6b7b ** 2 0 0 7 0 1W_1c2b6c7a *** 0 0 0 8 0 2W_1c2b6c7b *** 0 0 0 8 0 2


The data in Tables 9 and 10 is consistent with the Tables 7 and 8, recording a strongoverall performance for DEC Land and Site 1 + DEC land relative to other options:

» DEC land and Site 1 + DEC land are the only two options that are ranked bestunder any scoring scenarios, using any individual’s weights;

» DEC land was ranked best using the weights provided by individuals from theCommunity under 26 scoring scenarios, and equal best under another 2 scoringscenarios. Site 1 + DEC land ranked best in all other cases;

» DEC land was ranked best using the weights of individuals from the Water Corpunder 22 scoring scenarios, and performed equal best under another four scoringscenarios. Site 1 + DEC land ranked best in all other cases;

» The Pine Plantation and O’Connor Site were not ranked as the best option underany scoring scenarios using any Water Corporation individual’s weight, butperformed more strongly using individual Community weights. The Pine Plantationperformed second best under 16 scoring scenarios (i.e. was ranked best using thesecond greatest number of individual weights);

» The analysis using individual weights also showed more support for the O’ConnorSite than was apparent using the averaged weights, demonstrating how averagingweights can mask different views within a group. However, it is important to notethat in this case the individual analysis only influenced the third and fourth rankingoptions, and that DEC land and Site 1 + DEC land were still clearly found to be thebest options;

» If score scenarios in which one or more score was supported by only one person(i.e. those marked with at least one asterisk) are disregarded, the DEC land wasranked as the best option under all scenarios.

The MCA process therefore strongly suggests that the DEC land is the best option forthe location of the Mundaring WTP, followed by the combination of Site 1 + DEC land.The Pine Plantation can be considered the third best option from the Communityperspective, although it performed poorly using Water Corporation weights. O’ConnorSite, Site 1 + Pine Plantation and Site 1 + O’Connor Site performed relatively poorlyoverall.

To provide Water Corporation with the best possible information, it is important that thereasons for the relative performance of the options is analysed and reviewed. Thestrengths and weaknesses of the three best performing options are brieflysummarised, with reference to the scores provided in Table 4 as follows:

» DEC land performs well overall, being in the top three preferred options for allcriteria except cost (ranked 4) and community amenity (ranked 5 by 7 communitymembers, 3 by three community members and 6 by one community member).Although concerns have been expressed about the potential impacts of a WTP onDEC land with respect to recreational and tourism values and European heritage, itwas still considered to perform better than all configurations of Site 1 with respect tothese criteria;


» Site 1 + DEC land is a less costly option than DEC land alone (ranked 2) and alsoperforms slightly better with respect to embedded (embodied) energy andindigenous heritage. However, it performs poorly against recreation and tourismvalues and European heritage (ranked equal 4 with other Site 1 configurations). Itsperformance against community amenity is mixed (ranked 3 by 7 communitymembers, 6 by three community members and 4 by one community member). Inconclusion, there appears to be no strong reason why Site 1 + DEC land should beselected over DEC land alone.

» The Pine Plantation performs well with respect to hazardous chemical risks (ranked1), community amenity (ranked 1 or 2), recreation and tourism values (ranked 1 or2) and relatively well for both European and indigenous heritage (ranked 2).However, it is the worst performer in terms of cost, site flexibility and operability,and waterways and water cycles and performs poorly for embedded (embodied)energy (ranked 5) and potential for land degradation (ranked 4 by six members ofspecialist group and 5 by one member). In summary, almost without exception, thisoption performs well against criteria weighted highly by the Community and poorlyagainst criteria weighted highly by the Water Corporation, which explains thedifference in overall performance between the Community and Water Corporationanalysis. However, despite this it does not perform as well as DEC land or Site 1 +DEC land from either the Community or the Water Corporation perspective.


8. Conclusions

The MCA process identifies that the DEC land ranks highest as the preferred site forthe proposed Mundaring WTP, ahead of a combination of Site 1 + DEC land. Thisconclusion was reached based on analysis using average Community and WaterCorporation weights (Tables 7 and 8) and using individual Community and WaterCorporation weights (Tables 9 and 10).

The analysis was conducted for 36 distinct combinations of scores. These variationsarose because consensus was not reached on the scores for three criteria, and thattwo different scenarios of access route were scored. This is a form of sensitivityanalysis, since it allows the influence of different scores to be evaluated. Theconclusion that can be drawn is that variations in scores for these criteria made littledifference to the overall results in that DEC land and Site 1 + DEC land still performedbest overall, no matter how they were scored with respect to particular criteria bydifferent groups.

Tables 7 and 8, which were generated using average (not individual) Community andWater Corporation weights, show that DEC land is ranked 1 (best) under the vastmajority of score scenarios, and that Site 1 + DEC land is ranked best in all othercases. Where DEC land was ranked 1, Site 1 + DEC land was most commonly ranked2, and vice versa. The Pine Plantation was ranked equal 2 or 3 more often than anyother option from the Community perspective (Table 7) but performed relatively poorlyfrom a Water Corporation perspective (Table 8). The O’Connor site was ranked 6 moreoften than any other option using both the Community and Water Corporation averageweights, and can therefore be considered the least preferred option.

Tables 9 and 10 confirm these general trends using individual weights. The MCAprocess showed that DEC land was ranked 1 using individual weights by the mostpeople under most score scenarios, and that Site 1 + DEC land was ranked 1 in allother cases. A comparison between DEC land and Site 1 + DEC land indicates that thelatter performs better with respect to cost, but that there would otherwise be littlereason to select the second best performing option over the best performing.

Tables 9 and 10 were generated by applying individual weights to each of the 36 scorescenarios and then recording the number of times a particular site was ranked 1 (best).This analysis is important because averaging weights can mask distinctly differentviews within the groups. In this analysis, although DEC land and Site 1 + DEC landwere still by far the best performing sites, the Pine Plantation and O’Connor Sitepreformed better than was evident from the analysis using averaged weights. While thePine Plantation was not ranked 1 using any Water Corporation individual’s weightsunder any scenario, it performed relatively well using individual Community weights,and could therefore be considered the third best option, after DEC land and Site 1 +DEC land. Further analysis suggests that this is because it generally performed wellagainst those criteria that were weighted higher by the Community (communityamenity, recreational and tourism values and hazardous chemical risks) and poorlyagainst criteria that were weighted higher by the Water Corporation (cost, site flexibility


and operability, and waterways and water cycles). However, it is important to note thatdespite this, it was still only the third best performing site even using individualCommunity weights.

It should be noted that while the MCA process provides a detailed picture of therespective sustainability implications of the different site options, it does not indicatewhether the residual impacts are acceptable or not. Some sections of the communityhave expressed concerns about the potential impacts on recreation and tourism valuesand community amenity associated with the best performing option of DEC land, asreflected in the scores in Table 4 (not withstanding that the same concerns also applyto other sites). In the process of selecting its preferred site, Water Corporation mustevaluate and consider the significance of the impact of a WTP on this site with respectto these criteria.


9. References

Australian Interaction Consultants (2007) Report of an Archaeological Survey forAboriginal Sites in the area of the Proposed Water Treatment Plant and AssociatedInfrastructure at Mundaring, Western Australia, Perth, Western Australia

GHD (2007a) Report for Mundaring Water Treatment Plant Site Selection: PreliminaryComparative Environmental Impact Assessment, Perth, Western Australia

GHD (2007b) –Technical Note: Mundaring Water Treatment Plant Chlorine Storageand Dosing – Water Treatment Plant Location Sensitivities, Perth, Western Australia

Heritage and Conservation Professionals (2007) Mundaring Weir Water TreatmentPlant: Preliminary European Heritage Survey Heritage Audit for Alternate Sites, PerthWestern Australia


Appendix A

Making Good Decisions

MAKING GOOD DECISIONS

MAKING GOOD DECISIONS: A GUIDE TO USING DECISION-AIDING TECHNIQUES IN WASTE FACILITY SITING David Annandale and Ross Lantzke Institute for Environmental Science Murdoch University, Perth, Australia

2000

Supported by a grant from the Western Australian Waste Management and Recycling Fund

A GUIDE TO USING DECISION-AID ING TECHNIQUES IN WASTE FACIL ITY SIT ING


Table of Contents

1. INTRODUCTION..................................................................................................1

2. DECISION-AIDING TECHNIQUES AND IMPACT EVALUATION ..................2

2.1 Introduction...................................................................................................2

2.2 A Basic Decision-aiding Model .....................................................................3

2.3 Applications of Decision-aiding Techniques..................................................6

2.4 Strengths and Weaknesses of Decision-aiding Techniques.............................8

3. CONSTRUCTING A DECISION-AIDING EXERCISE ......................................11

3.1 Steps In Decision-aiding..............................................................................11

3.2 The Example Problem .................................................................................11

3.3 Specifying the Alternatives..........................................................................14

3.4 Specifying the Comparison Criteria .............................................................15

3.5 Scoring the Alternatives in Relation to Criteria ............................................18

3.6 Assigning Weights.......................................................................................22

3.7 Undertaking the Computation ......................................................................23

3.7.1 Additive Weighting ..........................................................................24

3.7.2 Concordance Analysis ......................................................................27

3.8 Dealing with Uncertainty.............................................................................31

3.9 Presentation of Results.................................................................................32

4. CONCLUSION ....................................................................................................32

5. REFERENCES.....................................................................................................34


1


1. INTRODUCTION

To many of us, making decisions is just 'common sense'. When we need to

decide about what kind of breakfast cereal to buy in a shop, or what University to

attend, or what kind of car to buy, we go through a similar decision-making

process each time, although this will vary depending on the psychology of the

individual in question.

Commonly, we would think about what the 'options', or 'alternatives' are. For

example, there might be 15 different types of breakfast cereal to choose from in

the shop, or four possible Universities to enroll in.

At the same time, we would also be thinking about how to compare the

alternatives available to us. If we are attempting to buy a car, we might have a

certain price, or size, or colour in mind. There will always be a certain number of

comparison criteria that we will use in attempting to make the 'best' decision.

Sometimes the decisions we make in our everyday lives are fairly straightforward.

We might only have a small number of alternatives to choose from, and perhaps a

handful of comparison criteria to help us make the choice.

This is rarely the case, however, for waste management planning decisions. If a

Government is trying to find a waste management solution for a significant

suburban community, there might be 20 to 30 combinations of site and

technology, and perhaps the same number of criteria by which to compare

alternatives. If you imagine this decision in the form of a matrix, with 20 to 30

alternatives along the X axis and the same number of comparison criteria along

the Y axis, you can begin to see how it might be difficult for an ordinary human

decision-maker to sort through, and order, the choices.

When decisions become this complex, we need special tools or techniques to help

us make sense of what can be a large amount of information. In addition,

complex environmental planning problems will almost always include value


2


judgements, public opinion, and controversy. So, the techniques need to deal

with more than just technical information. As was expressed by the now-defunct

Australian Resource Assessment Commission:

“The more complex the problem and the greater the number of players

involved, the greater the need to structure the decision-making process in a

systematic way, both to improve the quality of the decision and to justify

any action taken” (Resource Assessment Commission 1992).

Many different techniques have been developed over the years to assist in

situations such as these, and a number will be introduced later in this manual.

The main purpose of this manual is to introduce waste management decision-

makers to how decision-aiding techniques work, and how they can be used in the

evaluation stage of environmental impact assessment.

The manual has three parts. The first part is a brief overview of the role that

decision-aiding techniques can play in impact evaluation. The second part

outlines the steps taken when decision-aiding techniques are used to help guide

decision-making. It also provides details about specific decision-aiding

techniques. The final part of the manual includes a conclusion, and a list of

references.1

2. DECISION-AIDING TECHNIQUES AND IMPACT EVALUATION

2.1 Introduction

The term "decision-aiding techniques" is not precisely defined and is sometimes

synonymously termed, ‘multi-objective decision support’, ‘multi-criteria

decision-making’, or 'multi-criteria analysis'. All of these definitions have a basic

set of common components being:


3


• a given set of alternatives;

• a set of criteria for comparing the alternatives; and

• a method for ranking the alternatives based on how well they satisfy the criteria.

As is implied by the previous section, all of the techniques are not necessarily

new. For example, cost-benefit analysis, which was the predominant method for

project evaluation in the post-war period, is a type of decision-aiding technique

where the comparison criteria are all measured in dollars. However, there has

been a significant expansion in the use of decision-aiding techniques in the last 15

years, much of which can be attributed to methodological and ethical concerns

around whether it is right to convert all comparison criteria to dollars.

One of the perceived advantages of decision-aiding techniques is that they can

represent comparison criteria in their original units, without having to convert

them to a 'standard' commensurable unit such as dollars. The basic model

presented in the next section makes this point clearer.

2.2 A Basic Decision-aiding Model

Perhaps the best way to introduce decision-aiding techniques is to work through a

basic model. Consider a relatively common waste management policy choice

faced by many local governments. There could be at least three alternative

approaches to dealing with solid waste. One might be traditional landfill, another

might be incineration, and another might be composting combined with

landfilling of residual waste.

Criteria used to compare these alternatives might include capital cost,

employment potential, area of land required, and the possibility of groundwater

pollution from each option. The best alternative is not necessarily obvious.

Incineration would likely result in the lowest risk of groundwater pollution, but

1 Much of this manual is based on a structure first presented in the Resource Assessment Commission's 1992 booklet titled, "Multi-criteria Analysis as a Resource Assessment Tool"


4


employs the fewest people and costs the most. Composting combined with

landfill might be the cheapest option and employ the most people, but it requires a

reasonable amount of land and still has a groundwater pollution risk.

Clearly, this is not a simple decision. It would be even more complicated if there

were more options to choose from, more criteria used to make the comparison,

and stakeholder conflict over the relative importance of the different comparison

criteria. Decision-aiding techniques can help in situations such as this by

comparing the advantages and disadvantages of each alternative, one against the

other.

Table 1 shows how the alternatives and comparison criteria can be easily

displayed. This kind of table is usually referred to as an 'effects table', or as an

'impact matrix'. The columns of the effects table represent the alternative ways of

dealing with solid waste, and the rows represent the criteria by which the

alternatives are to be evaluated. The entries in the cells of the matrix are usually

known as 'scores'.

The first three criteria can be relatively easily measured in quantitative terms,

perhaps by consultants. The last criterion could be determined by "expert

judgement" or by panels of community members. Decision-aiding techniques

allow for this type of input.

Table 1: Effects table for a choice of hypothetical landfill options.

OPTIONS

CRITERIA conventional landfill incineration composting &

residual landfill

1. Capital cost ($ million) 20 30 10

2. Employment (thousands) 20 10 50

3. Area of land required (ha) 100 10 30

(Resource Assessment Commission 1992). This reference is greatly acknowledged.


