Document [original]

Challenges in Sustainability |2018 |Volume 6 |Issue 1 |Pages 1–19

DOI: 10.12924/cis2018.06010001

ISSN: 2297–6477

Challenges in

Sustainability

Research Article

A Multi-Criteria Approach for Assessing the Sustainability of

Small-Scale Cooking and Sanitation Technologies

Ariane Krause1*, Johann K¨

oppel2

1Center for Technology and Society, Technische Universit¨

at Berlin, Berlin, Germany

2Environmental Assessment & Planning Research Group, Technische Universit¨

at Berlin, Berlin, Germany

* Corresponding author: E-Mail: [email protected]; Tel.: +49 1793274543

Submitted: 29 August 2017 |In revised form: 27 March 2018 |Accepted: 5 April 2018 |

Published: 11 June 2018

Abstract:

To reduce the consumption of firewood for cooking and to realise recycling-driven soil fertility

management, three projects in Northwest Tanzania aim to provide the local smallholder community with

cooking and sanitation alternatives. The present study proposes an integrated approach to assess

the sustainability of the small-scale cooking and sanitation technologies. Based on the multi-criteria

decision support approach (MC(D)A), we developed a decision-specific, locally adapted, and participatory

assessment tool: the Multi-Criteria Technology Assessment (MCTA). Pre-testing of the tailored tool was set

up with representatives of Tanzanian and German partners of case study projects. From a methodological

perspective, we conclude that the MCTA uses a set of relevant criteria to realise a transparent and replicable

computational Excel-tool. The combination of MC(D)A for structuring the assessment with analytical

methods, such as Material Flow Analysis, for describing the performance of alternatives is a promising path

for designing integrated approaches to sustainability assessments of technologies. Pre-testing of the tool

served as a proof-of-concept for the general design of the method. Future applications and adjustments

of the MCTA require the inclusion of end-users, a reasonable and participatory reduction of criteria, and

an increase of feedback loops and group discussions between participants and the facilitator to support a

common learning about the technologies and thorough understanding of the perspectives of participants.

Keywords:

Biomass stoves; decision support; development of appropriate technologies; ecological

sanitation; energy-sanitation-agriculture nexus; multi-objective evaluation; sustainability assessment;

technology assessment

Preliminary Note

Additional information is presented in the Appendix and

the Supplements; Figures and Tables are indicated with

Figure A.x and Table S.x. Additional notes are indi-

cated in the text (e.g. [x]) and provided at the end of

the manuscript, alongside a list of non-standard abbre-

viations and the references. We use italic letters to

emphasize and indicate discipline-specific definitions,

established terms upon initial use, terms applied from

languages other than English, and quotations covering

more than two lines.

2018 by the authors; licensee Librello, Switzerland. This open access article was published

under a Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/).

librello

1. Introduction to the Context

1.1. Towards a Sustainable Anthroposphere

There is broad consensus that—from a global perspective—

the excessive use of natural resources by human beings

(in particular the resource consumption of the richer coun-

tries) has manifold consequences [

]. The anthropogenic

overexploitation of natural resources results, for example, in

(i) shortages of resources, such as oil, coal, sand, or rock

phosphate [

], (ii) climate change [

], or (iii) decrease of soil

fertility [

]. By various means, we are crossing the planetary

boundaries [

]. With respect to biogeochemical flows and

genetic diversity, the planetary boundaries are stressed to

such an extent that we risk non-reversible changes of the

environment and even the habitability of the earth. Agricul-

ture, in particular—respectively the production and use of

synthetic fertilizers, pesticides, or seeds—contributes to this

risk [

]. How, therefore, can we change the situation? Or,

more specifically: How can we meet the (human) demand

for food and energy without overexploiting our planet?

Recognized recommendations for a sustainable use of

resources include, for example [

–

]: (i) establishing cir-

cular economies; (ii) using effective, efficient, adapted, and

affordable technologies; and (iii) utilizing locally available

resources for soil fertility management. Measures to estab-

lish circular economies comprise inter alia (i) the concept

of bioenergy, hence the provision and use of energy from

biomass [

], respectively renewable resources or organic

waste materials, and (ii) the implementation of ecological

sanitation (EcoSan). With the latter, various material flows

are collected separately, treated specifically, and, if possi-

ble, ultimately recycled to agricultural land [

]. Both topics,

bioenergy and EcoSan, are well-studied and have already

been brought into practical applications.

1.2. A Sustainable Approach to the

Energy-Sanitation-Agriculture Nexus in Tanzania

Against this backdrop, and given that over 50% of global

food products are produced by small-scale farmers [

the sustainability potentials of smallholder communities in

the Global South is of major relevance. In the Northwest

of Tanzania (TZ), a representative region for Sub-Saharan

Africa (SSA), the main energy carrier is firewood, mostly

used for cooking on a three-stone fire, whilst pit latrines are

the most common sanitation approach [

]. This pattern,

alongside a steady population growth, increasingly stresses

natural resources such as arable land, aquifers or other

water sources, and forests [15,16].

To counteract local deforestation and soil nutrient deple-

tion, two Tanzanian farmers initiatives have started three

projects located in Karagwe, Kagera region, together with

German partners and donors [

]. The main objective of the

technological development cooperation was to contribute

to sustainable community development by adapting and

introducing small-scale and locally adequate cooking and

sanitation technologies [

]. At the same time, a circular re-

source economy shall be implemented to improve local soils

and food production. Therefore, residues from cooking and

sanitation are used to increase the recycling of nutrients

and the recovery of carbon for restoring soil humus.

Previous work covers how these locally adapted tech-

nologies in smallholder households contribute to environ-

mental health by reducing deforestation, climate gas emis-

sions, and nutrient leaching to aquifers [

]. We further

showed that using residues from cooking and sanitation

could stimulate the productivity of smallholder farming by

improving the acidic and nutrient-depleted local soil [

Subsequently, we widened our perspective to evaluate the

technologies studied according to multiple sustainability

dimensions.

1.3. Assessing Sustainability with a Participatory

Multi-Criteria Approach

The challenge of any sustainability assessment [

] is to

translate defined and basically agreed upon principles into

an operational model [

]. However, even though measur-

able criteria can be applied to operationalize sustainability,

it is individuals or societies who evaluate the plausibility

and credibility of relevant assessments. If the process in-

volves multiple actors, differing opinions may also cause

trade-offs between stakeholders. Multi-Criteria Decision

Analysis (MCDA) attempts to address this challenge [

–

]. In the first instance, any assessment of alternatives

against multiple criteria, irrespective of who participates in

the analysis, can generally be called a Multi-Criteria Anal-

ysis (MCA) [

]. Moreover, MCDA indicates a specific

MCA, addressing participatory decision-making processes

[

]. Both approaches, MCA and MCDA, are applied in

planning, decision, and evaluation processes and to assess

likely consequences of decisions [

]. In the following, we

use the abbreviation MC(D)A when referring to characteris-

tics of both approaches.

According to [

], the MC(D)A is generally suitable

for decisions related to sustainable development because

the method is largely transparent, participatory, and inter-

disciplinary. Furthermore, the MC(D)A is often applied in

environmental decision making [

] and strategic plan-

ning for natural resources management [

]. While the

number of MC(D)A applications has clearly grown over the

last two decades, the method was mainly applied in sci-

entific communities in the Global North [

]. Among the

view examples of MC(D)A-studies from SSA are (i) com-

parisons of different sanitation approaches in Uganda [

]

and in Burkina Faso [35], (ii) an assessment of cook stoves

in TZ [

], and (iii) an evaluation of appropriate farming

techniques in TZ [38].

With this study, we want to contribute to advancing

the applicability of MC(D)A in the regional context of SSA

as well as in the topical context of technology-driven sus-

tainable community development in the Global South and

Global North alike.

1.4. Research Objectives & Questions

Based on the MC(D)A methodology, our aim is (i) to de-

velop and propose an integrated method for a participatory

sustainability assessment, (ii) to translate into an applicable

tool, and (iii) to test the tool with selected stakeholders (pre-

testing). Our specific focus addresses small-scale cooking

and sanitation technologies alongside the use of residues

in smallholder farming, and on the example of Tanzanian

case studies.

The specific research objectives are: (i) to identify a

set of criteria and applicable methods within the framework

of MC(D)A; (ii) to develop and propose a handy tool for

an ex-ante, participatory, and multi-perspective technology

assessment; (iii) to conduct a pre-testing of the tool and

assess the sustainability of selected technologies.

The research questions are as follows: (RQ1) What are

the most relevant criteria and applicable methods for as-

sessing sustainability of small-scale cooking and sanitation

technologies? (RQ2) How did the stakeholders involved in

pre-testing rated the technologies at hand with regard to

relevant criteria and to overall sustainability? (RQ3) How

proved specific stakeholder groups perceptions? The RQ1

thereby refers specifically to the methodological develop-

ment of the tool whilst RQ2 and RQ3 refer to pre-testing of

the tool.

2. Used Methodology

Firstly, in Section 2.1, we introduce the study area, the

case study projects, and the technologies to be analysed.

In Section 2.2, we elaborate on our specific approach to

develop an integrated MC(D)A-method and tool for a par-

ticipatory sustainability assessment of small-scale cooking

and sanitation technologies. We introduce the main concept

of MC(D)A and explain the process how we have adapted

the methodology for our specific context and converted into

a tool. In Section 2.3, we shortly present the process ap-

plied for pre-testing the tool. Further details on methods,

computational works (including equations), and the tool are

provided in the Appendix.

2.1. Introduction of the Study Area & Case Studies

The study area is the Karagwe district, in the Kagera region

(lat. 01

◦

33’ S; long. 31

◦

07’ E; alt. 1500-1600 m.a.s.l.).

Kagera forms a part of the Lake Victoria basin and is lo-

cated about 1,500 km away from the Tanzanian capital Dar

Es Salaam. The regional economy is dominated by small-

holder agriculture with about 90% of households trading

agricultural products grown on their farms [14].

Three Karagwe projects serve as our case studies: (i)

Biogas Support for Tanzania (BiogaST), which aims to im-

plement small-scale biogas digesters to use bioenergy from

harvest and kitchen residues for cooking; (ii) Efficient Cook-

ing in Tanzania (EfCoiTa), which disseminates advanced de-

signs of improved cook stoves (ICS), such as microgasifier

cook stoves; and (iii) Carbonization and Sanitation (CaSa),

which works with EcoSan, including a urine-diverting dry

toilet (UDDT), thermal sanitation of faeces, and composting

of human excreta mixed with other organic wastes includ-

ing biochar. The latter material is a residue from cooking

with microgasifiers. Cooking and sanitation are connected

through the use of residues for soil fertility management,

which makes the energy-sanitation-agriculture nexus even

more interesting to study. In prior publications, we intro-

duced the case studies [

] and described the technologies

[

]. All technologies shall be implemented and used in

both, households and institutions (e.g. in schools or hos-

pitals). The local non-governmental organisation (NGO)

Mavuno Project - Improvement for Community Relief and

Services (MAVUNO) facilitates the BiogaST- and CaSa-

projects. The Programme for Community Habitat Environ-

mental Management (CHEMA) runs the EfCoiTa-project.

German project partners are Engineers Without Borders

(EWB) Berlin/Germany and Technische Universit

at (TU)

Berlin. Financing of the projects is provided by German

foundations BayWa (BiogaST) and Heidehof (CaSa and Ef-

CoiTa). After completing pilot studies, a strategy and a plan

for implementing technologies in the smallholder community

needs to be developed. Hence, the respective technolo-

gies are the subject of decisions at hand. Decision makers

are either the NGOs, implementing the technologies on a

project-basis, or single households, purchasing them, for

example, from local manufacturers or at local markets.

2.2. Designing an Integrated Tool for a Multi-Perspective

Technology Assessment

The method shall: (i) enable a systematic analysis;

(ii) consider

multiple perspectives related to sustainability;

(iii) involve

local communities and authorities in the assess-

ment; (iv) integrate available data; and (iv) be conducted

ahead of the technologies implementation in the community.

Against this backdrop, we chose to follow the systematic

concept of MC(D)A.

According to [

], MC(D)A consists of three pillars,

which are: (i) the presence of multiple criteria; (ii) the formal

approach with a set of analytical methods; and (iii) the

involvement of individuals or groups of individuals in the as-

sessment (for MCA) or in the decision process (for MCDA).

The MC(D)A further combines a range of methodologies to

use quantitative and qualitative data for (i) measuring sets

of criteria, (ii) considering consequences of decisions, and

for (iii) sequencing alternatives [

]. Applied MC(D)As often

combine (i) qualitative methods for problem structuring,

(ii) quantitative methods for problem analysis, and (iii) soft

methods for stakeholder participation [

]. Aggregation of

an assessment is possible from the level of different criteria

into an overall performance of alternatives and, likewise,

from the level of individual preferences into a common,

average preference [

] (For the interested reader, a sum-

mary of the fundamental terms used in MC(D)A can be

found in Table A.1). Based on a comprehensive literature

review, especially [

–

] inspired the design of our

MC(D)A approach. The dynamic process of developing the

specific design of our method and tool will be described

in the following paragraphs (with more details provided in

Appendix Section A.2).

(A) Framing the context:

Formulating the decision problem

involved learning about the context of the decision, such

as livelihood of the community, specific approaches of the

case studies, and ways of using technologies in daily life.

The main activities, therefore, included: participating in

projects; short- and long-term stays in Karagwe; teamwork

with project workers; reading project reports, governmental

reports, and non-governmental reports; and communicating

with scientists and practitioners in the region. Based on

this, we described the environment of the decision including

driving forces (Table 1) and motivations (Table 2) that repre-

sent the objectives of the projects initiators to develop and

implement ‘new’ technologies in the Karagwe smallholder

community.

In addition, two Process Flow Diagrams (PFD) foster

a better understanding of the technologies and possible

nutrient recycling approaches when interacting with people.

Both PFDs indicate how the technologies analysed are

integrated in on-farm resource management and how they

interact with the natural environment in terms of resource

consumption and emissions. One PFD is a more system-

atic illustration (Figure 1); the other PFD is a more pictorial

illustration (Figure A.1).

(B) Creating alternatives:

Alternatives to be analysed are

either (i) discrete technologies or (ii) a mix of technologies

[

]. In the present study, we decided to analyse discrete

technologies that are defined by the case studies (Table

3; cf. Tables A.3–A.5 for pictures and descriptions). Cur-

rently, the most common combination of technologies for

cooking and sanitation found in Karagwe smallholdings is

a three-stone fire and pit latrine [

]. The alternatives as-

sessed are further technologies that are locally available

(i.e. technically and commercially) for implementation. We

compared (i) technologies represented in the case study

projects, and (ii) other widespread alternatives. Regarding

the first point, cooking alternatives include: a system com-

posed of a biogas digester and a biogas burner, developed

and adapted to Karagwe conditions by the BiogaST-project;

and two types of ICSs promoted by the EfCoiTa-project,

such as the rocket stove and the microgasifier. MAVUNO

endorses implementation of biogas systems at households

through development cooperation projects whilst CHEMA

produces and disseminates rocket stoves and microgasi-

fiers at their workshop. Sanitation alternatives include to

ecological approaches, namely, “EcoSan” that refers to us-

ing an UDDT and “CaSa” that refers to using the UDDT in

combination with a sanitation oven for thermal treatment

of excreta pursuant to the practices of the CaSa-project.