5


4. Possibility of groundwater pollution. high very low low

A significant advantage of most multi-criteria methods is that they can allow for

evaluation criteria to be measured in either quantitative terms (i.e. with numerical

values) or qualitative terms (i.e. rank ordering or 'pluses and minuses'). The

ability of some multi-criteria methods to deal with qualitative scores or a mixture

of quantitative and qualitative scores provides greater flexibility compared with

techniques such as cost-benefit analysis that require all values to be quantified.

Sometimes merely assembling an effects table and determining the scores for

each matrix cell helps to clarify the decision-making process and might be enough

to produce a clear ranking of the choices.

When this is not the case, usually because there are many possible choices and/or

comparison criteria, the next step is to apply to the effects table a mathematical

procedure that results in a ranking of the alternatives. Some of these procedures

are introduced in the next two sections.

Before the mathematical procedures can be applied, however, the people involved

in the decision-making process need to attach some measure of relative

importance to the comparison criteria. Clearly, different people will 'weight'

comparison criteria differently, depending in part on their individual interests.

For example, a local authority bureaucrat might be more concerned about the cost

of an alternative, than about whether it generates employment. A person living

near a proposed landfill might be very concerned about groundwater pollution,

and not as interested in the cost of an alternative. A decision-aiding technique

should therefore allow for different 'weights' to be placed on comparison criteria,

so as to allow for different points of view to be explored. Weights can be

expressed in a number of ways, although the most common is allow participants

to apply a value as a number on a scale (perhaps from '0' to '10').


6


Some suggestions as to how weighting processes might be conducted will be

outlined later in this manual.

2.3 Applications of Decision-aiding Techniques

Decision-aiding techniques originated from business planning and project

development. These disciplines relied strongly on cost-benefit analysis in

industrialized economies during the post-World War II expansionary times, but

also ran into criticism during this period because of perceived difficulties

associated with placing a monetary value on environmental impacts. Cost-benefit

analysis was also criticized for its treatment of equity issues.

Around the same time that cost-benefit analysis was being challenged as the

preeminent technique for options choice, disciplines other than business planning

and project development - for example, urban planning and natural resource

management/EIA - began to look for methods to assist with difficult decisions.

Urban planners needed techniques which would allow them to make complex

comparisons between alternative development proposals. Two of the best known

"competitors" in this field are the Planning Balance Sheet (PBS) method (a

British development) and the Goals Achievement Matrix (GAM) approach, which

has also been heavily used in Britain and the USA.

PBS was developed by Nathanial Lichfield in Britain and extensively used by the

planning profession in the 1960's and 1970's. It is an adaptation of cost-benefit

analysis and shares its basic theory and methods. PBS goes beyond traditional

cost-benefit analysis in two ways. First, it records detailed information on the

distribution of costs and benefits among different groups of people affected by a

proposal (differentiating them as either producers or consumers). Second, it

formally accommodates intangibles and unmonetized impacts by providing them

with symbols for recording them in evaluation tables alongside monetized

impacts.


7


Goals Achievement Matrix (GAM) was developed in the late 1960's in reaction to

the perceived deficiencies of CBA and PBS. It gained some currency in Britain

during the early 1970's where it was used in a number of major regional structure

plans.

The main difference between GAM and PBS is that, in GAM, impacts are

categorized according to a set of explicitly stated community "goals" (such as

clean air or quiet surroundings), and further subdivided according to the

community groups affected. Each community goal, and each affected group, is

assigned a weight by a panel of experts. In addition, quantification is stressed,

but not monetization.

Examples of the workings of PBS and GAM are provided in McAllister (1980)

and Westman (1985).

The final decision-aiding technique derived from the urban planning discipline to

be discussed here is the McHarg Overlay method, which produces a graphical

output based upon the concept of spatial, physical constraints. McHarg's

constraint mapping "sieving" approach has been easily adapted and improved by

computerised GIS systems. In essence the overlay method produces a separate

"sheet" for each for each impact being assessed. Colours are used to represent

significance through intensity of shading. The advent of computerised GIS

systems has allowed the overlay method to include many more impacts than were

possible using a manual approach. The computerised approach to overlay has

been often used in the selection of "corridor" routes for facilities such as

powerlines and transport corridors.

The main advantage of overlay methods is their ability to show the spatial

dimension of impacts. They are also capable of including a weighting system,

indicating which variables are considered to be more important than others. The

disadvantages, however, include the fact that they are not successful in dealing

with impact characteristics such as probability, time, and reversibility. In


8


addition, overlay maps tend to make strict demarcations between boundaries. In

natural systems boundaries between vegetation and soil types, for example, may

sometimes be indistinct.

As with urban planning, the natural resource management "discipline" has

generated a wide range of decision-aiding techniques for use in areas such as

water resource planning, land use planning, timber harvesting, and the valuation

of significance associated with the environmental impact assessment of projects.

Perhaps the best known of these techniques are the "matrix" methods, such as the

Leopold Matrix approach, and the so-called Environmental Evaluation System

(EES). Most of the important decision-aiding techniques designed specifically

for use in natural resource management are outlined in Westman (1985).

Some of the earliest uses of decision-aiding techniques in natural resource

management have been in water resource planning in the USA, and transport

planning in Europe. The concepts have been relatively slow to be taken up in

Australia, but have gained ground in recent years as a result of promotion by the

Resource Assessment Commission through its Forests and Timber Inquiry, and its

use in various Queensland transport planning projects. It has also been used

sporadically for transport planning in WA, and is currently being applied to

recreational land use planning by local authorities in the Perth hills area.

2.4 Strengths and Weaknesses of Decision-aiding Techniques

It is sometimes possible to mistake decision-aiding techniques for "decision-

making". It needs to be stressed that decision-aiding techniques do not "make"

decisions. No technique can eliminate the need for governments to make difficult

decisions that must weigh the contending views and values of various segments of

the population. Instead, decision-aiding techniques help those involved in

decision-making to more easily compare options and make choices by making

problems more transparent, and by rationally ordering the decision-making

process.


9


While this Introductory Guide implicitly supports the use of such techniques,

there are weaknesses as well as strengths associated with their use. Table 2 lists

some of the advantages and disadvantages of decision-aiding techniques.


10


Table 2: Strengths and weaknesses of decision-aiding techniques

STRENGTHS

• provides structure for decision making while still allowing flexibility;

• particularly useful for complex problems where the amount of information exceeds the integrative capacity of the human brain;

• follows naturally from the way people tend to approach problems with multiple objectives—A familiar analogy of the effects table is a table comparing specifications for several models of motor car;

• flexible data requirements—methods are available for qualitative data, quantitative data, or a mixture of both;

• allows different points of view to be dealt with explicitly through the use of weights;

• allows information that is agreed upon by all parties to be distinguished from areas of contention (indicated by different weights);

• amenable to sensitivity analysis to determine how robust the final results are to changes in the underlying assumptions and methods;

• does not require assignment of a monetary value to all quantities;

• can identify where additional data would be useful and where additional data would have little impact on the final decision.

WEAKNESSES

• does not overcome fundamental problems associated with comparing quantities that some would argue are not comparable, but does provide more flexibility than is available with, say, benefit-cost analysis;

• variety of evaluation methods available without any clear indication that one is better than another;

• since many of the methods are complex and remain a ‘black box’ to the decision maker they can lead to either mistrust or excessive faith in the results;

• concentration on the definition of explicit weights can provide a false sense of objectivity about the remainder of the analysis—there are opportunities for introducing implicit weights at all stages of the analysis and these may remain undetected;

• considerable effort is needed to obtain the information for the effects table and the weights;

• methods for incorporating uncertainty explicitly into the analysis are not yet well


11


developed.

Source: Resource Assessment Commission 1992

3. CONSTRUCTING A DECISION-AIDING EXERCISE

3.1 Steps In Decision-aiding

Although there are a great number of decision-aiding techniques, most follow a

consistent application process. To make decision-aiding work as a rational

evaluation tool, the following issues need to be addressed:

• specifying alternatives;

• specifying criteria;

• scoring alternatives;

• assigning weights to criteria;

• undertaking the computation;

• dealing with uncertainty; and

• presenting results.

• The rest of this section will introduce what happens in each of these steps, using a transport planning example.

3.2 The Example Problem

In the early 1990's the Western Australian Government decided that it wanted to

improve the provision of public transport between the capital city of Perth

(population 1.5 million) and the growing coastal regional centre of Mandurah

(population 50,000) which is approximately 80kms to the south.

The Government had no preferred public transport technology in mind, and was

aware of the fact that finding a route for a bus, tram, or train would be

controversial. Through its Department of Transport, the Government

commissioned an environmental planning consulting company to help it find the


12


best combination of public transport technology and route. The project was given

the name 'South West Area Transit" (SWAT).

As Figure 1 indicates, SWAT was a complex study. It involved the aggregation

of outcomes from four separate investigations into economics (Brief 8),

environmental impacts (Brief 9), social impacts (Brief 10), and impacts on urban

development (Brief 11). These were brought together in a consolidated decision-

aiding exercise for Brief 12.

Figu

re 1

: Out

line

of th

e So

uth

Wes

t Are

a Tr

ansit

Stu

dy (S

WA

T)

Note

: Br

ief 8

: Fi

nanc

ial a

nd E

cono

mic

Eva

luat

ion

Br

ief 9

: En

viro

nmen

tal I

mpa

ct A

sses

smen

t

Brie

f 10:

So

cial

Impa

ct A

sses

smen

t

Brie

f 11:

Im

pact

on

Urb

an D

evel

opm

ent

Note

: Wei

ghtin

gs a

nd

rank

ings

in B

rief

s #8

to 1

1 to

be

igno

red

for

Brie

f #12

Raw

scor

es o

n Br

ief #

8 iss

ues

Raw

scor

es o

n Br

ief #

9 iss

ues

Raw

scor

es o

n Br

ief #

10 is

sues

Ra

w sc

ores

on

Brie

f #11

issu

es

Issu

es w

eigh

tings

from

focu

s gro

ups

RAN

K O

PTIO

NS

BASE

D O

N T

HE

WEI

GH

TIN

G O

F IS

SUES

BY

TH

E FO

CUS

GRO

UPS

PRES

ENT

FIN

DIN

GS

TO U

RBA

N T

RAN

SPO

RT C

OM

MIT

TEE

TO

ASS

IST

THE

COM

MIT

TEE

TO F

ORM

AN

OPI

NIO

N O

N T

HE

RELA

TIV

E M

ERIT

S O

F TH

E O

PTIO

NS

– IN

TH

E A

BSEN

CE O

F PR

IVA

TE S

ECTO

R SY

STEM

PRO

VIS

ION

PRO

POSA

LS

CARR

Y O

UT

SEN

SITI

VIT

Y T

ESTS

BY

VA

RYIN

G W

EIG

HTI

NG

S O

BSER

VE

ROBU

STN

ESS

OF

RESU

LTA

NT

OPT

ION

RA

NK

ING

13


14


The consolidation exercise consisted of 32 comparison criteria, with 18

route/technology combinations to choose from. The rest of the example will

introduce you to how decision-aiding techniques were used in Brief 9

(environmental impacts).

3.3 Specifying the Alternatives

In a sense, selection of alternatives can be thought of as being not particularly

controversial. However, agreement on alternatives can be a crucial aspect of

decision-aiding techniques, especially if interested stakeholders exist. This is

because interest groups can quickly discount the legitimacy of a multi-criteria

exercise if they believe that viable options have been excluded.

In the SWAT study alternative public transport technology/route combinations

were determined by the consultants in collaboration with SWAT bureaucrats.

Despite what was said in the previous paragraph, this did not result in conflict,

probably because so many options (18) were presented and so interest groups

were satisfied that their own choices were represented.

Experience has suggested, however, that the best way to determine alternatives

for a decision-aiding exercise is to involve stakeholders and allow them to offer

as many alternatives as they see fit. This is sometimes a threatening concept for

decision-makers, but it is the first step in gaining legitimacy for the process.

The 18 options generated for the SWAT study included heavy rail, light rail and

busways. All of these technologies had different combinations of route

alternatives.

This was obviously quite a long list of alternatives. Often multi-criteria exercises

attempt to reduce the range of alternatives to a manageable set by eliminating

those that do not satisfy an initial screening criterion. In the SWAT case, for

example, it might be that a particular wetland is so significant that it cannot be


15


impinged upon by any public transport route. To do so would be to eliminate the

alternative. No such screening criterion existed in this example.

3.4 Specifying the Comparison Criteria

Decision-aiding techniques are founded on the idea that decision-makers usually

attempt to satisfy more than one objective simultaneously. Often these objectives

can meet all of the so-called 'triple bottom line' aims of social, economic, and

environmental outcomes. You should bear in mind that the SWAT example

focuses only on environmental aims.

The label 'criterion' covers a continuum from relatively well defined and easily

measured quantities, such as area of land required for a train line, to less well

defined concepts such as visual amenity.

Some commentators suggest that the number of criteria used should be restricted

because of the cognitive limitations of the human mind, and because of the need

to gather the necessary information for each alternative. Others say that a forced

limiting of criteria increases potential for conflict, similar to presenting a limited

number of alternatives.

There are some pre-conditions for criteria that have developed over time. Keeney

and Raiffa (1976) summarize these as follows:

Complete. If two alternatives have the same score for each criterion then it must

be agreed that the two alternatives are equivalent. In other words, there should not

be any additional basis for distinguishing between alternatives. In the SWAT

case, the criteria would not be complete if one or more parties agreed that two

alternatives were identical with respect to the criteria listed but then argued that

visual amenity must also be considered.

Operational. The set of criteria should be able to be used in some meaningful

manner in the ensuing analysis.


16


Decomposable. It should be possible to simplify the analysis by disaggregating

the decision problem. Suppose the criteria include biological diversity and the

area of disturbed land, and an increase in biological diversity is considered

desirable when it occurs in natural areas but undesirable when it occurs in

disturbed areas because then it results from invasion of weedy species and pests.

The set of criteria is not decomposable because biological diversity cannot be

evaluated independently of the area of disturbed land. In the extreme case in

which the score for each criterion is dependent on the score for every other

criterion, the multi-criteria structure is lost completely.

Non-redundant. No aspect of the problem is accounted for more than once. This

is sometimes difficult, as people will have different ideas about the use of natural

resources. Take an example of conflict over the use of forest reserves. One party

might argue that an area of land placed in conservation reserve and opportunities

for recreation are closely linked and the use of two criteria results in this factor

being counted twice. Others might argue that conservation reserves have a

significance beyond their contribution to recreation and it is essential that there be

two separate criteria. Similar arguments can be applied to possible overlaps

between environmental quality, economic growth and quality of life. Decision-

aiding techniques cannot resolve these arguments, but it does provide a

framework within which the arguments can be articulated and examined.

Minimal. There should be no other smaller set of criteria satisfying the preceding

conditions.

In the SWAT case, 32 comparison criteria were developed by the consultants

acting in collaboration with SWAT bureaucrats. After the event, the consultants

realized that this was a mistake, as consultations over weighting (see below)

indicated that an important criterion had been left out. As a general rule,

therefore, it seems that stakeholders should be allowed to offer their own list of

criteria.