MAVUNO is engaged in testing and promoting both ap-

proaches. Regarding the second point, charcoal stoves

(also called “charcoal burners”) are commonly available

in Karagwe on markets and from local stove sellers and

a combination of a flush toilet and septic tank can be ac-

cessed through local plumbers. The charcoal and biogas

alternative also include the charcoal production and biogas

digester. These processes provide the energy carrier used

in households. Rocket or microgasifier stoves, in compari-

son, make direct use of firewood or sawdust. The latter is an

available waste resource in many anthropogenic settings.

Table 1.

Definition of driving forces behind the technology development of case study projects. These definitions had

been pre-formulated by the planner and were then proofed and agreed on by participants during the design process.

Social driver Environmental driver Economical driver

•Provide new technologies to the community,

which have progressive reputation. •Reduce pressure on natural

forests by providing cooking technologies that

use less firewood or alternative fuels.

•React on social drivers through

increased agricultural productivity.

•Increase food security through

increased soil fertility and crop productivity.

•Reduce poverty through increased

farm income generation.

•Recycle nutrients recovered from residues

available after cooking or from sanitation facilities.

•Provide organic fertilizers to

replenish nutrients in exhausted local soils.

Table 2. Consented definition of the motivations behind the technology development of case study projects

Energy Sanitation Agriculture

Use available resources appropriately

to meet the energy demand of smallholder

farming households.

Improve the hygiene conditions in

farming households while promoting the recycling

of nutrients.

Sustain the quality and productivity

of the local soil in order to ensure food

sovereignty and farm income.

Figure 1. Systematic illustration of the analysed context featured as the “Resource Management Trifecta” [44].

Table 3.

Alternatives defined for the sustainability assessment of small-scale cooking and sanitation technologies that

are, technically and commercially, available to smallholders in Karagwe, Tanzania

Cooking alternatives Sanitation alternatives

Current state (not analysed) Three-stone fire Pit latrine

Alternatives analysed 1. Charcoal stove (incl. charcoal production) 1. EcoSan: with UDDT only

2. Rocket stove 2. CaSa: UDDT and sanitation oven for heat

treatment of solids3. Microgasifier

4. Biogas digester and biogas burner 3. Water toilet and septic tank

Non-common abbreviations: CaSa: project “Carbonization and Sanitation”; EcoSan: ecological sanitation; UDDT: urine-diverting dry toilet.

In order to operationalize sustainabil-

ity, relevant and feasible, thus appropriate, criteria need to

be identified to describe the multiple dimensions. According

to MC(D)A methodology, main-criteria describe the objec-

tives, or highest-level criteria, and sub-criteria describe

the attributes, or lowest-level criteria [

]. To identify

such criteria, we performed a comprehensive literature

review. Our focus was specifically on comparable work

conducted (i) for topics such as bioenergy, EcoSan, and

soil fertility or natural resource management, and (ii) in

a regional context of TZ, East-Africa, or SSA. We further

collaborated with practitioners, academic professionals,

and other experts in bioenergy and/or sanitation in an East-

African-context. Activities and methods applied include: (i)

semi-structured interviews conducted in TZ and Uganda;

(ii) group discussions based on the world caf

e method [

]

in TZ and Germany; and (iii) discussion with the coordinator

of BiogaST and CaSa projects (A. Bitakwate) to conclude a

final set of criteria.

(D) Collecting data:

To describe how the alternatives per-

formed against the criteria we needed to collect relevant

data. Most data used derive from prior studies related

to the case studies [

–

]. Additional data originate

from project documents of the pilot studies, judgements of

local expert, and literature as well as internet reviews. De-

scriptions provided to participants also include information

about data sources and the estimated quality of data (cf.

Table A.9).

(E) Analysing stakeholders and selecting participants:

Par-

ticipants of a MC(D)A can be split into two groups: (i) the

planner; and (ii) stakeholder representatives [

]. The plan-

ner facilitates the whole MC(D)A and, therefore, pre-selects

criteria, prepares the tool and descriptions of the alterna-

tives beforehand, and moderates the assessment process.

Stakeholder representatives participate, in particular, in

conducting the weighting of criteria and the scoring of alter-

natives. According to [

], relevant stakeholders that should

be considered when conducting a participatory MC(D)A

include: (i) funding agencies; (ii) governmental authorities;

(iii) specialists (e.g. stove technicians, EcoSan-experts, or

strategic advisors); (iv) (future) staff for operation and main-

tenance; (v) end-users of the technologies and/or of the

by-products; (vi) legislator and enforcement agencies; (vii)

research institutions; (viii) local, national and international

NGOs; and (ix) site residents (if any). In our case, local

smallholders are probably the stakeholders being most

affected by the technologies assessed.

(F) Preparing the assessment tool:

The following require-

ments had been defined for the assessment tool: (i) free

of charge; (ii) intuitive to use; (iii) transparent (e.g. that it

is possible to see and understand also the computational

model); and (iv) possible to use off-line for participants be-

cause internet access is not always reliable in Karagwe.

Unfortunately, most available commercial or free software

was not appropriate for several reasons. Most of them were

either (i) too cost-intensive, such as V

A, (ii) too com-

plex, such as PROMETHEE I because visualization is too

scientific, or (iii) working only online, such as Decisionarium

or Web-HIPRE.

As a consequence, we developed a hand-made and

Excel-based assessment tool. The tool is tailored to our

specific application and based on a template of [

]. In

total, the designed tool comprises sheets for (i) providing

information to participants, (ii) collecting judgments of par-

ticipants during weighting and scoring, and (iii) calculating,

summarizing, and visualizing results. The latter include so-

called Performance Matrices (PM) that summarize weights,

scores, and results for each participant and all alternatives.

To elicit individual preferences from participants, we se-

lected applicable methods for participatory weighting and

scoring from [

]. The weighting further comprises

ranking and rating of main- and sub-criteria. Ranking is

the ordering of criteria according to individually perceived

importance. Rating means assigning individual numeric

weights as points ranging from 0 to 100 to each criterion in

order to differentiate between the criteria.

2.3. Pre-Testing of the Tool

In order to test the developed method, we performed a

pre-testing of the tool with a selected group of participants.

Subjects of the pre-testing are the alternatives as described

above. However, the technologies currently most used are

not assessed in order to keep the total amount of alterna-

tives manageable.

Against the backdrop in which there is a significant lan-

guage gap between Tanzanian farmers and German par-

ticipants, who mostly do not speak Swahili, we considered

direct participation of farmers as not feasible. Furthermore,

and pursuant to [

], farmers may hesitate to freely ex-

press their thoughts if white researchers and/or donors are

present during the assessment. Overall, stakeholders par-

ticipating in pre-testing the MC(D)A tool include: (i) the

facilitating NGOs, represented by the management, project

coordinators, and other staff members; (ii) the coopera-

tion partners, represented by researchers, volunteers, and

donor agents. In total, 10 people committed their partic-

ipation and the number of participants in each group is,

respectively, six (four of MAVUNO and two of CHEMA) and

four (four scientists from TU; three members of EWB; and

one member of the advisory board from Heidehof founda-

tion; including double affiliations). Planner and moderator

was A. Krause. In the pre-testing, participating staff mem-

bers of local NGOs also act as representatives of the local

smallholders. (The tool itself should, however, support farm-

ers in decision-making within the smallholder community.)

Pre-testing followed a “9-steps-approach”, where the ac-

tual assessment of alternatives is conducted in a stepwise

and participatory procedure, including the following nine

steps:

Presenting: Introduction of the context of the assess-

ment, of the MC(D)A method, and of different activi-

ties for participants and the facilitator in the assess-

ment’s course.

Agreeing: Presentation of pre-formulated driving

forces and motivations for project initiations and re-

quest to comment, agree, or disagree.