17


Figure 2 presents a small 'snaphot' outline of the effects table used for Brief 9 of

SWAT.


18


Figure 2: Selection of alternative and comparison criteria for SWAT

Criteria

energy use

air pollution

noise

vibration

lakes & wetlands

system six areas

vegetation

Alternatives

metro

rail

east

(A)

metro

rail

west

(B)

access

rail

east

(C)

access

rail

west

(D)

bus-

way

east

(F)

MRS

amend-

ment

(G)

bus-

way

west

(E)

South West Area Transit (1993)

3.5 Scoring the Alternatives in Relation to Criteria

To complete the effects table, some value or score must be assigned to each

alternative indicating its performance in relation to each criterion.

There are three types of measurement scales: ordinal, interval and ratio. An

ordinal scale provides information on order only. It can indicate that one

alternative scores higher than another alternative, but not by how much. An

interval scale provides a measure of the difference between two alternatives but

does not indicate actual magnitude. For example, on an interval scale the scores 3

and 6 are equivalent to the scores 23 and 26. The scores imply that one

alternative scores 3 units higher than another, but whether this represents a

doubling or just a tiny change cannot be determined. A ratio scale has a natural

origin (zero value) and provides a measure of both difference and magnitude. On


19


a ratio scale the scores 3 and 6 would indicate a doubling of the score, whereas

the scores 23 and 26 would indicate a relatively small increase of 13 per cent.

Scores obtained by directly measuring ‘real’ attributes will generally have a ratio

scale. Ordinal or interval scales are more common when scores are constructed by

other means.

Since measurements such as: +++, ++, +, or "low", "medium", "high" must be

translated to at least an ordinal scale if they are to be used in any useful way,

‘qualitative’ and ‘ordinal’ are often used interchangeably and ‘quantitative’ is

reserved for interval and ratio scales. In Figure 1, the first three criteria are

measured in ratio scale, and the last criterion is measured in an ordinal scale.

Capital cost, employment, and area of land required are regarded as being

quantitative, and "possibility of groundwater pollution" is seen as qualitative. This

nomenclature differs slightly from standard mathematical practice, where

‘quantitative’ is used to refer to any measurement expressed as a number and

therefore includes ordinal scales.

In the SWAT example, "vibration" was measured in an ordinal scale, and all other

criteria were measured in ratio scale.

The type of scale is important because it determines which mathematical

operations are meaningful and therefore the types of evaluation methods that can

be applied. No mathematical operations are valid with ordinally scaled data.

Interval scaled data can be added and subtracted, but not multiplied or divided.

All operations (addition, subtraction, division, and multiplication) can be applied

to ration-scaled data. As Vansnick (1990) states,

"Either you tackle the difficult problem of working with poor

information, or you tackle the difficult problem of getting additional

information in order to obtain an interval scale or a ratio scale and

consequently…[be able to] use a more sophisticated mathematical

treatment".


20


Before applying some evaluation methods, in particular those in which the

numerical value of the score is used directly to rank alternatives, it is necessary

for all criteria scores to be reduced to a comparable or standardised basis.

Standardisation should not be confused with weighting: Standardisation is

intended to eliminate effects of scale that would otherwise introduce a weighting.

The aim is to avoid introducing weights at any place in the analysis other than

explicitly through the weight vector.

A simple example of standardisation is ensuring that dollar amounts are all

expressed as actual dollars and not dollars for some criteria and thousands of

dollars for others. If the scores were not expressed on the same scale and the

method for ranking alternatives was to simply sum the scores over all criteria, the

criteria expressed in dollars would be weighted 1000 times more heavily than the

criteria expressed in thousands of dollars.

A less trivial example is deciding how to standardise energy use and area of lakes

and wetlands taken, as in Figure 2. Various standardisation methods have been

proposed, but there is no obvious reason for selecting one method over another.

Since the objective of the standardisation is to enable comparisons between

criteria originally measured on quite different scales (MJ/km versus hectares, for

example) the method that appears best suited to do this has to be determined on a

case-by-case basis. In some situations the choice of method may have little

impact on the final results. In other situations it could have a significant effect.

An example of a commonly used standardisation formula is provided Section 3.7,

which introduces two decision-aiding techniques.

A final note worthy of comment in relation to standardisation is that for the

formula to work properly it needs to be clear whether a criterion and its score is a

"cost", or a "benefit". With a "cost" criterion, the lower the score the better the

outcome. For example in the SWAT case, if we are concerned with lakes and

wetlands, the best alternative would be the one that affects the smallest area

(measured in hectares). This criterion would therefore be defined as a "cost".


21


When undertaking decision-aiding with environmental criteria, most criteria will

be "costs". This is because we will tend to compare alternatives according to how

they impact negatively on environmental aspects. An example of a "benefit"

might be "jobs created", or "hectares of land rehabilitated". Here, a higher score

is a better outcome.

Standardisation formulae treat "costs" and "benefits" differently. Section 3.7 will

better explain this concept.

Finally, standardisation is an issue only for evaluation methods that use the actual

numerical value of a criterion to obtain a ranking of the alternatives.

Scoring is one of the few areas in decision-aiding where value judgements are

not significant. As a consequence, it is usually possible for participants to 'hand

over' scoring to technical professionals. This is what happened in the SWAT

case, where the consultants were responsible for undertaking scoring. It needs to

be remembered, however, that not all criteria are easily measured in ratio-scaled

units. Some criteria will require the application of value judgement. For

example, it is difficult to imagine how the impact of a new rail line or busway on

visual amenity or individual well-being, could be 'measured' in an objective sense

by technical experts. With criteria such as these it is likely that scoring will need

to be done by decision-makers, or their consultants, or by stakeholders, in some

kind of consultative fashion.

For the criteria listed in Figure 2, the following units were used:

• energy use: MJ/km

• air pollution: kgCO2/km

• noise: decibels (db)

• vibration: ordinal scale from '0' to '2'

• lakes and wetlands: hectares affected (ha)

• system six areas: hectares affected (ha)


22


• vegetation: hectares affected (ha)

3.6 Assigning Weights

Assigning weights to the criteria is possibly the most valuable aspect of decision-

aiding because it allows different views and their impact on the ranking of

alternatives to be expressed explicitly. The weights defined in this step are

referred to as ‘explicit weights’ to distinguish them from implicit weights that

enter the analysis unintentionally and possibly unnoticed.

Weights represent social preferences, and therefore the question of who specifies

the weights becomes important. Davis (1984) suggests that the need to assign

weights is a disadvantage because it is unclear whether it should be done by the

analyst or by politicians. More recent attitudes to decision-aiding , however, have

tended to move away from the idea of an analysis based on a single ‘correct’ set

of weights. Maimone (1985) uses three different sets of weights representing

three different points of view, a business-economic, a national and an

environmental point of view, in evaluating solid waste plans.

One of the strengths of decision-aiding is that the application of microcomputers

allows the incorporation of a large number of weight sets. It is theoretically

possible for each individual's own set of weights to be factored into a decision-

aiding computation.

Weights can be assigned directly by the individual carrying out the analysis to

represent hypothetical points of view, or they can be based on data collected from

opinion polls, focus groups, public meetings/workshops, or other direct forms of

sampling public or expert opinion. Some evaluation methods require all weights

to be quantitative, some use either quantitative or qualitative weights (rankings)

and others use only qualitative weights.

Ideally, the weights assigned in this step would be the only part of the analysis

sensitive to value judgments. In practice, this is impossible to achieve because

each of the previous and subsequent steps requires decisions that cannot be purely


23


objective. Implicit weights can enter the analysis through choice of criteria. For

example, selecting four criteria to represent environmental quality and only two

criteria to represent economic growth could introduce a weighting in favour of

environmental quality. The scale on which the criteria are scored can also

introduce an implicit weighting. Standardisation is intended to reduce this effect

but may not always be completely successful.

In Brief 9 of the SWAT case, weights were determined by a focus group of 15

people. Focus group participants were provided with 100 points and asked to

assign them across all criteria. Another alternative would be for weights to be

assigned from a "1-to-10" scale for each criterion. This method allows for each

criterion to be assessed against all others.

The computation, which will be outlined in Section 3.7, established three sets of

weights consisting of minimum values provided by the group for each criterion,

average values, and maximum values. The purpose was to assess the impact of

using different weights on the outcome of the options choice. In this way, multi-

criteria techniques can incorporate "sensitivity analysis". This concept is

discussed in more detail in Section 3.8.

3.7 Undertaking the Computation

The alternatives are evaluated by applying a mathematical procedure to the effects

table and the criteria weights, to produce a ranking of alternatives.

The choice of evaluation method is dictated to some extent by the nature of the

effects table and the type of weight information. Some methods require

quantitative data, some are designed for qualitative (ordinal) data, and some can

deal with a mixture of quantitative and qualitative data. For any problem,

however, there may be several possible methods and no obvious reason for

choosing one method over another. One approach to this dilemma is to apply

several different methods. If the results lead to similar conclusions one can be

reasonably confident that the choice of method is not critical. If different methods


24


give very different results one may be forced to conclude that the information in

the effects table is not capable of distinguishing between the alternatives in an

unambiguous way.

There are at least 40 or 50 different decision-aiding methods. In this Guide we

will introduce you to two of these. The first of these methods is called "additive

weighting", and the second is known as "concordance analysis" or "ELECTRE".

3.7.1 Additive Weighting

Additive weighting is possibly the easiest MCA technique to come to terms with.

It involves determining a standardised matrix, then multiplying weights by scores,

and finally summing the column scores to obtain rankings of alternatives.

The method begins with an effects table. This is shown for the SWAT Brief 9

case as the first matrix (“main matrix”) in Figure 3.

Looking back to Figure 2 provides more detail as to the makeup of the seven

alternatives, the seven criteria, and their measurement units. The final column on

the right hand side indicates that each criterion is a "cost", meaning that a low

score is a "good" outcome.

There are also three columns of weights, representing the "minimum", "average",

and "maximum" values taken from the focus group exercise outlined in Section

3.6.

Section 3.5 introduced the idea that there is a range of possible standardisation

formulae that can be used. The formulae used in this additive weighting

computation are presented below.

Below the effects table is the "standardised matrix". When the scores are

standardised, the value in each new cell lies between "0" and "1.00". There is a


25


range of possible standardisation formulae that can be used. The formulae used in

this additive weighting computation are presented below—they are different for

benefit and cost criteria:

Benefit:

cell value ! min value in range

max !min values [= range of values]

Cost: 1!

cell value ! min value in range

max!min values [=range of values]( ) Examine the following worked example. Alternative B for criterion 1 is a "cost"

criterion, so the standardised value is:

1!

cell value ! min value in criterion range

max!min value in the range

"

# $ %

& ' = 1!

52533 ! 32400

129238 ! 32400

" #

% &

= 1 ! 0.208 = 0.79

This can be confirmed in Figure 3.

Having produced a standardised matrix, the final step in additive weighting is to

apply the weights to the scores for each criterion and each alternative. This is

done by multiplying each score by an appropriate weight. Figure 3 indicates that

this example used the middle column of weights … the "average" weight

provided by focus group participants.

Using standard Excel spreadsheet software, the final matrix in Figure 3

("Weighted 2 Matrix") shows the outcome of the computation. The highest

ranked alternative is E (Busway West as shown in Figure 2), because it has the

highest column score. The column sums for the other alternatives indicate their

relative rankings.


26


Figure 3: Additive Weighting Example Using the SWAT Case Main Matrix

Criteria Weights A l t e r n a t i v e s Cost c, or

1 2 3 A B C D E F G Benefit b

1 0 7.6 35 48672 52533 32400 35301 80925 86430 129238 c

2 0 7.2 13 9504 8989 3911 3928 4856 5186 17752 c

3 0 3.1 10 110.7 92.9 120.9 113.8 11 9.2 21.6 c

4 0 1.8 8 2 2 1 1 0 0 2 c

5 10 12.8 20 92.5 67.2 73.2 47 47 73.2 65.2 c

6 5 11 20 181.7 166.7 81 118 118 82 263.6 c

7 5 11.4 20 205 195 248 185 137 248 178.4 c

Standardised Matrix

Criteria Weights A l t e r n a t i v e s

1 2 3 A B C D E F G

1 0 7.6 35 0.83 0.79 1.00 0.97 0.50 0.44 0.00

2 0 7.2 13 0.60 0.63 1.00 1.00 0.93 0.91 0.00

3 0 3.1 10 0.09 0.25 0.00 0.06 0.98 1.00 0.89

4 0 1.8 8 0.00 0.00 0.50 0.50 1.00 1.00 0.00

5 10 12.8 20 0.00 0.56 0.42 1.00 1.00 0.42 0.60

6 5 11 20 0.45 0.53 1.00 0.80 0.80 0.99 0.00

7 5 11.4 20 0.39 0.48 0.00 0.57 1.00 0.00 0.63

Weighted (2) Matrix

Criteria Weights A l t e r n a t i v e s

1 2 3 A B C D E F G

1 0 7 . 6 35 6.32 6.02 7.60 7.37 3.79 3.36 0.00

2 0 7 . 2 13 4.29 4.56 7.20 7.19 6.71 6.54 0.00

3 0 3 . 1 10 0.28 0.78 0.00 0.20 3.05 3.10 2.76

4 0 1 . 8 8 0.00 0.00 0.90 0.90 1.80 1.80 0.00

5 10 1 2 . 8 20 0.00 7.12 5.43 12.80 12.80 5.43 7.68

6 5 1 1 20 4.93 5.84 11.00 8.77 8.77 10.94 0.00

7 5 1 1 . 4 20 4.42 5.44 0.00 6.47 11.40 0.00 7.15

Sum = 20.25 29.75 32.13 43.70 48.32 31.17 17.58


27


Unfortunately, there is one significant problem associated with the use of additive

weighting in the SWAT Brief 9 example. As was specified in Section 3.5,

ordinally scaled data cannot be multiplied, which is what happened in Figure 3 for

criterion 4 ("vibration").

In cases such as these, where criteria are measured in both ratio scale and ordinal

scale, different decision-aiding techniques designed to deal with both quantitative

and qualitative data must be used.

The example outlined in Section 3.7.2 is one such technique.

3.7.2 Concordance Analysis

Concordance analysis is a pair-wise comparison technique. This involves

comparing the cell values between two alternatives across the full range of

criteria, and for all combinations of pairs. The power of concordance analysis is

that it does not work directly with scores. Instead, all mathematical operations

are applied to weights. This enables the technique to be applied to mixed

quantitative and qualitative (ordinal) data.

The computation begins with the same effects table as was used in the additive

weighting example. This is presented as the first matrix in Figure 4. However,

what happens in the next step is quite different.

The values in the "Concordance (2) Matrix" are developed by adding together all

the weights associated with the criteria, where the value of the principal cell is

"better" than the value of the alternative cell, then dividing the sum of those

weights by the total of all the weights. Here "better" is defined differently for

benefit and cost criteria. For a benefit criterion the larger the number the better

the value; for a cost criterion the lower the number the better the value.