Self-assessment: Participants disclose their role as

stakeholder defined as their personal role, power, in-

terest, and means of intervention in each of the case

study projects.

Weighting: Participants express their individually per-

ceived importance of criteria by (i) ranking and rating

main-criteria and (ii) simple rating of sub-criteria.

Knowledge-exchange: Presentation and discussion

of prior research work and, in particular, those data

used to formulate the descriptions of alternatives.

Scoring: Participants indicate the individually per-

ceived value of an alternative based on descriptions

provided by the planner and completed by individual

expertise.

Calculating: The planner (i) calculates weighted

scores of all sub- and main-criteria, (ii) deduces ag-

gregated overall results, and (iii) visualizes results.

Conclusion: Final presentation for sharing results with

all participants; if possible, including group discus-

sions and the possibility to adjust individual scorings.

Evaluation: To evaluate the process, a questionnaire

is given to participants for providing feedback and

criticism, and to formulate lessons learned.

Unfortunately, due to time constraints and because partic-

ipants had not been in one place, rather located in both

countries, Tanzania and Germany, discussion and adjust-

ments of scoring as part of step 8, was not possible. Plan-

ner/facilitator and participants communicated in English

and mainly via e-mail or by using Skype or phone calls on

demand, for example to clarify questions. Presentations

were shared as PDF-documents and via file hosting service.

(More details provided in Appendix Section A.3).

3. Results & Discussion

This section starts with a discussion of results as far as the

methodology development is concerned whilst addressing

RQ1 (Section 3.1.). Thereafter, we present and discuss

results of the designed tool’s pre-test, and refer to RQ2

and RQ3 (Section 3.2.). Finally, we reflect on the adapted

assessment process and tool (Section 3.3.).

3.1. Results from Adapting the MC(D)A Method for the

Specific Context

In order to systematically assess the sustainability of small-

scale cooking and sanitation technologies, we propose

a tool called the Multi-Criteria Technology Assessment

(MCTA). We designed the MCTA by adapting MC(D)A and

developing a tool, which is adequate for the sustainability

assessment of small-scale cooking and sanitation tech-

nologies. The process included activities as described in

Section 2.2. Results with respect to the identification of

most relevant criteria and applicable methods for assessing

the sustainability of small-scale cooking and sanitation

technologies (RQ1) are presented and discussed in the

following paragraph.

Selection of criteria:

To define the main-criteria considered

in the MCTA, we merged sustainability categories of [

]

with aspects from the Integrated Sustainable Waste Man-

agement approach of [

] that describes “the enabling en-

vironment of sustainability”. Finally, we chose the follow-

ing main-criteria as being most relevant: technological-

operational, environmental, health/hygiene, socio-cultural,

socio-economic/financial, and political/legal criteria. Pur-

suant to [

], we use a six-pointed so-called “sustainability

star” (Figure 2) to visualise how we define sustainability in

the MCTA. Each of the six dimensions represents one of

the six main-criteria and thus reflects an objective of sus-

tainability that is addressed in the analysis of cooking and

sanitation technologies. Furthermore, the categories “ac-

ceptability”, “affordability”, and “reliability” are assigned to the

main-criteria. Derived from Sustainable Development Goal

no. 7 of the United Nations, these indicate characteristics,

which feature a sustainable use of energy technologies [

Figure 2.

The sustainability star that was used to visualize

and summarize the six main-criteria considered in the MCTA.

The final selection of sub-criteria (Table A.8) constitutes

a synthesis of criteria from [

] (cf. Table A.6).

From a total of 84 sub-criteria considered in the MCTA, 80

and 75 are applied for assessing cooking and sanitation

technologies, respectively (Table A.7).

Overall, we consider the chosen criteria as significant

to systematically identify strengths and weaknesses of

technologies and to assess their sustainability. However,

criteria are not thoroughly independent from each other.

Some criteria describe different objectives even though

they are related to each other. For example, two of the

technological-operational criteria are “preferably high use

of locally available resources” and “preferably low use of

industrial resources”. Both sub-criteria are related to the

socio-economic/financial criterion “preferably low cost for

implementation” because the material use influences pro-

duction costs. Another reason for preferring a substantial

use of locally available resources is, however, a higher

degree of independence from international markets (and

main providers of industrial materials). To sum up, most

of the sub-criteria applied are intermediate or so-called

“means-to-an-end” objectives rather than fundamental ob-

jectives [

]. Nevertheless, according to [

] this approach

is not contradicting the MC(D)A concept.

Collection of data:

Overall, our study integrates quantita-

tive data of recent scientific findings and explorative prac-

titioners’ experiences with qualitative data collected from

literature, oral explanations by farmers, learning from partic-

ipating in the projects, and expert judgements. Quantitative

data of prior studies, where we employed Material Flow

Analysis (MFA) [

] also in combination with Soil Nutrient

Balancing (SNB) [

], was particularly important to describe

environmental attributes. The MFA identifies and quanti-

fies sources, sinks, directions, magnitudes, connections,

dependencies, and shifts of material flows between the

anthroposphere and the environment [

]. The SNB specif-

ically measures, calculates, and balances various input and

output flows of nutrients to and from agricultural land [52].

Against the backdrop that reaching environmental sus-

tainability requires a systematic reduction of the physical

degradation of nature [

], we consider MFA and SNB as

highly suitable methods to be integrated in a multi-criteria

sustainability assessment. By combining qualitative and

quantitative methods for data collection, the MCTA realises

an integrated approach as recommended by [24].

Data collection, however, was challenging. Available

and accessible data in the given context was not sufficient

to estimate and to describe performances of most technolo-

gies. In order to generate a profound basis for the MCTA,

we needed significant additional efforts and approximately

two years for conducting performance tests of the technolo-

gies, elaborating a field trial and MFA and SNB modelling.

Hence, describing the performance of the alternatives was

challenging and time-consuming, which resulted in certain

shortcomings of the pre-test.

Calculations:

The computational model applied in the MCTA

is set up through combining the Multi-Attribute Value The-

ory (MAVT) and Simple Additive Weighting (SAW) out of a

wide range of existing MC(D)A methods [

]. The MAVT

basically uses multiple attributes, for comparing alternatives

according to their value. SAW describes the mathemati-

cal approach to determine such values through systematic

aggregation.

Computations pursue the following process: Firstly, rank-

ing and rating of main-criteria is done by applying the

SWING-method [

]; sub-criteria are then simply rated.

From the numeric weights (i.e. points ranging from 0 to

100) given by participants during rating, the tool supports

the planner in determining the relative weights of sub- and

main-criteria (Equations A.1–A.5). Subsequently, scoring

is conducted for each sub-criterion and each alternative

separately. Participants therefore assign numeric scores in

points ranging from

−

10 to +10 (Table 4) that should reflect

their individually perceived performance of an alternative

and the associated value. In the MCTA, participants also

have the option to waive scoring, by indicating a * instead

of assigning points, if they consider available information

as insufficient and/or feel unsure about scoring. In offering

this *-option, we attempt to not force participants into an

arbitrary judgement.

Pursuant to the SAW, the assigned scores are firstly

weighted on the lowest level of single attributes. The tool,

therefore, determines the weighted score for each sub-

criterion as the product of (calculated) relative weight and

(assigned) score (Equation A.6). Computations are adapted

for the *-case as described in the Appendix (Equations

A.4 and A.6). Subsequently, plain addition determines the

weighted scores for each main-criterion (Equation A.7). To

deduce the overall value of an alternative, weighted scores

are then further aggregated to result in a final Sustainability

Index (SI). The SI is determined for each person (Equation

A.8) along with an average SI of all participants for each

alternative (Equation A.9). To evaluate significant differ-

ences between alternatives, value ranges characterised by

the mean value

standard error of the mean (SEM) are

compared pursuant to [54].