Figure 4: Concordance Analysis Example Using the SWAT Case


28


Main Matrix

Criteria Weights A l t e r n a t i v e s Cost c, or

1 2 3 A B C D E F G Benefit b

1 0 7.6 35 48672 52533 32400 35301 80925 86430 129238 c

2 0 7.2 13 9504 8989 3911 3928 4856 5186 17752 c

3 0 3.1 10 110.7 92.9 120.9 113.8 11 9.2 21.6 c

4 0 1.8 8 2 2 1 1 0 0 2 c

5 10 12.8 20 92.5 67.2 73.2 47 47 73.2 65.2 c

6 5 11.0 20 181.7 166.7 81 118 118 82 263.6 c

7 5 11.4 20 205 195 248 185 137 248 178.4 c

Concordance (2) Matrix

(for weights 2)

A B C D E F G

Metro rail east A 0.85 0.74 0.94 0.86 0.65 0.51

Metro rail west B 0.15 0.50 0.94 0.86 0.42 0.51

Access rail east C 0.26 0.50 0.51 0.53 0.31 0.50

Access rail west D 0.06 0.06 0.49 0.51 0.29 0.26

Busway west E 0.14 0.14 0.47 0.49 0.27 0.00

Busway east F 0.35 0.58 0.69 0.71 0.73 0.44

MRS amendment G 0.49 0.49 0.50 0.74 1.00 0.56

Av. cell value = 0.50

C- Dominance (2) Matrix

(for weights 2)

A B C D E F G

Threshold 1: A 1 1 1 1 1 0

0.6 B 0 0 1 1 0 0

C 0 0 0 0 0 0

D 0 0 0 0 0 0

E 0 0 0 0 0 0

F 0 0 1 1 1 0

G 0 0 0 1 1 0

Sum = 0 1 2 4 4 1 0

Comparison of Concordance Dominance for Different Thresholds

(created manually) A l t e r n a t I v e s

A B C D E F G

0.3 2.00 4.00 6.00 6.00 6.00 4.00 4.00

0.4 1.00 4.00 6.00 6.00 6.00 3.00 4.00

Threshold Values 0.5 0.00 2.00 4.00 5.00 6.00 2.00 2.00

0.6 0.00 1.00 2.00 4.00 4.00 1.00 0.00

0.7 0.00 1.00 1.00 4.00 4.00 0.00 0.00

0.8 0.00 1.00 0.00 2.00 3.00 0.00 0.00

0.9 0.00 0.00 0.00 2.00 1.00 0.00 0.00

0.95 0.00 0.00 0.00 0.00 1.00 0.00 0.00


29


In other words, in the case of a “cost” criterion (lowest is best), if the value for the

“principal” cell is less than the alternative cell, then the relevant cell in the new

matrix is given the value of the weight for that criterion. In the case of a "benefit"

criterion (highest is best), if the value for the "principal" cell is more that the

alternative cell, then the relevant cell in the new matrix is given the value of the

weight for that criterion.

Where scores are equal, the weights are added and then divided by two.

So, the value of the new cell will be:

! weights for each benefit criterion

where the cell score > cell score

of the alternative.

"

#

$ $ $

%

&

' ' '

+

! weights for each cost criterion

where the cell score < cell score

of the alternative.

"

#

$ $ $

%

&

' ' '

+

! weights divided by 2 for

criteria where cell scores

are equal.

"

#

$ $ $

%

&

' ' '

! all weights

Consider the following example calculation of a Concordance (2) Matrix cell, for

the principal alternative C, compared to alternative G, for weight 2 (hence the

term "Concordance [2] Matrix").

Criterion Weights

b or c If b, then is C>G?

If c, then is C<G?

If any yes, then use weight

1 7.6 c yes 7.6 2 7.2 c yes 7.2 3 3.1 c no 4 1.8 c yes 1.8 5 13 c no 6 11 c yes 11.0 7 11 c no Sum: 27.6


30


The value of cellgc = 27.6 divided by the sum of all weights = 27.6/54.9 = 0.502.

This number appears in the Concordance (2) Matrix cell in Figure 4, in the third

column … second cell from the top.

The significance of this number (ie 0.502) is that alternative C dominates

alternative G for 50.2% of the weights.

Further manipulation has to occur, however, before the data presented in the

Concordance (2) Matrix can be used to determine a ranking of alternatives. The

next step is to develop what is known as a "concordance dominance" matrix. In

this matrix (the third from the top in the Figure 4 example), the dominant

alternative is given a “1” and the subservient “0”, when also compared against an

arbitrary "threshold" level. In the Figure 4 example, the threshold has been set at

0.60.

An examination of the third column in the Concordance (2) Matrix shows that

only the first cell (0.74) and the second-last cell (0.69) have a value that is greater

than the threshold of 0.60. These two cells are thereby converted to "1s" in the

Concordance Dominance Matrix.

When the columns in the Concordance Dominance Matrix are summed, we can

gain some insight as to the ranking of alternatives using the concordance analysis

technique. The C-Dominance (2) Matrix in Figure 4 indicates that alternatives E

and D are equally ranked.

This result is interesting because it equates to that obtained in the additive

weighting example, where alternative E was slightly higher ranked than

alternative D according to the summed column scores.

Some final explanatory words are needed about concordance analysis. First, it is

useful because it permits comparison between decision criteria without requiring

the building of a mathematical. model of interrelationships. In addition,


31


information is only required on which to make judgements of an ordinal (relative

rank) nature.

However, a word of warning is also required. Concordance is useful for

indicating how much better one alternative is than another. For criteria where one

alternative is worse than another, the technique provides no information. So,

using concordance we could have one alternative better than another on all

criteria except one. This doesn’t necessarily mean that this alternative is the best

one, because the one “bad” criterion could be really bad!

3.8 Dealing with Uncertainty

One of the strengths of decision-aiding is its ability to take uncertainty into

account. It is possible to deal with uncertainty by varying weights and actual

methods. Both of techniques presented in this section had three sets of weights

that could have been operated upon. Our computations chose the focus group's

"average" weight set for the sake of example. If we had chosen a different weight

set, we may have seen a different outcome.

What this means is that, for example, different interest groups could have their

own weight sets incorporated into a computation to see what happens to the

outcome. It may be that in this SWAT case, a calculation undertaken using an

environmental group's weight set might result in a different ranking of

alternatives than if a transport engineer's weight set were to have been used.

This testing of outcomes using different inputs is called "sensitivity analysis".

Sometimes it so happens that inputting different weight sets from different

interest groups does not substantially affect the ranking of alternatives. In cases

such as this, the outcomes are said to be "robust".

Another possible sensitivity test (used only in concordance analysis) is to set the

threshold in the C-Dominance Matrix to different levels. This allows for a


32


"sifting" of alternatives if a number happen to be equally ranked. In Figure 3,

alternative E only stands out from alternative D when the threshold is set at 0.80.

In the SWAT case that we have worked with in this section, varying of technique

seems to have made no great difference. Alternatives D and E come out well for

both methods. There is not the space here to undertake a sensitivity test based on

using different weights. In the actual case, varying weights did not significantly

affect the outcome either.

3.9 Presentation of Results

In the example presented in this section, we have shown how decision-aiding can

be structured to incorporate public participation during the choice of alternatives,

the listing of comparison criteria, and the application of value weights.

These are significant inputs, especially when compared to the limited "comment"

procedures usually allowed for in traditional project development. Experience is

showing that decision-aiding is a powerful conflict resolution technique because,

if organised properly, it allows people to feel a degree of "ownership" of the

process. Commitment to the process of decision-aiding assigns a level of

legitimacy to it. As a consequence, it is then much harder for participants to

disagree with the outcome.

However, for this point to be reached, it is necessary for stakeholders to be

involved from the start of the process and in all phases of alternatives selection,

criteria listing, and weighting. Finally, it is also important to allow stakeholders

to manipulate the computerised models and see how they work.

4. CONCLUSION

The aim of this manual has been to introduce decision-aiding techniques and how

they might work as a useful tool in the evaluation phase of environmental impact

assessment.


33


Two techniques have been introduced in this manual, although many more exist

in practice.

Enough information has been provided here for interested students of decision-

aiding techniques to begin using existing software, or to programme their own if

so inclined. Annandale and Lantzke have undertaken some programming

themselves, and have obtained access to a small selection of proprietary software.

Examples may be viewed at the website for a Murdoch University School of

Environmental Science course called, "Techniques for Environmental Impact

Assessment". The URL for this course is:

http://science.murdoch.edu.au/teach/n422/


34


5. REFERENCES

David, B. 1984, 'Management of non-wood values in state forests: should the use

pay?, Australian Forestry, vol. 47, pp.143-147.

McAllister, D.M. 1986, Evaluation in Environmental Planning: Assessing

Environmental, Social, Economic, and Political Trade-Offs, The MIT Press,

Cambridge, Massachusetts, 308 pp.

Keeney, R. and Raiffa, H. 1976, Decisions with Multiple Objectives: Preferences

and Value Trade-Offs, John Wiley, New York.

Maimone, M. 1985, 'Using multi-criteria evaluation in developing solid waste

plans', in A Faludi and H Voogd (eds), Evaluation of Complex Policy

Problems, Delftse Uitgevers Maatshcappij, Delft.

Parkin, J. 1994, Judging Policies and Plans, Avebury Press.

Resource Assessment Commission 1992, Multi-criteria Analysis as a Resource

Assessment Tool, RAC Research Paper No.6, Commonwealth of Australia.

Vansick, J-C, 1990, 'Measurement theory and decision aid', in C.A. Bana e Costa

(ed), Readings in Multiple Criteria Decision Aid, Springer, Berlin.

Varis, O. 1989, ‘The analysis of preferences in complex environmental

judgements—A focus on the analytic hierarchy process’, Journal of

Environmental Management, vol.28, pp.283-294.

Westman, W.E. 1985, Ecology, Impact Assessment, and Environmental Planning,

John Wiley & Sons, New York, USA, pp. 532.


Appendix B

Attendees at Water CorporationPreliminary Workshop 8th May2007


MCA Planning Workshop Attendees

Mundaring Water Treatment Plant Site Selection

MultiCriteria Analysis Planning Workshop

8th May 2007 9 am – 1 pm

Attendees

Robert Ng: Water Corporation

Jessica Teh: Water Corporation

Ken Walter (part): Water Corporation

Julia Krsnik: Water Corporation

Natalie Reilly: Water Corporation

Steve McKenzie: Water Corporation

Peter Marchesani: Water Corporation

George Malita: Water Corporation

Suzanne Brown (part): Water Corporation

Lynette Lund: Water Corporation

Neil Formosa: Water Corporation

Megan Dilly: GHD

Jenny Pope: GHD


Appendix C

Summary of Sites

61/16611/71624 Water Corporation – Mundaring Water Treatment Plant Site SelectionSummary of Site Information for Shortlisted Sites July07

Summary of Site Information for Shortlisted Sites, July 2007IntroductionThis document contains a summary of site information for the four short listed sitesunder consideration for the location of the Mundaring WTP. The document has beenprepared by GHD using information provided by the Water Corporation and variousexternal consultants. The detailed reports from which the summary information istaken are mentioned in the summary table and are available if further detail is required.

The information is to assist community participants complete their individual ranking onthe four short listed sites and also describes the sustainability criteria that the WaterCorporation will use to undertake its own multicriteria assessment of the four sites.

The four remaining sites are:

Site 1

Below Mundaring Weir, on Mundaring Weir Road, south of the Helena River. This siteis located in Reserve 5342 and includes Reserve 39644, which is vested in the WaterCorporation for the purpose of Water Supply Mundaring Weir, and is zoned Parks andRecreation in the Metropolitan Region Scheme (MRS).

Note that if Site 1 is selected, the chlorine store and the drying beds will be located atone of the other three sites.

Pine Plantation

Located to the north of Mundaring Weir, alongside an airstrip (managed by DEC).The site lies off Allen Road, to the east of Mundaring Weir Road. This site iswithin State Forest and is zoned Water Catchments and State Forest in the MRS.

DEC Land (formerly “CALM Land”)

DEC Land is a 20 ha site located at the current DEC depot to the east of MundaringWeir Rd, at Mundaring Weir, WA. The site is located on State Forest and Freeholdland owned by DEC.

O’Connor SiteThe O’Connor site is a 39.7ha site located in State forest at the junction of MundaringWeir Road and Firewood Road, north of the weir. The site is east of Mundaring WeirRoad and north of Firewood Road.

This document firstly describes the 11 sustainability criteria that will form the basis ofthe multicriteria analysis (MCA) that the Water Corporation will use to assist in itsassessment of the sites. It then provides a summary table of the results of the studiesundertaken to assess the impacts of each of these sites with respect to thesustainability criteria. A table with a summary of the major advantages anddisadvantages is also included.


Description of Sustainability Criteria to be used in the MCA for theMundaring WTP Site Selection

Biodiversity

This criterion relates to the biodiversity values of the site, including the clearingfootprint required. Biodiversity is the variety of all life forms, the different plants,animals and microorganisms, the genes they contain, and the ecosystems of whichthey form a part.

In the case of the proposed Mundaring WTP, the issues to be considered under thesustainability criterion of impact on biodiversity are:

» The overall ecological value of the sites;

» The conservation significance of the sites;

» The clearing footprint required at the sites (the amount of native vegetation that willbe cleared);

» The value of the site for native flora; specifically the diversity of native flora at thesite, any significant flora species or vegetation types at the site and the health of thevegetation at the site;

» The value of the site for native fauna; specifically, the potential of the site tosupport native species, including significant species; the value of the site as faunahabitat and the use of the site as a habitat linkage.

Embedded energy of assets

This criterion refers to the amount of energy that has been used to produce thematerials from which the WTP will be constructed. Low embedded energy equates tolow energy use and also low emissions of greenhouse gases, and is therefore adesirable environmental goal. Embedded energy relates to the quantity and types ofmaterials used in the construction of the WTP, and in this case will be stronglycorrelated with the lengths of pipe required for each site (i.e. less pipe means lessembedded energy).

Please note that the amount of energy required for operation of the sites has not beenconsidered as this will be nearly identical for all sites and thus does not add any valueto the comparison of sites.


This criterion relates to the potential impacts of the WTP on waterways, which for thispurpose includes river and stream systems and wetlands, and on watercycles. Thedesirable environmental goal for this criterion is that waterways (including minordrainage lines) and associated riparian vegetation, should be protected anddevelopment should be excluded from the buffer area of a waterway. The desirableenvironmental goal for water cycles is that the project would minimise impacts andalterations to the natural watercycles. This incorporates consideration of catchmentissues. Protection of catchments occurs through restricting certain types of


development in catchment areas and by implementing management measures forpractices that carry a risk of contaminating surface and groundwater.