Table 4. The scoring system applied in the MCTA

Scoring points Performance of the

alternative

Value of the

alternative

−10 Extremely weak/poor Strongly unfavourable

−5Poor/fair; major

improvements needed Unfavourable

0 Good, ordinary Acceptable

5Very good; still needs

some improvement Favourable

10 Excellent Very favourable

* Not clear to me Not assessable

Presentation of results:

To understand the variety of as-

sessments of individual participants, the visualization of

results is an essential element of the MCTA. In total, the

tool provides the following visualizations, which are inspired

by [

]: (i) a colour-coded scatter plot showing the rela-

tive weights assigned to main-criteria; (ii) bar diagrams

ranking the alternatives with respect to the average SI; (iii)

colour-coded scatter plots showing the average SI for all

alternatives in one graph; (iv) colour-coded scatter plots

showing assessment results on the level of main-criteria;

(v) colour-coded bar diagrams showing the distribution of

the SI as the balance of positive and negative weighted

scores on the level of main-criteria; and (vi) a summary of

scoring and weighting results on the level of main-criteria.

The latter three types of visualizations summarize the voting

of all participants in one graph for each alternative. The

colour-coded scatter plots indicate assessment results of

different stakeholder groups alongside the mean of all par-

ticipants in different colours. The first and fourth graphs

are designed by using a net diagram and by arranging the

axes that represent main-criteria according to the sustain-

ability star. The sixth graph visualises connections between

the perceived importance of a certain dimension of sus-

tainability and the assigned performance of an alternative.

The former is indicated by numeric weights assigned to

respective main-criteria plotted on the x-axis and the latter

by numeric scores that an alternative receives for certain

main-criteria plotted on the y-axis. The area, where certain

main-criteria score negatively (numeric score

0) while, at

the same time, these main-criteria are considered important

(numeric weight

50), is marked as “red area”. Results

are presented for all main-criteria (colour-coded symbols)

and for all participants (number of signs) in one graph for

each alternative.

We acknowledge that the computational approach of

the MCTA is rather simple, however, in accordance with

basic principles of MC(D)A modelling [

]. For example,

on the one hand, the final single index makes alternatives

directly commensurable [ibid.]. On the other hand, the ap-

plied methods, MAVT and SAW, belong to compensatory

MC(D)A techniques [ibid.]. This means that a very bad scor-

ing for one attribute can be compensated by a very good

scoring for another and vice versa. Furthermore, instead of

determining specific value functions for each participant, we

assumed linearity. The main reason was to avoid overstrain-

ing participants. Furthermore, the statistical approach is

rather simple but viable. As featured by [

], who demand a

“more transparent, simple, and easily accessible participa-

tory modelling paradigm and process”, we thus consider the

MCTA as a valid methodological extension of the MC(D)A.

Finally, visualisations support communication of results to

and between participants. Graphs are designed so that

they can effectively help (i) to reveal issues that need to be

addressed before considering a technology suitable for sus-

tainable community development, (ii) to identify the specific

fields where further improvements of the technology are

needed, and, finally, (iii) to avoid potential project ruptures

in the future.

3.2. Selected Results from Pre-Testing the Tool

In order to perceive the feasibility of the developed method

and tool, we performed a pre-testing of the MCTA. In

this section we firstly show how participants weighted the

different main-criteria describing the overall sustainability.

Subsequently, we present selected results of the sustain-

ability assessment of cooking and sanitation alternatives.

We provide additional graphical visualizations alongside

plot data in the Supplements.

Weighting of main-criteria:

In the course of MCTA’s pre-test,

the choice group of participants assigned, on average, the

highest weights to the environmental main-criterion; the

political-legal objective is perceived as least important (Fig-

ure 3). Second in level of importance are the remaining

four main-criteria with no significant differences (Figure S.2).

We further find a high variance in the distribution of weights

of all participants. Weightings among participants are most

divergent for socio-cultural criteria (∆

max

min ≈

20%) and least

divergent for political/legal criteria (∆max

min ≈11%).

Comparing weights assigned by participants, there is

a tendency that Tanzanians consider health/hygiene and

environmental criteria more important in comparison to

Germans. German representatives, in contrast, perceive

technological-operational and socio-cultural criteria as more

important compared to Tanzanians.

We explain the finding that stakeholders consider the

environment to be the most important dimension of sus-

tainability because the environmental main-criterion include

many sub-criteria related to agriculture, and is consequently

perceived as highly important for smallholders. On the con-

trary, according to [

], residents in the vicinity of Arusha/TZ

do not consider environment as an important objective when

choosing renewable energy technologies for electrification.

Only one interviewee out of about 40 respondents, who

were mainly farmers, addressed an environmental criterion

as an influencing factor to purchase a solar home system

[ibid.]. Respondents instead mentioned health and hygiene

as the most important objectives for their families [ibid.].

Additionally, [

] state that environmental criteria are

less relevant for residents when assessing sanitation sys-

tems in a peri-urban area in Burkina Faso. In order to

perceive a certain sanitation system as appropriate, fac-

tors such as costs and service quality are most relevant

for participants [ibid.]. Nevertheless, we understand our

finding that environmental criteria are most relevant, by the

fact that (i) our study area is a rural area farther away from

urban settlements compared to the study areas of [

] or

[

] and (ii) that many participants in our study are closely

related to environmental settings (e.g. farmers with off-farm

income, agricultural technicians, environmental scientists,

and environmental activists).

In other comparative studies of small-scale bioenergy

technologies from SSA, however, highest weights are as-

signed to the technological-operational criterion. For exam-

ple, in an assessment of ICSs available on the Tanzanian

market, criteria related to the construction or operation of

stoves ranked highest, such as manufacturability, durability,

or portability [

]. Furthermore, stakeholders participating

in an assessment of a biogas system in Bahir Dar, a city

in northwest Ethiopia, consider technological-operational

criteria as most important dimension of sustainability [23].

In our study, lowest priority is assigned to the politi-

cal/legal main-criterion. We reason that there is little gov-

ernmental support in TZ and especially in the study region,

which is recognized by stakeholders, and thus influences

their judgements. National programs for promoting renew-

able energies and supporting access to such technologies

especially in rural areas of TZ include the Rural Energy

Fund, the Rural Energy Agency, and the Tanzanian Domes-

tic Biogas Program (TDBP) [

]. However, to our knowl-

edge, these programs have not yet reached Karagwe. For

example, TBDP has not yet chosen or appointed implemen-

tation partners in the region. According to [

], Tanzanian

standards for cooking technologies are only available for

charcoal stoves whilst existing standards are not enforced.

Furthermore, governmental initiatives, intended to support

bioenergy implementation, are not coordinated and, thus,

largely inefficient [ibid.]. With respect to sanitation, [

]

also criticizes a strong lack of legislation in the sector of

non-sewer sanitation service in TZ.

The finding that only average importance is given to

socio-cultural aspects is comparable to studies of [

] and

[

] and in the study of [

] only minor importance is as-

signed to socio-cultural criteria.

Figure 3.

Colour-coded scatter plot of the relative weights of the main-criteria (in%) of Tanzanian (red) and German

(blue) participants alongside the mean of all participants (grey). [The sum of all six main-criteria is 100%.]

Sustainability assessment of cooking alternatives:

When

pre-testing the MCTA, all analysed alternatives reach aver-

age SI-values between 0 and +1 (Figure 4). Hence, the

choice group of participants perceives cooking alternatives

overall as “acceptable” and attributes a “good, ordinary”

performance to respective technologies (cf. Table 4). Mean

values of the four alternatives, however, do not significantly

differ from each other, as overlapping error bars in Figure 4

indicate [54].

Regarding the six dimensions considered to assess sus-

tainability, all stoves receive positive (weighted score

assessment results mainly for health/hygiene, socio-cultural

or technological-operational (except biogas system) functions

(Figure 5). Negative (weighted score

0) assessment results

refer to socio-economic/financial (for charcoal and biogas

alternatives) and environmental (for charcoal only) criteria.