Land degradation is a serious environmental problem in Western Australia and can bedefined as the decline in condition or quality of the land as a consequence of humanactivities. For the purposes of the project this criterion addresses potential issuesassociated with the construction and operation of the WTP as they relate to erosionand soil degradation at the site and the potential to introduce diseases and weeds tothe general area. When considering this criterion it is necessary to assess the currentstatus of the land and the extent of degradation of the site, ie what soil degradationalready occurs and whether weeds and diseases (specifically dieback) already occur atthe site. It also assesses the sensitivity of the site to land degradation factors.

Hazardous chemical risksWhile a number of chemicals are used in the water treatment process, the chlorineused for disinfection of the water is the most significant in terms of potential hazardsoutside the WTP boundaries. Chlorine is therefore the focus of the assessment ofhazardous chemical risks. The risk assessment process involves identifying differentscenarios under which chlorine could be released; modeling the resulting concentrationof chlorine at various distances from the release; and assessing the acceptability withinjury limits. While most of the sites meet the WA risk criteria, even in the event of amajor chlorine release, some sites are closer to residences and recreational facilitiesand have therefore been assessed as having a slightly higher risk.

Community amenity

Community amenity is a general term often used to describe the potential impacts of adevelopment on the lifestyle of the local community. In the case of the proposedMundaring WTP, the issues considered under community amenity were: visualamenity, i.e. what the WTP looks like and what impacts it might have on the outlookfrom residents’ properties, including impacts of plant lighting; noise, i.e. whether thenoise generated by plant operations will affect local residents; and traffic, i.e. whethertraffic entering and leaving the WTP site will create impacts in terms of noise,congestion or road safety.


This criterion relates to the potential impacts of the WTP on the users and managers ofrecreational and tourism facilities in the area, in terms of convenience and enjoyment.Some of these impacts may be direct, for example, some walking trails pass throughproposed WTP sites or their buffer zones and these would have to be rerouted. Otherimpacts are less tangible, for example, the WTP may impact on the visual amenity ofpeople using facilities such as picnic areas, museums, or scenic lookout.

European heritageConsideration of potential impacts of the WTP on European heritage has manydimensions. The construction of the WTP on some sites could disrupt historical


artifacts or areas where it is expected that artifacts could be found based uponknowledge of the site’s history, or could require old buildings to be moved from theiroriginal locations, which is undesirable from an interpretive perspective. On the otherhand, as recognised by the National Trust, the WTP could have a positive contributionto heritage values, representing the latest chapter in the story of Mundaring Weir andthe Goldfields water supply. Both potential negative aspects and positive enhancementopportunities have been taken into consideration in assessing the overall Europeanheritage impacts of each site.

Indigenous heritage

The assessment of the indigenous heritage values of each site is based on thesignificance of the sites to the indigenous groups in the area, and the presence ofarcheological sites containing indigenous artifacts.

Site flexibility and operabilityThis criterion relates to the planning, design and operational phases of the project froman engineering perspective. Site flexibility is a reflection of the range of optionsavailable to planners and designers in configuring the WTP on the site, as well as thesite’s potential to accommodate changing circumstances and new technologies.Operability relates to the ease with which a plant located on the site could be operatedand maintained on a daily basis, which is related to its layout and design. In assessingthe flexibility and operability of each site the following factors were considered:expandability, topography, plant size, accessibility, security, ease of fire management,proximity to raw water and ease of discharge of backwash water.

Costs

The costs associated with the WTP at each potential site have been calculated byWater Corporation and expressed in terms of net present value (NPV), which is astandard method for the financial appraisal of longterm projects. It takes intoconsideration all future cash flows, capital and operating, and discounted them to apresent value after tax. The total NPV figure for each site incorporates planning,design, construction, operating and maintenance costs over the life of the plant. WaterCorporation uses a discount rate of 7.1%, an escalation (inflation) rate of 2.5% and atax rate of 30%.

Operating costs depend on two major operating components; pumping and treatingwater. The significant difference in elevation (about 210m) between Mundaring Weirand the Sawyers Valley Tank contributes to nearly all the lifts required by the pumps.The frictional head losses are small. Treatment costs are similar for all sites. Theamount of water to be treated is the same for all sites.

Because pumping and treatment costs are similar for all sites, the total operating coststherefore will be similar for all four sites. And the present value of capital expenditureswill provide a good basis to the comparison of sites.


Summary Table of Sitespecific Information (Four shortlisted sites)

Sustainability Criteria Site One DEC Land Pine Plantation O’Connor Site

Environmental

Biodiversity

Source:

GHD (2007) Mundaring Water TreatmentPlant Site Selection: Preliminary Flora andFauna Assessment – Site One, June 2007

GHD (2007) Mundaring Water TreatmentPlant Site Selection: Preliminary SpringFlora and Fauna Assessment – DECLand, June 2007

GHD (2007) Mundaring Water TreatmentPlant Site Selection: Preliminary SpringFlora and Fauna Assessment – PinePlantation Site, June 2007

GHD (2007) Mundaring Water TreatmentPlant Site Selection: Preliminary Flora andFauna Assessment – O’Connor Site, June2007

NB: The Bush Forever VegetationCondition ratings scale was used to ratethe vegetation condition of the site(Government of WA, (2000). Bush ForeverVolume 1. Policies, Principles, Processes.Department of Environmental Protection,Perth, Western Australia)

Site One has been subject to a longhistory of disturbance. Most of the site isdegraded with a high level of weedinvasion.

The site contains remnant scatteredmature Flooded Gum and Marri trees.The granite system in the east and theriparian vegetation near Helena River arestill in relatively good condition. Theconceptual plant layout shows that themajority of these areas will be retained.

There may also be impacts on biodiversitydepending on which of the other sites thechlorine store and drying beds will belocated on.

According to the conceptual layouts theWTP at Site One will require clearing ofthe following amount of vegetation ratedcondition Good (4) or better:

» Site One plus Chlorine and drying bedsat DEC Land: 2 ha in condition Good(4) or better.

» Site One plus Chlorine and drying bedsat Pine Plantation: 2.24 ha in conditionGood (4) or better.

» Site One plus Chlorine and drying bedsat O’Connor Site: 5.68 ha in conditionGood (4) or better.

A Priority 3 flora species was recorded onthe site but is outside the WTP footprintand impacts on this species can beavoided.

The value of the site for native fauna islimited due to its degraded nature.

If the WTP was located at Site One theexisting pipelines from the weir can beutilised to carry the treated water from theweir to the site of the chlorine store. Allthe other sites will require a third pipelinenorth of the weir and this may requiresome clearing of native vegetation andhave some impacts on biodiversity.

A significant portion of the site has beencleared in the past, with most of thecentral section being highly degraded andcontaining predominantly introducedspecies. There are some areas of nativevegetation remaining, with good qualityvegetation in the north of the site (gullysystem and granitic herblands) and to theeast of the site (JarrahMarri Forest).

The conceptual plant layout shows theWTP situated in the central area of thesite where the vegetation is highlydegraded. The layout impacts on 0.65 haof vegetation rated condition Good (4) orbetter.

The majority of the site has beensignificantly altered due to the oldsettlement that was located on it andwould provide little value as habitat tonative fauna species. Some species,particularly bird species could utilise theintroduced garden plants for foraging, butin general the disturbances on the sitehave reduced its value for fauna.

The northern section of the site has highhabitat value as it contains a number ofdifferent habitat types, including graniteoutcrops and a gully system. Theconceptual plant layout shows that thisarea will be retained.

The vegetation at this site is fairly uniform,with two main vegetation types occurring;a JarrahMarri forest in the northwest partof the site and a pine plantation in thesoutheast of the site.

The pine plantation was generallyCondition 5 (Degraded) as it containedpredominantly introduced species, withsome scattered natives.

The conceptual plant layout shows theWTP situated primarily in the pineplantation with very little impact on theareas of native vegetation. This layoutimpacts on only 0.52 ha of vegetationrated as condition Good (4) or better atthe site. However, there will be impactson native vegetation from the requirementto widen the access track into the site.This will require vegetation adjacent to theexisting Allen Rd to be cleared over thealmost 1km of track into the site.

The pine plantation offers limited value tonative fauna, it provides some feedinghabitat for species but is of less valuethan the native vegetation surrounding theplantation.

This site requires a substantial length ofaccess road, as well as pipelines. Thiswill result in clearing of native vegetationand may also impact on fauna linkages asthe access track and pipelines may resultin a substantial barrier to faunamovement.

The majority of the O’Connor Sitesupports Jarrah and Marri Forest. Thesite has been logged previously and thevegetation is regrowth.

Small sections of the site, particularlyaround the access track in the centre ofthe site are highly disturbed, and containsome planted nonnative species. Thecondition of the vegetation improves awayfrom the road and while there isdisturbance evident in these areas thevegetation structure remains intact andthe site retains the ability to regenerate toa better condition.

The conceptual plant layout shows theWTP centred around the access track,which is the most degraded section of thesite. However, this layout will still requireclearing of native vegetation. This layoutimpacts on 12.98 ha of vegetation ratedas condition Good (4) or better at the site.

This site contains the largest contiguousarea of native vegetation of all the sites.While this vegetation has been previouslydisturbed it still has the highest flora andfauna values of all the sites beingconsidered in this assessment.



Embedded energy

Source: Relevant Water Corporationexperts

Pipe Lengths

» Site One plus Chlorine and drying bedsat DEC Land: 3.2km, Rank 2nd

» Site One plus Chlorine and drying bedsat Pine Plantation: 5.1km, Rank 4th

» Site One plus Chlorine and drying bedsat O’Connor Site: 2.1km, Rank 1st(best)

Pipe Lengths: 3.4km, Rank 3rd Pipe Lengths: 5.3km, Rank 5th Pipe Lengths: 9.9km, Rank 6th (worst)


Source:

GHD (2007) Mundaring WTP SiteSelection: Preliminary ComparativeEnvironmental Impact Assessment,Report prepared for Water Corporation,July 2007.

Two waterways (drainage lines) passthrough this site. One flows through thegully in the centre of the site and oneflows through the western section of thesite. Both waterways drain into theHelena River. The Helena River passesthrough the northwest corner of the site.The conceptual plant layout shows thatimpacts on the Helena River and thewestern drainage line can be avoided butthe central waterway will be impacted bythe WTP.

There are two waterways (drainage lines)through this site, within the two gullysystems that run eastwest. Thesedrainage lines are tributaries of BendingGully, which runs southwest through FredJacoby Park (to the west of the site).

The conceptual layout shows that thelarger drainage line in the northern gullywill not be impacted by the WTP.However, the smaller drainage linethrough the central gully will be impactedby the plant layout and the drainage in thecentral section of the site will be altered.

There is one waterway (drainage line) thatflows southwest through the pineplantation site and drains to MundaringReservoir. This drainage line flowsthrough the minor gully that is to the southof the central access track.

This site is situated within a ReservoirProtection Zone. The Department ofWater has indicated that if the WTP is tobe located here substantial drainageworks will be required to engineer the siteout of the catchment. This will require anartificial drainage system that will drain theentire site to a sump. The drainage waterwould then need to be pumped out of thecatchment area. This is a significantalteration of the hydrology of the site andwill be very expensive to construct andoperate. This process will also haveongoing energy requirements.

This site will have the highest impact onwaterways and watercycles

This site contains one waterway, in a gullyalong the southern boundary of the site.This waterway is a tributary of BendingGully.

The conceptual plant layout for this siteshows that impact on waterways at thissite have been generally avoided, withonly the access track impacting on thedrainage line in the south of the site.

This site is expected to have the leastimpact on waterways and watercycles.


Source:


Site One has been historically disturbedand is subject to a number of ongoingdisturbances. Much of the site is clearedor contains weedy species and thepotential for land degradation at this site islow in comparison to the other sites.

Some areas of the site are sloped andhave the potential to lead to landdegradation. While much of thevegetation at the site is introduced and itcontains a number of weedy species, theclearing of this vegetation has thepotential to lead to erosional impacts.

While the WTP layout at this site isprimarily within pine plantation it issurrounded by native vegetation.Additionally, native vegetation will need tobe cleared for the access track into thesite.

Clearing of this site and the constructionand operation of the WTP have thepotential to cause land degradation of theareas of native vegetation adjacent to thesite, particularly through the risk of theintroduction of weed species and dieback.

This site is in the most natural condition ofall the sites. It has the most completenative vegetation cover and has the leastweed invasion. There is potential forconstruction and operation of the WTP tocause land degradation at this site.

Social




Source:

GHD (2007) Technical Note: MundaringWater Treatment Plant Chlorine Storageand Dosing – Water Treatment PlantLocation Sensitivities

Meets all EPA safety requirements.However, the Western Power Substation,Museum and Mundaring Weir Picnic Areaand children's playground are within theinjury impact zone, as are sections of theKattamorda Trail, Munda Biddi Trail andBibbulmun Track in the vicinity of Site 1.However, should Site 1 be selected as thepreferred site, Water Corporation intendsthat the chlorination plant would belocated at one of the other three sites,separate from the main water treatmentplant. If no hazardous chemicals werestored at Site 1, EPA requirements wouldclearly be met and the chance ofimpacting members of the public would beminimal.

Meets all EPA safety requirements. Theinjury impact zone encroaches intosections of the Fred Jacoby Park, MundaBiddi Trail, Kep Track and MundaringLoop Bike Trail, but this does notcontravene EPA requirements for activeopen spaces.

Meets all EPA safety requirements. Theinjury impact zone only encroaches ontothe airstrip south of the proposed site, andthis does not contravene EPArequirements for active open spaces. Furthermore, the airstrip would only beused during an emergency, which makesany impact on the public extremelyunlikely.

Meets all EPA safety requirements. Theinjury impact zone encroaches ontosections of the Munda Biddi Trail, KepTrack and Mundaring Loop Bike Trail, butthis does not contravene EPArequirements for active open spaces.Furthermore, walkers would be at low riskof injury since they would be unlikely to bestationary within the impact zone.

Community amenity Minimal, as no houses are located in thevicinity.

The main impact on community amenitywill be from transport to the site.

However, depending on which site is usedfor the chlorine store and the drying bedsthere may be some impacts on communityamenity from this additional area.

There are no private residences in theimmediate vicinity of the site and impacton community amenity will be low.However, there are residences to thesouth and to the north of the site andpotential indirect impacts on theseresidences are possible.

Minimal, as no houses are located in thevicinity.

The main impact on community amenitywill be from transport to the site.

This site is the closest of all the sites toresidential properties. A buffer area hasbeen retained between these propertiesand the site but amenity impacts forresidence may still be an issue at this site.

Tourism and recreational values

Source:


Within Site One there are picnic sites andpublic recreation areas. There are anumber of walking trails, including theBibbulmun Track, the Munda Biddi Trailand the Kattamorda Trail, that run veryclose to the site, both to the west and tothe east along the weir wall.

To the north of the site is a heritage andtourism area, which includes a museum,the C Y O’Connor memorial andwalktrails. The Mundaring Hotel/Mundaring Hall are located approximately350m north from the site.

The WTP at this site would be visible frompublic areas, including the weir wall,lookouts and picnic sites.