When crosschecking the assessment results with liter-

ature, we consider our findings for cooking alternatives as

highly comparable to [

]. In both studies, microgasifiers

and rocket stoves receive positive scores because of their

potential to reduce fuel consumption and greenhouse gas

(GHG) emissions.

Interestingly, Tanzanian participants in particular as-

signed positive scores to the biogas system and for sub-

criteria on GHG-emissions despite the fact that the de-

scription clearly stated that GHG-emissions would increase

compared to cooking on a three-stone fire. We reason that

Tanzanian stakeholders are aware of the minor influence of

TZ on global GHG-emissions, compared to Germany, other

European countries, or the United States. We further argue

that there is a strong and positive pre-conception of the

biogas system as an “environmental-friendly technology”.

Regarding the usability of cooking technologies, all stake-

holders participating in the MCTA agreed that the biogas

system performs the worst of all alternatives analysed. Ma-

jor problems identified are thereby (i) water requirements for

a sound operation of the digester and (ii) lack of robustness

of the system towards changes in climate conditions and

user abuse. Also with respect to costs (including costs for

implementation, operation, and maintenance), the biogas

technology is assessed as the most disadvantageous cook-

ing alternative. To underline scorings of costs, investment

costs of locally available alternatives in Karagwe are as

follows: costs for implementing a biogas system are ap-

proximately

1,200 , which makes possessing a biogas

digester a substantial investment for smallholders. In com-

parison, acquisition costs of charcoal stoves or ICSs range

from 2 to about

25 (cf. Table A.5). Furthermore, [

]

identify high installation and maintenance costs as key bar-

riers for biogas implementation in TZ, despite the fact that

households are willing to adopt the technology. Hence, [

]

conclude that implementing biogas in rural areas requires

financial support by national programs through subsidies

and/or loans.

Figure 4.

The average sustainability index (SI) of cooking alternatives analysed, presented as mean value of all

participants and in ranked order. The SI-values range from

−

10 to +10 (cf. Table 4). Grey bars indicate the standard

error of the mean (SEM).

Considering the perceptions of stakeholder groups, we

find a very high variance in individual valuations of par-

ticipants. On the example of the rocket stove, weighted

scores of main criteria range from

−

5 (“unfavourable”) to

+7 (“clearly favourable”) between participants (Figure 5).

With respect to the SI, most Tanzanian representatives

assess charcoal burners, rocket stoves, and microgasifiers

negatively and below the average SI (Figure S.3). German

participants, however, assess these technologies with pos-

itive and above-average valuations. A reverse tendency,

however, is depicted for the biogas systems: judgements

of German participants are rather negative and below the

average SI, whilst Tanzanians assess the technology pos-

itively and above the average SI. Most prominently, staff

members of CHEMA ascribe the highest positive points

to rocket and microgasifier stoves, which are the cooking

technologies that CHEMA disseminates. Likewise, staff

members of MAVUNO rate ICSs with rather negative scores

compared to the biogas system, which is the technology

MAVUNO promotes in Karagwe.

Sustainability assessment of sanitation alternatives:

Pre-testing the MCTA, all alternatives reach average SI-

values in the range of about

−

2 to +2 (Figure 6). Both

ecological approaches to sanitation are rated significantly

better compared to the septic alternative. There is, however,

no significant difference in the assessment of EcoSan and

CaSa because respective error bars in Figure 6 overlap [

Overall, the performance of EcoSan and CaSa approaches

is assessed as “good or ordinary”, which means that the

choice group of participants values both approaches as

“acceptable” to “slightly favourable” alternatives. On the con-

trary, participants assess the septic alternative as “slightly

unfavourable” alternative.

Regarding the six sustainability dimensions, EcoSan

and CaSa receive positive assessment results mainly

for technological-operational, environmental, and socio-

economic/financial functions (Figure 7). Negative as-

sessment results refer, in particular, to the political/legal

criterion. Shortcomings of the septic alternative mainly

refer to the environmental, health/hygiene, and socio-

economic/financial criteria.

Crosschecking our results with literature, the observa-

tion that a clearly better socio-economic/financial perfor-

mance is assigned to the ecological approaches compared

to the septic alternatives, is in line with [

]. For the case

of Burkina Faso, [

] report that in particular low- and

middle-income households perceive the UDDT as more

financially appropriate compared to septic tanks. Strategic

thinking, which includes considering operational costs as an

important factor, influences households’ judgements [

In contrast, high-income households consider the septic

system as more appropriate compared to the UDDT, be-

cause for this social group factors such as service quality

are more important than costs [ibid.]. Farmers in Karagwe

belong, however, mainly to low- and middle-income house-

hold groups [

]. (Table A.5 summarizes costs of sanitation

alternatives.)

The pre-test further indicated certain doubts about the

approaches, in particular on the sub-criteria “safe working

conditions” and “safety in operation through avoiding risks

of infection to users”. In contrast, literature commonly pro-

motes EcoSan as a safe sanitation alternative because of

the multi-barrier approach applied [

]. To our knowl-

edge, there were, however, no instances of sickness during

the course of the CaSa pilot project that ran more than four

years. Further investigations might follow-up this significant

observation (cf. Section 4.2.). In total, the performance

of ecological alternatives against the health/hygiene main-

criterion is neutral. Our finding ultimately supports integrating

health/hygiene as criterion when assessing sanitation tech-

nologies. This helps to identify areas of conflict, as shown

in our case. Health/hygiene criteria should, therefore, not

be excluded from analyses as it was, for example, the case

in studies by [56] or [60].

Furthermore, we find that socio-cultural performances

assessed for EcoSan and CaSa alternatives are compa-

rable to assessment results for the septic system. We

attribute this finding to the practical demonstration of UDDT-

technology in the region. On the contrary, [

] identify

UDDT as less socio-culturally accepted compared to a sep-

tic system and, therefore, request more alternatives and

their advantages and disadvantages before assessing the

technologies. This was exactly the case in the CaSa pilot

project’s course.

Perceptions of stakeholder groups with respect to the

average SI assigned to sanitation alternatives (Figure

S.4) reveals as well high variations in assessment results.

The EcoSan alternative, for example, weighted scores of

main criteria range from

−

7.5 (“clearly unfavourable”) to

+6.5 (“clearly favourable”) between participants (Figure 5).

Nonetheless, the characteristic patterns of stakeholders’

valuations differ here. For both ecological alternatives, Tan-

zanian stakeholders assess negatively and below the av-

erage SI, whilst German participants rate the alternatives

clearly positive and above the average SI. In contrast, most

German participants assess the septic alternative as clearly

more negative and below the average SI compared to their

Tanzanian fellows. Only for the septic system, the assess-

ments of all participants are rather homogenous and less

differentiated compared to other sanitation technologies. In

addition, the septic alternative receives an average neg-

ative SI from all except one participant. Most prominent

is, however, that one German participant, who has been

strongly involved in designing, developing, and testing the

CaSa-approach, assigns the highest positive scores to that

sanitation alternative.

3.3. Reflection and Critique of the Assessment

Methodology and Process

In this section we firstly summarize experiences, insights,

and lessons learned, that derived from the participatory

pre-test of the MCTA regarding the assessment process

and tool. Next, we discuss certain shortcomings of the

methodology. Specific adjustments we recommend for fu-

ture applications of the MCTA are summarized in Section

4.2.

In the first instance, the participatory design of the

method we developed is a direct response to [

], who

stress the need to develop more participatory elements

at all levels of MC(D)A to make the method applicable in

real-world practices. The relevant problem addressed is

the ex-ante sustainability assessment of small-scale cook-

ing and sanitation technologies in the exemplary context of

smallholder communities in Karagwe, TZ.

Figure 6.

The overall sustainability index (SI) of sanitation alternatives analysed, presented as mean value of all

participants and in ranked order. The SI-values range from

−

10 to +10 (cf. Table 4). Grey bars indicate the standard

error of the mean (SEM).

Figure 5.

Colour-coded scatter plot for the four cooking alternatives analysed (charcoal stove, rocket stove, microgasifier,

biogas system) indicating the assessment results of Tanzanian (red) and German participants (blue) alongside the mean of

all participants (grey). Results are shown on the level of the six main-criteria. Values range from

−

10 to +10 (cf. Table 4).

Figure 7.

Colour-coded scatter plot for the three sanitation alternatives analysed (EcoSan, CaSa, and septic system)

indicating the assessment results of Tanzanian (red) and German participants (blue) alongside the mean of all participants

(grey). Results are shown on the level of the six main-criteria. Values range from −10 to +10 (cf. Table 4).

We further learned that the visualisations of results, as

integrated part of the MCTA-tool, are highly supportive to

shed light on individual perspectives and valuations of stake-

holders for the assessed alternatives. The visualisations

help to make subjectivity in scoring results transparent.

Here, we agree with [

] who emphasises that showing

stakeholders each other’s preferences is rather one of the

most relevant results of conducting an MC(D)A. The visu-

alisations further allow for a target-oriented improvement

to overcome shortcomings of the technologies. Due to the

low sample size and the high variance we observed in

the tool’s pre-test, some visualization are not discussed

thoroughly in this article. They are, however, presented in

the Supplements (cf. Figures S.5–S.8).

As other benefits of the MCTA, we concluded that the

assessment process promotes collaborative learning and a

better understanding of the technologies assessed. Final

feedback revealed that five out of six participants articulated

that the MCTA supported them in forming their own opinion

about the technologies (Figure S.9). Hence, our method

reacts on [

] who promote “embedded learning”, that is

integrating recent findings and sustainability assessment.

This integration also constitutes an important part of the

analytical tool we developed. Information that is provided to

participants during the assessment process and describe

various impacts of the technologies assessed, originate in-

ter alia from recent action research conducted in Karagwe.

Despite these benefits, we further experienced the chal-

lenge of “balancing three tensions: (i) maintaining scientific

credibility, (ii) assuring practical saliency, and (iii) legitimis-

ing the process to multiple participants” [

]. In particular,

we identify certain limitations referring to (I) the choice of

alternatives and participants, (II) the realised means of par-

ticipation, (III) time consumption, communication effort and

complexity, and (IV) gender balance.

(I) Firstly, we acknowledge that technologies mostly ap-

plied for the time being, which are three-stone fire and pit

latrine, are not among the alternatives assessed. One could

further criticise the fact that farmers were not involved in

the assessment even though they represent the party most

concerned and that the technologies assessed were devel-

oped to improve livelihoods of smallholders. Reasons for

conducting a pre-test of the MCTA-tool with local repre-

sentatives limited to NGOs’ staff members are disclosed

in Section 2.3. Our proceeding is supported by [

] who

also recommend establishing intermediaries between re-

search institutions and farmers when evaluating agricultural

projects. Culture and social standing of research and/or

NGOs can inhibit reciprocal communication between re-

searchers and farmers.

(II) We acknowledge that the MCTA process is not par-

ticipatory throughout as intended. This is a consequence

of additional empirical data generated and that we faced

a conflict between the scope of the work and the available

time budget when actually starting the pre-test. Hence, we

had to adapt the 9-steps-approach. Initially we wanted to

include more feedback loops during the process, for ex-

ample, in order to discuss experiences after scoring, and,

thereafter, offering participants the opportunity to adapt

their valuations. Moreover, the assessment was conducted

separately, meaning that Tanzanian participants were in TZ

and German participants in Germany. This fact results, in

limited insights into the individual perceptions and, thus,

inhibits joint understanding and learning of participants.

(III) We understand that high time consumption is of-

ten reported as a problem when conducting a participa-

tory MC(D)A [

]. Nonetheless, in the final feedback,

most participants judged the scope of the MCTA, defined

by time budget, details, number of criteria, clarity of the

task, amount of information, etc., as good and to cope

with (Figure S.9). Moreover, extensive communication (via

e-mail and Skype) between the planner and individual par-

ticipants was required to identify and solve possible misun-

derstandings. For example, participants scored alternatives

significantly different although the performance of these al-

ternatives was described as strongly comparable. Differing

backgrounds of participants might explain this, as well as

the problems they faced in understanding all (scientific and

non-scientific) terms used in the descriptions. Participants

might also simply not agree with the description provided,

due to contrasting personal experiences and knowledge.

However, because of the decentralized process we applied,

it was not possible to directly address these observations,

for example in individual or group discussions.

(IV) Lastly, we acknowledge that we pre-tested the

MCTA-tool with in total two females and eight males; both

women white and with academic background. Hence, we

failed to reach a gender balance and, in particular, to in-

clude female representatives of Tanzanian partners. We

attribute this shortcoming (i) to choosing English as work-

ing language, and (ii) to a general dominance of men at

the level of project coordination and/or management in the

participating NGO communities.

With regards to differences in stakeholders’ perceptions,

our findings are further in line with [

], who state that ap-

plying a MC(D)A-approach does not create an objective,

unbiased, and value-free analysis. In a nutshell: individual

perspectives and subjective judgements are “normal”. It is

hence a methodological challenge to make these differences

visible instead of neglecting them. For example, we found

that assessments of participants in the pre-test seem substan-

tially biased towards the technology they promote. Visualizing

these differences expediently is only the first step. The lack

of group discussion (as part of the step 8; cf. Section 2.3.)

is a clear shortcoming. For example, we have no specific

insights into why differences exist and which arguments

underlie the variations observed. Interpretation of results

(by participants and the authors) is thus largely based on

guesswork.

According to the final feedback, two thirds of participants

believe that the MCTA is also applicable within their com-

munity if further adapted (cf. Section 4.3). In particular,

the tool can help (i) to clarify existent doubts and expected

benefits, (ii) to stimulate discussions, and (iii) to compare

varying assessments of projects facilitating NGOs and local

communities.

4. Conclusions & Recommendations

In this section, we firstly summarize overall conclusions

from the work presented with respect to developing the

MCTA (Section 4.1.) and the pre-test (Section 4.2). We

close the article with a summary of recommendations to

adapt the tool for future real-world applications in Karagwe

or elsewhere (Section 4.3).

4.1. Scientific Relevance of the Methodological Work

With respect to the objective to develop an integrated

method for a participatory sustainability assessment, we

conclude the following strengths and benefits of the MCTA

method and tool:

•

Inclusion of a set of relevant criteria for assessing

sustainability of small-scale cooking and sanitation

technologies, which can also serve as a catalogue

from which specific criteria for a given context can be

chosen.

•

Adequate organization of available data that describe

various effects of technology implementation in re-

lation to sustainability and to a particular objective-

driven decision, such as sustainable community de-

velopment.

•

Realization of a transparent computational approach

with a combination of applicable methods and a repli-

cable Excel-tool.

•

In-depth exploration of different alternatives with re-

spect to their sustainability and in an iterative effort

and moderated process.

•

Instructive presentation of the current understanding

of the analysed technologies by aggregating judge-

ments of individual participants into a ranking of alter-

natives.

•

Target-oriented mapping of individual perceptions in

order to effectively track conflicting stakeholder opin-

ions and to react to conflicts at an early stage, for ex-

ample, before a decision for implementation is made.

•

Target-oriented identification of the most relevant as-

pects where improvements of the analysed technolo-

gies are still needed.