Under the Mundaring Weir MasterplanSite One is within a proposed heritagetourist precinct.

There are a number of tourist andeducational facilities surrounding the DECLand, including the Kookaburra OutdoorCinema adjacent to the site, the HillsForest Discovery Centre, about 100across Allen Rd from the site, and FredJacoby Park, across Mundaring WeirRoad.

A number of walking trails pass through ornear to the site, including the KattamoordaTrail, Kep Track and the Mundaring LoopBike Trail, which pass along the westernedge of the site and then through itsnorthern corner. Additionally, other majortrails in the near vicinity are the MundiBiddi trail, approximately 300m west, andthe Bibbulmun track, approximately 50msouth. The site is around 550m northeastof the picnic sites associated withMundaring Weir and the Mundaring WeirHotel is situated about 500m from the site.

The pine plantation site is relativelyremoved from the high use tourism andrecreation areas around Mundaring Weirand impacts on recreation and tourismshould be relatively low. However, a WTPsite at this site may cause visual impactson users of the Bibbulmun Track, whichruns through the northwest corner of thesite.

This site is surrounded by state forest andsituated away from the recreational,tourism and heritage areas associatedwith the weir. However, the site isadjacent to Mundaring Weir Road and theKattamoorda Trail, Kep Track, MundaBiddi Trail and Mundaring Loop Bike Trailare all located along the western boundaryof the site.



Indigenous heritage

Source: Villiers, L (2007) Report of anArchaeological Survey of the ProposedWater Treatment Plant and AssociatedInfrastructure at Mundaring, WesternAustralia. Report prepared by AustralianInteraction Consultants for WaterCorporation

No archaeological sites found. Preferredsite, as it is already highly disturbed.

No archaeological sites found. Retainssome native vegetation and therefore it ispossible that intact archaeologicalmaterials could be encountered duringearthworks.

No archaeological sites found. Areacurrently under pines is already disturbedand therefore suitable for development.However, for area not under pine ( thenorthwest portion of the site) it is possiblethat intact archaeological materials couldbe encountered during earthworks.

No archaeological sites found. Leastdisturbed site and therefore there is apossibility of encountering archaeologicalmaterials during earthworks.

European heritage

Source: Rosario, R. (2007) MundaringWeir Water Treatment Plant HeritageAdvice: Preliminary European HeritageSurvey Heritage Audit for Alternate Sites.Report prepared by Heritage andConservation Professionals for WaterCorporation

Contains a number of significantarcheological sites. Close to MundaringWeir, which is included in Shire ofMundaring’s municipal heritage inventory.Due to its scale, it would be difficult todevelop the WTP in manner sympatheticto the existing cultural precinct. Leastpreferred site by the National Trust andheritage consultant.

Significant heritage values. Location ofone of the first divisional headquarters ofthe Forests Department. Contains anumber of heritage buildings dating fromthe 1920s, some of which have beenmoved to the site. Construction of theWTP on this site would result in loss orrelocation of the heritage buildings.Historically the cottages built by the forestdepartment have been relocated and itwould be possible to relocate any of theremaining buildings on this site if analternative location is identified. However,the opportunity to interpret the history ofthe settlement in its original location wouldbe lost.

No specific heritage issues associatedwith site. However, site is not favoured bythe National Trust because it is too farfrom Mundaring Weir to allow forinterpretation of the WTP as part of theongoing story of Mundaring Weir and thegoldfields water supply (i.e. to achieve apositive heritage outcome).

No existing heritage value. Opportunity forpositive contribution to heritage values.

Economic


Source: Workshop with relevant WaterCorporation experts

Rates poorly due to site’s constrainedsize, rugged topography and the need forremote location of the chlorination plant.

Good or acceptable performance withrespect to all issues considered.

Rates lowest overall due to stringent watermanagement requirements (due tolocation in Reservoir Protection Zone);poor fire management characteristics (dueto dense vegetation and limitedaccessibility); and high security risk (dueto relatively remote location).

Rates highest overall due to large areaavailable and site accessibility. Goodperformance with respect to all issuesconsidered, except for proximity to rawwater.



Costs

Source: Relevant Water CorporationExperts

» Site One plus Chlorine and drying bedsat DEC Land:

Capital Cost $130 million

Operating Cost $158 million

Total Cost $288 million

» Site One plus Chlorine and drying bedsat Pine Plantation:




» Site One plus Chlorine and drying bedsat O’Connor Site:














Summary Advantage/Disadvantages Table (four shortlisted sites)

Site Advantages Disadvantages

Site 1 » Close to Mundaring Weir,with cost and operationalbenefits

» Site has been previouslycleared and ispredominantly degradedso biodiversity impacts willbe low (though there maybe biodiversity impacts onthe additional site requiredfor chlorine and dryingbeds)

» Minimal impact oncommunity amenity(depending on where thechlorine store and dryingbeds are located)

» Site is already highlydisturbed and thereforelittle chance of disruptingindigenous artefacts

» Site One is the only sitethat will not require anadditional main pipelinefrom the weir and thus willnot have the associatedimpacts from the pipelinein a constrained area(potential social andenvironmental impacts)

» Chlorination plant must belocated remotely, due tospace and safetyconsiderations. Thisadditional location mayhave environmental andsocial issues associatedwith it.

» Site has ruggedtopography and isconstrained in size

» Contains a number ofsignificant Europeanheritage archeologicalsites.

» Difficult to develop theWTP in mannersympathetic to the existingcultural precinct. Willimpact on existinglanduses of the site,including recreation areasand the Western Powersubstation.

» Will be visible from anumber of locations,including lookouts

DEC Land » Close to Mundaring –Kalgoorlie pipeline

» Opportunity for positivecontribution to heritagevalues as part of theongoing story ofMundaring Weir and theGoldfields water supply

» Much of the site isdegraded and containsintroduced species,impacts on flora and faunawill be minimal

» Good or acceptableperformance with respectto flexibility and operabilityof the site.

» Significant heritagevalues, including buildingsthat would be either lost orrelocated, so theopportunity to interpret thehistory of the settlement inits original location wouldalso be lost.

» Adjacent to communityfacilities, including FredJacoby Park, KookaburraOutdoor Cinema, ForestDiscovery Centre

» There would be a risk ofdisrupting indigenousartifacts

» A number of walking andcycling trails pass through


Site Advantages Disadvantagescycling trails pass throughsite, these may need to bererouted

PinePlantation

» WTP could be located in apine plantation withoutmajor direct impacts onadjacent nativevegetation.

» Low risk of disruptingindigenous artefacts, asarea under pine plantationis already disturbed

» No existing Europeanheritage values

» Minimal impact oncommunity amenity due toremote location

» Removed from heritageand tourism precinct

» Located within ReservoirProtection Zone andtherefore not supported byDepartment of Water.Would require extensivedrainage system (highcost and ongoing energyrequirements)

» Too far from MundaringWeir to allow forinterpretation of the WTPas part of the ongoingstory of Mundaring Weirand the Goldfields watersupply

» The Bibbulmun Trackpasses through the site(potential indirect impacts)

» Significant distance fromMundaring – Kalgoorliepipeline – additional pipingcosts

» Access route fromMundaring Weir Rd to sitemay impact on touristareas such as ForestDiscovery Centre

» The access track will alsorequire clearing of nativevegetation and mayimpact on fauna linkages

» Adjacent to DEC airstrip

O’ConnorSite

» No existing Europeanheritage values.Opportunity for positivecontribution to heritagevalues as part of theongoing story ofMundaring Weir and thegoldfields water supply

» Large site available, highlevel of flexibility

» Close to Mundaring –Kalgoorlie pipeline

» Would require clearing ofgood quality vegetation

» Least disturbed site andtherefore there is apossibility of encounteringarchaeological materialsduring earthworks

» Two private properties(including residences)adjacent to WTPboundary, plus otherprivate properties nearby

» A number of walking andcycling trails are adjacent


Site Advantages Disadvantagesto this site, these may besubject to aestheticimpacts or may need to bererouted

» Substantial lengths ofadditional pipelinerequired


Appendix D

Community Amenity


Mundaring Water Treatment Plant Site Selection MultiCriteria Analysis (MCA): Community Amenity ScoringProcessOutcomes from Workshop: 19th July 2007Facilitated by Jenny Pope (GHD)

Participants:

Irene Clark

Myles Harmer

Diana Frylick

Jodie Hutton

Jenny Johnson

Sharon Davies

Helen Dullard

Gerard van Didden

Fiona Jordan

Karen Shelley (Friends of Biala)

Sandra Bentley (Mundaring RPA)

Paul Benson (Sawyers Valley RPA)

Apologies:Mike Davies (Stoneville RPA)

Margaret Fowler (Glen Forrest RPA)

Community Amenity

Community amenity is a general term often used to describe the potential impacts of adevelopment on the lifestyle of the local community. In the case of the proposedMundaring WTP, the issues considered under community amenity were focused upon‘nuisance’ factors such as: visual amenity, i.e. what the WTP looks like and whatimpacts it might have on the outlook from residents’ properties, including impacts ofplant lighting; noise, i.e. whether the noise generated by plant operations will affectlocal residents; and traffic, i.e. whether traffic entering and leaving the WTP site willcreate impacts in terms of noise, congestion or road safety.

It was noted during the scoring process that some participants would have preferred abroader definition of community amenity, to include issues such as hazardouschemical risks, vegetation clearing and heritage values. However, these matters areincluded under other criteria in the MCA process, and therefore were excluded fromthe discussion of community amenity so as to avoid double counting.

General Issues Raised by Participants

Site One

While some participants rated Site One as having low impact on community amenitybecause of the relatively few residences in the vicinity, others ranked it as having ahigh impact on community amenity because its selection would mean that the WTPwould be located on two different sites, thus increasing overall impact. The questionwas asked as to whether it the section of Mundaring Weir Road closest to the Weircould be realigned to permit the whole plant to be located on Site One, but WaterCorporation advised that this was not possible and that the chlorination plant and


sludge drying beds would still have to be located at one of the three other sites if SiteOne were selected. Water Corporation also advised that the chlorination plant itself, iflocated remotely, would have minimal amenity impacts, as it would occupy a very smallspace, involve no noisy equipment, and require only minimal lighting at night. Thesludge drying beds would not be visible from residential areas and would not beodorous.

DEC Land

There are a number of residences in the vicinity of the DEC Land, including Tree TopsBed and Breakfast and properties along Mundaring weir Road. It was suggested bysome participants that a WTP located at this site would be more visible fromresidences along Mundaring Weir Road than a WTP on O’Connor Hill because theywould look down on the plant, but this view was not shared by all. Noise and lightingimpacts on these residents were also of concern.

Pine Plantation

The Pine Plantation site was considered by all participants to represent a low impacton community amenity because of its distance from any residences Therefore it rankedhighly overall.

O’Connor Site

Concerns were raised that the WTP would be visible from Mundaring Weir Road andfrom residences along Mundaring Weir Road and to the north and east of the proposedsite, including those along McCallum Road. It was pointed out that the map providedby Water Corporation excludes residential areas to the northeast of the proposed site.Noise and lighting impacts on these residents were also of concern.

General Issues

Several participants made the point that the increase in traffic movements associatedwith the WTP operation would be negligible in the context of existing traffic alongMundaring Weir Road, although this view was not shared by all. The point was alsomade that sound carries long distances in Mundaring, particularly at night, andtherefore Water Corporation was requested to adopt best practice noise reductiontechnology when designing the plant. The issue of construction noise was alsodiscussed, and Water Corporation indicated that construction could take approximatelytwo years.

Final Rankings: 1 is the best outcome and 6 is the worst outcome (most impact).The majority of participants reached consensus and where consensus was notreached the alternative rankings are included.

Seven participants ranked:

1. Pine Plantation

2. Site 1 and Pine Plantation

3. Site 1 and DEC land


4. Site 1 and O’Connor Site

5. DEC land

6. O’Connor Site

Three participants ranked:

1. O’Connor Site

2. Pine Plantation

3. DEC land




One participant ranked:

1. Pine Plantation




5. O’Connor Site

6. DEC land


Appendix E

Recreation and Tourism Values


Mundaring Water Treatment Plant Site Selection MultiCriteria Analysis (MCA): Recreation and Tourism ScoringProcessOutcomes from Workshop: 19th July 2007Facilitated by Megan Dilly (GHD)

Participants:

Anne Brake (National Trust)Gwen Plunkett (Bibbulmun Track Foundation)Jamie Ridley (Parks and Visitor Services, DEC)Jens Jorgenson (Mundaring Tourism Association)Jill Frances (Perth Hills National Parks Centre)Kirk Kitchen (Recreation Field Coordinator, Shire of Mundaring)Megan Griffiths (Executive Manager Community Service, Shire of Mundaring)Stefan de Haan (DEC)Stuart Harrison (DEC Trails and Recreation) – could only attend part of workshop

ApologiesJenniVerne Taylor (Munda Biddi Foundation)

Recreational and Tourism Values

This criterion relates to the potential impacts of the WTP on the users and managers ofrecreational and tourism facilities in the area, in terms of convenience and enjoyment.Some of these impacts may be direct, for example, some walking trails pass throughproposed WTP sites or their buffer zones and these may have to be rerouted. Otherimpacts are less tangible, for example, the WTP may impact on the visual amenity ofpeople using facilities such as picnic areas, museums, or scenic lookout.


Site One

Site One is currently used for tourism and recreation significantly more than any of theother sites and has the highest value of all the sites for this aspect. A WTP would notbe appropriate in this area, and would impact on recreation and aesthetic values,particularly through visual impact.

DEC Land

This site is very close to the tourism and recreation area. Light and noise from the sitemay impact on the campground, Kookaburra Outdoor Cinema, the Treetops Bed andBreakfast and walktrails. This site would reduce the reason why people use thesefacilities, ie. the natural setting.

This site has the potential to be used as a tourism and educational facility in the future,through use of the heritage values of the site. A WTP at this site would result in loss of


potential future tourism. However, this site would enable provision of educationalfacilities associated with the plant, as it is in close proximity to the other educationalareas.

A number of walk and cycle trails may be impacted by this site. The heritage trailsshould not be moved because they follow the old railline and have historic meaning inthe location that they are in.

Pine Plantation

The traffic to this site would need to go along Allen Rd, which passes the visitorscentre and campground. Last year there was 45 000 visitors to the centre, this is ahigh use tourism and recreation centre. Traffic on Allen Rd could be a risk issue, withpedestrian movements across Allen Rd and may also be a nuisance issue.

The site itself is remote from the tourism and recreation facilities. However, this maymean that potential educational value of the site is lost.

It was raised that if an alternative access to this site could be found (ie by upgradingFirewood Rd and existing firebreaks) that this site would have few tourism andrecreational impacts.

The Bibbulmun Track shouldn’t be impacted by this site, and if there was any indirectimpacts (ie. visual) then the track can be moved in this area.

O’Connor Site

This site was generally seen to have a low potential impact on tourism and recreationas it is removed from the main tourism and recreation areas. Potential issues with thissite are aesthetic impacts (particularly visual impact) on users of the walking andcycling trails that pass along the western boundary of the site and on users ofMundaring Weir Rd.

Final Rankings (1 is the best outcome and 4 is the worst outcome (mostimpact)). Consensus was reached on all rankings.

1 O’Connor Site

2 Pine Plantation (with access as Allen Rd)

3 DEC Land

4 Site 1 (all possible combinations of Chlorine store)

However, if an alternative access could be found to Pine Plantation that wouldmean that traffic would not need to use Allen Rd (ie. if Firewood Rd and existingfirebreaks were upgraded to allow access to the Pine Plantation from the north)then the rankings would be as follows:

1 Pine Plantation (with access north of Allen Rd)

2 O’Connor Site


3 DEC Land

4 Site 1 (all possible combinations of Chlorine store)

The participants were in all agreement that Site 1 and DEC Land would have the worstimpact on tourism and recreational values and that the impacts from these sites issignificantly greater than any impacts from the Pine Plantation and the O’Connor Site.


Appendix F

Biodiversity and Potential for LandDegradation


Mundaring Water Treatment Plant Site Selection MultiCriteria Analysis (MCA): Biodiversity Scoring Processand Potential for Land Degradation Scoring ProcessOutcomes from Workshop: 23rd July 2007Facilitated by Megan Dilly (GHD)

Participants:

Beth Schultz (WA Forest Alliance)Fiona JordanJael Johnson (EARTH)Paul BensonSandra Bentley (Reserve ProtectionGroup)Steve McKiernan (ConservationCouncil)Toni Burbidge (Shire of Mundaring)

Apologies:Allan and Anne Pilgrim (Eastern HillsWildflower Society)Paul Van de Beecke (Shire ofMundaring Environmental AdvisoryGroup)Penny Hussey (Helena RiverCatchment Group)

Biodiversity

This criterion relates to the biodiversity values of the site, including the clearingfootprint required. Biodiversity is the variety of all life forms, the different plants,animals and microorganisms, the genes they contain, and the ecosystems of whichthey form a part.

In the case of the proposed Mundaring WTP, the issues to be considered under thesustainability criterion of impact on biodiversity are:

» The overall ecological value of the sites;

» The conservation significance of the sites;

» The clearing footprint required at the sites (the amount of native vegetation that willbe cleared);

» The value of the site for native flora; specifically the diversity of native flora at thesite, any significant flora species or vegetation types at the site and the health of thevegetation at the site;

» The value of the site for native fauna; specifically, the potential of the site tosupport native species, including significant species; the value of the site as faunahabitat and the use of the site as a habitat linkage.

Potential for Land Degradation

Land degradation is a serious environmental problem in Western Australia and can bedefined as the decline in condition or quality of the land as a consequence of humanactivities. For the purposes of the project this criterion addresses potential issuesassociated with the construction and operation of the WTP as they relate to erosionand soil degradation at the site and the potential to introduce diseases and weeds tothe general area. When considering this criterion it is necessary to assess the current


status of the land and the extent of degradation of the site, ie what soil degradationalready occurs and whether weeds and diseases (specifically dieback) already occur atthe site. It also assesses the sensitivity of the site to land degradation factors.


Site One

Site One was generally seen as degraded, though the value of the granite outcrops(that are present on the site) as fauna habitat was discussed. One of the biggestissues with this site is the requirement for an additional site for the drying beds andchlorine store. It was generally agreed that impacting on two sites was worse thanretaining the impact at one site. Potential impacts on the Helena River adjacent to thesite was also discussed, particularly any potential downstream impacts.

One advantage of this site is that a new pipeline will not be required to transfer waternorth of the weir (though one will be required for transfer to the sludge drying beds)and so impacts on vegetation along the existing main conduit will be reduced.

DEC Land

The area required for the plant is degraded and impacts here were seen to be minor.Participants raised the issue of the number of weedy species within and adjacent to thesite and discussed whether Water Corporation could undertake weed control as amanagement measure, which could increase the value of the areas adjacent to theplant. The value of this site as fauna habitat was raised. The majority of the sitecontains introduced species, which have some value for fauna; however, it wasaccepted that the introduced species are of less value to fauna than native vegetation.

Pine Plantation

The majority of the plant at this site would be in an area currently under pines, which isan area of relatively low value; however, it was recognised that pines do providehabitat for some native species, including blackcockatoos.

While the impact on biodiversity from the site itself may be low, the potential impactsfrom the required access to this site, particularly due to clearing of vegetation adjacentto the existing road and potential fragmentation of habitat was seen as an issue.

This site is within a Dieback Disease Risk Area and is within areas of nativevegetation, away from the high areas of human activity near the weir. The WTP at thissite was seen to be potentially causing impacts in an area where other impacts arelimited, this may increase the risk of land degradation, particularly the spread ofDieback and weeds.

O’Connor Site

There was a very strong consensus that this was the worst site in terms of potentialimpacts on biodiversity and land degradation. While the site has been previously


disturbed, it still has the highest biodiversity values of all the sites and will requiresubstantially more clearing of native vegetation than the other sites.

General Issues

The requirement for drying beds was an issue that was discussed and the point raisedthat if native vegetation is required for clearing for drying beds then it should beassessed to determine if another form of drying (such as mechanical drying) would bepreferable.

Final Rankings (1 is the best outcome and 6 is the worst outcome (mostimpact)). The majority of participants reached consensus and where consensuswas not reached the alternative rankings are included.

Biodiversity:

Four participantsranked:

1. DEC Land

2. Site One and DECLand

3. Pine Plantation

4. Site One and PinePlantation

5. Site One andO’Connor Site

6. O’Connor Site

Two participantsranked:

7. DEC Land

8. Pine Plantation

9. Site One andDEC Land

10. Site One andPine Plantation


12. O’Connor Site


7. Site One and DECLand

8. DEC Land

9. Pine Plantation

10. Site One and PinePlantation


12. O’Connor Site

Potential for Land Degradation

Six participants ranked:

13. DEC Land

14. Site One and DEC Land

15. Site One and Pine Plantation

16. Pine Plantation

17. Site One and O’Connor Site

18. O’Connor Site


19. Site One and DEC Land

20. DEC Land

21. Site One and Pine Plantation

22. Site One and O’Connor Site

23. Pine Plantation

24. O’Connor Site


Appendix G

Site Flexibility and Operability


Mundaring Water Treatment Plant Site Selection MultiCriteria Analysis (MCA): Site Flexibility and OperabilityScoring ProcessOutcomes from Workshop: 6th July 2007Facilitated by Jenny Pope (GHD)

Participants (all Water Corporation):

Andrew Crawford A/Senior Engineer, Strategy

George Malita Principal Engineer, Water Treatment

Peter Marchesani Program Manager, Capital Investment

Peter Mcallister Scheme Operations Manager, Scheme Operations

Andrew Mcdonough Senior Engineer, Surface Water

Anthony Mclaughlin IWSS Development Consultant, Asset Mgmt

Kingsley Quartermaine Treatment Operations Manager, Treatment Operations

Brian Robertson Program Manager, Capital Investment

Noel Winsor Principal Engineer, Water Treatment

Chris Dolley Program Manager, Capital Investment

Noel Turner Service Delivery Representative, Asset Delivery

Apologies:

Peter Minson Surface Water Operations Manager, Surface WaterOperations

Site Flexibility and Operability

This criterion relates to the planning, design and operational phases of the project froman engineering perspective. Site flexibility is a reflection of the range of optionsavailable to planners and designers in configuring the WTP on the site, as well as thesite’s potential to accommodate changing circumstances and new technologies.Operability relates to the ease with which a plant located on the site could be operatedand maintained on a daily basis, which is related to its layout and design.

The first aim of the workshop held on 6th July was to clarify the dimensions of siteflexibility and operability. The workshop participants identified the following factors tobe considered: expandability, topography, plant succinctness, accessibility, security,ease of fire management, proximity to raw water and ease of discharge of backwashwater.



Site One

Site 1 is constrained in size, which means that if this site were selected for the WTPthe chlorination plant and sludge drying beds would have to be located at one of theother three sites. For this reason, Site 1 scored poorly for expandability and plantsuccinctness. The site is also steep and rocky and therefore also scored poorly fortopography. Relatedly, some concern was raised about the Mundaring Weir Roadbeing steep and winding close to the Weir, and hence Site 1 scored poorly foraccessibility. However, being adjacent to Mundaring Weir, it was considered to be thebest site in terms of proximity to raw water and also for ease of disposal of filterbackwash water and fire management. No distinction was made in the discussionbetween the three possible configurations of Site 1.

DEC Land

The DEC land site was considered to be acceptable with respect to each of the siteflexibility and operability considerations, and was not found to be either the best or theworst site in any case.

Pine Plantation

The Pine Plantation site generally rated poorly for site flexibility and operability, andwas not considered to be the best site with respect to any of the considerations. Itsrelative remoteness, distance from Mundaring Weir Road, and location within theReservoir Protection Zone meant that the Pine Plantation was considered the worstperformer with respect to security, fire management and ease of discharge ofbackwash water. It scored reasonably well in terms of expandability and topography,and was considered adequate in terms of accessibility, proximity to raw water.

O’Connor Site

The O’Connor site was considered the best option in terms of expandability andaccessibility, due to its location adjacent to Mundaring Weir Road but away fromcommunity infrastructure. It scored poorly for proximity to raw water, and wasconsidered adequate with respect to all other issues.

General Issues

Several site flexibility and operability considerations served to distinguish one site fromthe other three. For example, all sites were considered equally adequate fortopography and plant succinctness except Site 1, and all sites were considered equallyadequate from the perspective of security except the Pine Plantation. Despite this,however, the engineers attending the workshop felt that all sites could be acceptablefrom an engineering perspective, but that costs (outside the scope of this discussion)would vary accordingly.


Final Rankings: 1 is the best outcome and 3 is the worst outcome (most impact).

Rankings were determined after the workshop by analysing the comments made byparticipants, focusing on the identification of sites clearly better or clearly worse thanothers, and the ranking of the sites where this was undertaken, with respect to eachsite flexibility and operability consideration. Based on this analysis, it was consideredthat DEC land and the O’Connor site ranked equally well, followed by Site 1, and thePine Plantation, which was considered the worst performing site for this criterion. Thefinal rankings are therefore:

1. DEC land and O’Connor site

2. Site 1 (all configurations)

3. Pine Plantation

61/16611

Appendix H

Waterways and Water Cycles

61/16611

Mundaring Water Treatment Plant Site Selection MultiCriteria Analysis (MCA): Waterways and WatercyclesScoring ProcessOutcomes from meeting with Frances Guest (Department of Water (DoW) andMegan Dilly (GHD) on the 6th of August 2007 and meeting with Steven McKiernan(Water Policy Officer – Conservation Council) and Megan Dilly (GHD) on the 8th

of August 2007.

Waterways and Watercycles

This criterion relates to the potential impacts of the WTP on waterways, which for thispurpose includes river and stream systems and wetlands, and on watercycles. Thedesirable environmental goal for this criterion is that waterways (including minordrainage lines) and associated riparian vegetation, should be protected anddevelopment should be excluded from the buffer area of a waterway. The desirableenvironmental goal for water cycles is that the project would minimise impacts andalterations to the natural watercycles. This incorporates consideration of catchmentissues. Protection of catchments occurs through restricting certain types ofdevelopment in catchment areas and by implementing management measures forpractices that carry a risk of contaminating surface and groundwater.

General Issues Raised

Site One

Site One is of concern due to its proximity to the Helena River and the two tributariesthat run through the site into the Helena River. Generally, DoW would require a 100 mbuffer between the WTP and the waterways. This site was not seen as preferable dueto these issues. However, it was noted that the river has been highly disturbed in thisarea.

DEC Land

All impacts on the northern tributary should be avoided (as indicated in the preliminaryplant layout) and management measures will be required for drainage at the site,particularly for the small tributary of Bending Gully that originates in the site.

Pine Plantation

This site is major problem and DoW would not support at all. Site is within theReservoir Protection Zone (RPZ) and no public access should be allowed in this area.

O’Connor Site

The major issue at this site is related to the requirement for clearing and the potentialto increase salinity in the catchment. Clearing controls are present within thecatchment and clearing is generally not permitted (see Department of Water, HelenaRiver Salinity Situation Statement, Water Resource Technical Series: Report No. WRT34, May 2007.). If clearing was permitted there would be the requirement torevegetate equivalent amount of native vegetation within the catchment. The issue

61/16611

with clearing requirements for the drying beds was raised by Steven McKiernan andwould need to be addressed further if this site was chosen.

Final Rankings by DoW and GHD (1 is the best outcome and 6 is the worstoutcome (most impact)).

1 DEC Land

2 O’Connor Site

3 Site One and DEC Land

4 Site One and O’Connor Site

5 Pine Plantation

6 Site One and Pine Plantation

61/16611

Appendix I

Community Weighting WorkshopMaterial

61/16611

Summary of Sustainability CriteriaBiodiversity

» Issues include:

– The overall ecological value of the sites

– The conservation significance of the sites

– The clearing footprint required at the sites (the amount of native vegetation thatwill be cleared)

– The value of the site for native flora; specifically the diversity of native flora at thesite, any significant flora species or vegetation types at the site and the health ofthe vegetation at the site

– The value of the site for native fauna; specifically, the potential of the site tosupport native species, including significant species; the value of the site asfauna habitat and the use of the site as a habitat linkage


» Means the decline in condition or quality of the land as a consequence of humanactivities

» Includes consideration of:

– Soil erosion and degradation

– Potential to introduce diseases and weeds

– Current state of site

– Sensitivity of site to land degradation

Embedded (embodied) energy of assets» Refers to the amount of energy that has been used to produce the materials from

which the WTP will be constructed

» Embedded (also known as embodied) energy correlates to greenhouse gasemissions generated by production of the materials

» Does not include energy required for operation of the sites

– This will be nearly identical for all sites and thus does not add any value to thecomparison of sites

61/16611



– Rivers

– Streams

– Wetlands

– Minor drainage lines

– Associated riparian vegetation

– Catchment protection


» Various chemicals will be used

» Chlorine is most hazardous chemical and therefore the basis of the preliminary riskassessment

» Issue relates to potential health and safety impacts on:

– Residents

– Visitors to parks, museums and other facilities

– Users of walking trails

Community amenity» In this case, community amenity can be equated to ‘nuisance’, i.e. impacts on

lifestyle of local people in their homes


– Visual impacts

– Plant operational noise

– Traffic noise

– Traffic nuisance (congestion etc)

– Light spillage

61/16611


» In this case, community amenity can be equated to ‘nuisance’, i.e. impacts onlifestyle of local people in their homes


– Visual impacts

– Plant operational noise

– Traffic noise

– Traffic nuisance (congestion etc)

– Light spillage

European heritage

» Includes both positive and negative dimensions:

– Disruption of artefacts or areas where it is expected that artefacts could be found

– Disruption of locations of historical significance

– Requirement for relocation of buildings or artefacts (undesirable from aninterpretive perspective)

– Positive opportunities for WTP to be interpreted as next chapter in the story ofMundaring Weir and the Goldfields water supply (recognised by National Trust).

Indigenous heritage

» Includes:

– Archeological issues (presence or potential for presence of artefacts)

– Ethnographic issues (cultural significance of areas, stories etc)

– Related to Native Title

61/16611


» Relates to planning, design and operational phases of the project from anengineering perspective

» Issues include:

– Expandability

– Topography

– Plant size

– Accessibility

– Security

– Ease of fire management

– Proximity to raw water

– Ease of discharge of backwash water

Costs

» Includes:

– Capital costs

– Operating costs

Weighting ProcessWeighting means asking: ‘how important is each criterion with respect to each othercriterion in selecting the best site for the Mundaring WTP?’

Weighting process facilitated by Ross Lantzke (Independent MCA specialist fromMurdoch University)

Process to follow:

25. Review suite of criteria

26. If you think any are not relevant at all, weight this criterion 0

27. Identify the least important remaining criterion and weight it 1 (may have more thanone criterion weighted as 1)

28. Weight each remaining criterion in relation to the criterion (criteria) that scored 1 bycompleting Table 2

» i.e. identify how many times more important it is

» Use Table 1 as guidance

29. Record the number on your weighting worksheet for future reference

61/16611

Table 1 Guidance for Allocating Weights

Noimportancewhatsoever

Lowestimportance

A little moreimportant

Moderatelymoreimportant

Stronglymoreimportant

Verystronglymoreimportant

Extremelymoreimportant

0 1 2 3 4 5 6 7 8 9 10


Table 2 Community Forum Weighting Worksheet (Number )

Criteria My weight

Biodiversity





Community amenity


European heritage

Indigenous heritage


Cost


Appendix J

Water Corporation WeightingWorkshop Attendees


Water Corporation Weighting Workshop

Mundaring Water Treatment Plant Site Selection

Water Corporation Weighting Workshop

24th July 2007 23 pm

Attendees

Ken Walter (Manager Special Duties)

Bob Humphries (Sustainability Manager)

Natalie Reilly (Manager Sustainability Strategy)

Louisa Kinnear (Communication Officer)

Keith Cadee (General Manager Water Technologies)

Steven Mackenzie (Communication Officer)

Graham Cargeeg (General Manager Asset Management)

Miles Dracup (Environmental Strategy & Policy Manager)

Neil Formosa (Financial Analyst)

Robert Ng (Supervising Engineer)

Apologies

Lloyd Werner (Manager Pricing & Evaluation)

Julia Krsnik (Senior Environmental Officer)

Mark Leathersich (Principal Engineer)

Peter Marchesani (Program Manager)


Appendix K

Community and WaterCorporation Weights


Community Weights

Original Weights Criteria \ Serial # 2 3 4 5 12 14 15 16 18 19 20 21 22 23 26 27 28 37 40 41 42 43 45 47 48 49 50 53 54 55 56 57 59 60 64 65 67 691. Biodiversity 6 8 10 8 8 10 5 8 10 10 10 0 5 10 9 10 5 4 9 6 0 1 4 2 4 10 10 10 10 10 9 7 10 8 1 10 5 42. Potential for

land degradation 5 4 10 9 6 9 5 6 9 9 10 0 4 5 9 10 4 2 9 8 0 2 3 2 2 8 8 10 10 10 9 7 10 6 1 10 6 1

3. Embedded energy 2 3 1 1 6 2 1 2 6 1 2 0 2 0 7 10 1 2 7 1 0 1 0 0 1 6 6 9 6 8 5 0 10 1 1 10 0 24. Waterways and

water cycles 5 8 8 9 4 8 5 4 10 8 5 0 4 7 10 10 2 5 8 5 1 3 4 2 5 1 1 5 5 5 4 7 10 7 1 10 10 6

5. Hazardous chemical risks 4 2 10 8 10 3 2 2 10 10 9 0 9 10 10 10 6 1 10 6 0 0 7 1 8 10 10 7 4 4 10 6 10 5 1 10 1 16. Community amenity 3 2 10 8 8 9 4 4 9 6 5 0 8 10 8 10 2 10 10 9 0 3 2 3 10 10 10 10 10 10 10 6 10 8 1 10 2 107. Recreation and

tourism values 3 2 2 2 6 9 4 3 9 4 2 0 2 8 6 1 6 10 7 5 1 6 3 4 6 1 1 0 2 0 1 0 1 4 1 10 8 10

8. European heritage 3 2 2 1 2 1 4 2 10 4 5 0 5 2 1 1 1.5 2 5 4 5 6 7 3 1 1 1 0 2 2 1 1 1 3 1 10 5 29. Indigenous heritage 5 5 8 1 2 6 4 2 10 9 8 0 6 2 1 10 1.5 1 10 4 0 6 6 3 0 8 8 1 2 2 2 1 10 3 1 10 1 110. Site flexibility & operability 2 1 3 3 10 2 4 4 8 4 3 0 3 1 10 10 4 2 9 4 1 3 3 1 6 1 1 1 1 1 2 2 10 2 1 10 4 711. Cost 2 3 0 0 4 8 2 4 8 8 8 10 0 2 1 10 0 8 5 1 0 3 0 1 9 8 8 1 0 0 7 0 10 0 1 10 8 7

Totals 40 40 64 50 66 67 40 41 99 73 67 10 48 57 72 92 33 47 89 53 8 34 39 22 52 64 64 54 52 52 60 37 92 47 11 110 50 51

Normalized Weights1. Biodiversity 0.15 0.20 0.16 0.16 0.12 0.15 0.13 0.20 0.10 0.14 0.15 0.00 0.10 0.18 0.13 0.11 0.15 0.09 0.10 0.11 0.00 0.03 0.10 0.09 0.08 0.16 0.16 0.19 0.19 0.19 0.15 0.19 0.11 0.17 0.09 0.09 0.10 0.082. Potential for

land degradation 0.13 0.10 0.16 0.18 0.09 0.13 0.13 0.15 0.09 0.12 0.15 0.00 0.08 0.09 0.13 0.11 0.12 0.04 0.10 0.15 0.00 0.06 0.08 0.09 0.04 0.13 0.13 0.19 0.19 0.19 0.15 0.19 0.11 0.13 0.09 0.09 0.12 0.02

3. Embedded energy 0.05 0.08 0.02 0.02 0.09 0.03 0.03 0.05 0.06 0.01 0.03 0.00 0.04 0.00 0.10 0.11 0.03 0.04 0.08 0.02 0.00 0.03 0.00 0.00 0.02 0.09 0.09 0.17 0.12 0.15 0.08 0.00 0.11 0.02 0.09 0.09 0.00 0.044. Waterways and

water cycles 0.13 0.20 0.13 0.18 0.06 0.12 0.13 0.10 0.10 0.11 0.07 0.00 0.08 0.12 0.14 0.11 0.06 0.11 0.09 0.09 0.13 0.09 0.10 0.09 0.10 0.02 0.02 0.09 0.10 0.10 0.07 0.19 0.11 0.15 0.09 0.09 0.20 0.12

5. Hazardous chemical risks 0.10 0.05 0.16 0.16 0.15 0.04 0.05 0.05 0.10 0.14 0.13 0.00 0.19 0.18 0.14 0.11 0.18 0.02 0.11 0.11 0.00 0.00 0.18 0.05 0.15 0.16 0.16 0.13 0.08 0.08 0.17 0.16 0.11 0.11 0.09 0.09 0.02 0.026. Community amenity 0.08 0.05 0.16 0.16 0.12 0.13 0.10 0.10 0.09 0.08 0.07 0.00 0.17 0.18 0.11 0.11 0.06 0.21 0.11 0.17 0.00 0.09 0.05 0.14 0.19 0.16 0.16 0.19 0.19 0.19 0.17 0.16 0.11 0.17 0.09 0.09 0.04 0.207. Recreation and

tourism values 0.08 0.05 0.03 0.04 0.09 0.13 0.10 0.07 0.09 0.05 0.03 0.00 0.04 0.14 0.08 0.01 0.18 0.21 0.08 0.09 0.13 0.18 0.08 0.18 0.12 0.02 0.02 0.00 0.04 0.00 0.02 0.00 0.01 0.09 0.09 0.09 0.16 0.20

8. European heritage 0.08 0.05 0.03 0.02 0.03 0.01 0.10 0.05 0.10 0.05 0.07 0.00 0.10 0.04 0.01 0.01 0.05 0.04 0.06 0.08 0.63 0.18 0.18 0.14 0.02 0.02 0.02 0.00 0.04 0.04 0.02 0.03 0.01 0.06 0.09 0.09 0.10 0.049. Indigenous heritage 0.13 0.13 0.13 0.02 0.03 0.09 0.10 0.05 0.10 0.12 0.12 0.00 0.13 0.04 0.01 0.11 0.05 0.02 0.11 0.08 0.00 0.18 0.15 0.14 0.00 0.13 0.13 0.02 0.04 0.04 0.03 0.03 0.11 0.06 0.09 0.09 0.02 0.0210. Site flexibility & operability 0.05 0.03 0.05 0.06 0.15 0.03 0.10 0.10 0.08 0.05 0.04 0.00 0.06 0.02 0.14 0.11 0.12 0.04 0.10 0.08 0.13 0.09 0.08 0.05 0.12 0.02 0.02 0.02 0.02 0.02 0.03 0.05 0.11 0.04 0.09 0.09 0.08 0.1411. Cost 0.05 0.08 0.00 0.00 0.06 0.12 0.05 0.10 0.08 0.11 0.12 1.00 0.00 0.04 0.01 0.11 0.00 0.17 0.06 0.02 0.00 0.09 0.00 0.05 0.17 0.13 0.13 0.02 0.00 0.00 0.12 0.00 0.11 0.00 0.09 0.09 0.16 0.14

Totals 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00


Water Corporation Weights

Original WeightsCriteria \ Serial # 1 2 3 4 5 6 7 8 91. Biodiversity 10 10 6 9 10 8 2 4 72. Potential for land degradation 6 9 6 7 9 7 2 3 63. Embedded energy 6 6 6 6 5 1 5 5 54. Waterways and water cycles 10 10 6 6 4 5 3 4 75. Hazardous chemical risks 1 6 6 8 0 2 3 1 36. Community amenity 6 8 4 4 3 6 1 4 67. Recreation and tourism values 6 8 1 1 5 6 1 1 18. European heritage 8 8 1 4 5 2 1 1 29. Indigenous heritage 8 8 4 4 0 3 1 2 210. Site flexibility & operability 10 6 2 8 7 8 5 4 511. Cost 6 4 2 9 8 10 5 6 3

Totals 77 83 44 66 56 58 29 35 47

Normalized Weights1. Biodiversity 0.13 0.12 0.14 0.14 0.18 0.14 0.07 0.11 0.152. Potential for land degradation 0.08 0.11 0.14 0.11 0.16 0.12 0.07 0.09 0.133. Embedded energy 0.08 0.07 0.14 0.09 0.09 0.02 0.17 0.14 0.114. Waterways and water cycles 0.13 0.12 0.14 0.09 0.07 0.09 0.10 0.11 0.155. Hazardous chemical risks 0.01 0.07 0.14 0.12 0.00 0.03 0.10 0.03 0.066. Community amenity 0.08 0.10 0.09 0.06 0.05 0.10 0.03 0.11 0.137. Recreation and tourism values 0.08 0.10 0.02 0.02 0.09 0.10 0.03 0.03 0.028. European heritage 0.10 0.10 0.02 0.06 0.09 0.03 0.03 0.03 0.049. Indigenous heritage 0.10 0.10 0.09 0.06 0.00 0.05 0.03 0.06 0.0410. Site flexibility & operability 0.13 0.07 0.05 0.12 0.13 0.14 0.17 0.11 0.1111. Cost 0.08 0.05 0.05 0.14 0.14 0.17 0.17 0.17 0.06

Totals 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00


Appendix L

Worked Examples of Normalisation andConcordance Analysis


Example of the how the Weights are Normalised

Criteria Weights asallocated by anindividual

Calculation:Divide the Weight bythe Sum of all theWeights

Normalised Value(shown here as aPercent)

Criterion A 5 5/29.5 17

Criterion B 2 2/29.5 7

Criterion C 7 7/29.5 24

Criterion D 1 1/29.5 3

Criterion E 2 2/29.5 7

Criterion F 2 2/29.5 7

Criterion G 5 5/29.5 17

Criterion H 3 3/29.5 10

Criterion I 2.5 2.5/29.5 8

Sum of all the Weights 29.5 100%

This example is a modified version of the slide shown at the Community Forum on 30 July 2007. Therewere 11 criteria in the weighting process associated with the siting of the Water Treatment Plant. Theabove is only an illustration of the process of normalisation.


Example of the Calculation of one Pairwise Concordance Analysis value

Criteria + Weighting + Scores Matrix Compare Option O with Option 1DOptions

Criteria Wt D P O 1D 1P 1O

Costor

Benefit

Is the O scorebetter

than the IDscore?

If yes, give it theweight for that

criterion1 0.125 1 3 6 2 4 5 c No2 0.111 1 4 6 2 3 5 c No3 0.052 3 5 6 2 4 1 c No4 0.104 1 5 2 3 6 4 c Yes 0.1045 0.103 3 1 2 3 1 2 c Yes 0.1036 0.122 5 1 6 3 2 4 c No7 0.079 3 2 1 4 4 4 c Yes 0.0798 0.070 3 2 1 4 4 4 c Yes 0.0709 0.074 3 2 4 1 1 1 c No

10 0.068 1 4 1 3 3 3 c Yes 0.06811 0.091 4 6 5 2 3 1 c No

Sum theweights = 0.425

In the above table 'c' means the higher scores are a cost, thatis higher the score the less worth it has and the lower scorethe better it is.

= 42.5%of potential

weights

The value of 0.425 is then inserted into the Concordance Matrix as highlighted in the table below, whereoption O, the column title at the top of the table, is compared to option 1D, the row title on the left side.When option 1D is compared to option O the result is the reciprocal, namely 0.575, as illustrated in bolditalics.

Concordance MatrixD P O 1D 1P 1O

D 0.448 0.286 0.390 0.390 0.4429P 0.552 0.412 0.626 0.448 0.389O 0.714 0.588 0.575 0.678 0.627

1D 0.610 0.374 0.425 0.371 0.3921P 0.610 0.552 0.322 0.630 0.3931O 0.558 0.611 0.373 0.608 0.607

Sum 3.043 2.574 1.818 2.829 2.494 2.242

The higher the sum the better the result. The above process is repeated until all options are compared toall other options.

The outcome of the Concordance Matrix is often put through a further step to create a Dominance Matrix,however the sum of the concordance analysis gives a good indication of the rank of the options. In thiscase D is the preferred option, 1D is next, P third, 1P fourth & 1O fifth. O is the least preferred option.

water corporation mundaring water treatment plant site selection

Documents