From a methodological perspective, combining the MC(D)A

with analytical methods is a viable integrated approach to

sustainability assessment. On one hand, MC(D)A helps

to structure the assessment. On the other, the integration

of results from analytical studies, such as MFA, SNB, or

laboratory analysis, allows for illustrating the performance

of alternatives assessed through semi-quantitative objec-

tives of sustainability, and particularly the environmental

dimension.

Finally, the developed method provides hands-on sup-

port through a decent set of criteria and a basic assessment

tool. Basic skills in Excel are the main requirement for using

the MCTA-tool. Nevertheless, conducting an MCTA is an

extensive process that demands considerable time budgets

and a strong commitment from all participants.

To summarize, we conclude that the MCTA is an appro-

priate and applicable method to analyse potential effects

and interrelations of different cooking and sanitation tech-

nologies on household levels. The method also indicates

contradictions with respect to different sustainability goals.

In addition, the MCTA enhances transparency about individ-

ual preferences, values, doubts, and objectives among all

participants, respectively stakeholders.

4.2. Practical Relevance of Pre-Test Results

Even though the number of individuals participating in the

pre-test of the MCTA was rather small, we conclude:

•

All cooking and sanitation alternatives need further

improvements before considering the technologies as

appropriate and viable options for sustainable com-

munity development.

•

Further improvements include enhancing

the technological-operational and socio-

economic/financial performance of the technologies

and pushing political/legal actions that support the im-

plementation of cooking and sanitation technologies

on household levels.

•

Valuations of alternatives clearly differ between par-

ticipants and, in particular, with respect to different

groups of stakeholder representatives.

With respect to the latter, representatives of the German

project partners generally tend to be more enthusiastic

about the analysed technologies while representatives of

the Tanzanian partners are more sceptical. This observa-

tion holds true especially for the ICSs technologies, such

as microgasifier and rocket stoves, and the EcoSan and

CaSa sanitation alternatives. Technologies that are more

common and established in both countries, such as bio-

gas or septic systems, are assessed more similarly among

participants. Another example for varying assessments is

that representatives of the implementing NGOs and of the

technology-developers or scientific partners tend to favour

the alternative that they have supported and accompanied,

and, likewise, rate competing technologies below average.

From our hands-on experience in facilitating the MCTA

pre-test, we finally hypothesise that the higher the asses-

sors diversity with respect to education, language, class,

and scientific or practical experiences and knowledge, (i) the

more difficult the process because of uncertainties about

the common understanding and (ii) the more important the

process and the more interesting the results. Testing this

hypothesis could be subject of future scientific MC(D)A

work.

Pre-testing the MCTA with a choice group of participants

can serve as a basic proof-of-concept, with the method as

a simple but viable sustainability assessment tool for the

multi-perspective evaluation of cooking and sanitation alter-

natives. Hence, we conclude that, by providing a framework

and a tool, the MCTA can effectively support case stud-

ies and relevant strategic planning and decision-making.

Moreover, the MCTA is also practicable for other real-world

applications related to small-scale cooking and sanitation

technologies, such as in other regions and/or for other tech-

nologies.

4.3.

Recommendations for Future Real-World Applications

If the MCTA is applied in a different setting, necessary

adjustments include re-phrasing the descriptions of alterna-

tives and adapting the set of criteria. To make the MCTA

further applicable for its use in communities, methodological

challenges and problems, as described in the preceding

sections, need to be addressed. We thus recommend the

following with respect to the choice of participants:

•

Inclusion of farmers or other end-users of the tech-

nologies as participants while avoiding the domi-

nance of participants representing implementers, re-

searchers, or donors.

•

Inclusion of agency representatives in order to push

policy development and/or national support programs,

such as subsidies.

•Realisation of a larger number of participants.

•

Awareness of an adequate gender balance in the

assessment team.

In addition, we recommend the following with respect to

the assessment process and method:

•

Reduction of complexity and workload, for example

through intelligent reduction of the number of sub-

criteria.

•

If possible, conducting the MCTA with participants lo-

cated at one spot, at least during scoring or evaluation

of results.

•

Inclusion of more feedback loops through group meet-

ings and discussions at different steps of the MCTA.

•

Potentially, the assessment could also be conducted

in separate groups for men and women in order to

allow women to speak more openly.

Reducing the number of sub-criteria could follow a partici-

patory approach comprising, firstly, a moderated discussion

about the existing catalogue of sub-criteria and, thereafter,

a selection of those criteria considered as the most rele-

vant. Methodologically, the “silent negotiation” could be

applied in a moderated workshop, which helps to cope with

power-imbalances in the consortium [

]. If practitioners

do not aim at running through a whole assessment pro-

cess, the existing catalogue of criteria could also serve to

structure focus group discussions or community assemblies.

Moderated discussions between participants can help to

enhance the joint understanding of arguments that underlie

and explain differences in the quantitative assessments.

For example, a group discussion after reading the descrip-

tions can help to ease a common understanding about the

information provided before scoring the alternatives. At the

same time, such a discussion could help to alleviate the

time budget of the facilitator for clarifying open questions

from participants. With more time available, another option

could be to firstly conduct scoring of alternatives without

description, secondly with description prepared by the plan-

ner, and finally to compare and to discuss results together.

Finally, and with respect to the alternatives, we recommend

the inclusion of technologies that are most commonly ap-

plied so far to support a comparison of alternatives to the

actual real-world situation.

References and Notes

[1]

Status of the World’s Soil Resources (SWSR)–Main Report. Food

and Agriculture Organization of the United Nations and Intergovern-

mental Technical Panel on Soils, Rome, Italy. 2015;650.

[2]

Kiers ET, Leakey RR, Izac AM, Heinemann JA, Rosenthal E, Nathan

D, et al. Agriculture at a crossroads. Science. 2008;320(5874):320.

doi:10.1126/science.1158390.

[3]

Cordell D, White S. Peak phosphorus: clarifying the key issues of a

vigorous debate about long-term phosphorus security. Sustainability.

2011;3(10):2027–2049. doi:10.3390/su3102027.

[4]

Climate Change I. The physical science basis. Contribution of Work-

ing Group I to the Fifth Assessment Report of the Intergovernmental

Panel on Climate Change United Nations: Geneva, Switzerland.

2013;.

[5]

Montanarella L, Pennock DJ, McKenzie N, Badraoui M, Chude V,

Baptista I, et al. World’s soils are under threat. Soil. 2016;2(1):79–82.

doi:10.5194/soil-2-79-2016.

[6]

Steffen W, Richardson K, Rockstr

om J, Cornell SE, Fetzer I, Ben-

nett EM, et al. Planetary boundaries: Guiding human develop-

ment on a changing planet. Science. 2015;347(6223):1259855.

doi:10.1126/science.1259855.

[7]

Meier T. Planetary boundaries of agriculture and nutrition–an An-

thropocene approach; 2017. In: Proceedings of the Symposium

on Communicating and Designing the Future of Food in the An-

thropocene. Humboldt University Berlin. Bachmann Verlag, Berlin,

Germany.

[8]

Revised World Soil Charter. Food and Agriculture Organization of

the United Nations (FAO). 2015; Available at: http://www.fao.org/3/a-

i4965e.pdf (Assessed on 18 Apr 2017).

[9]

Campesina LV. Declaration of the International Forum for Agroecol-

ogy; 2015. Available at: https://viacampesina.org/en/index.php/main-

issues-mainmenu-27/sustainable-peasants-agriculture- mainmenu-

42/1749-declaration-of-the-international-forum-for-agroecology (As-

sessed on 15 Nov 2016).

[10]

Markwei C, Ndlovu L, Robinson E, Shah W. International assess-

ment of agricultural knowledge, science, and technology for develop-

ment (IAASTD) sub-Saharan Africa summary for decision makers.

Rep., Food Agric., Nat. Resour. Policy Anal. Netw., Pretoria, S.

Afr. http://www.fanrpan.org/documents/d00522/IAASTD summary

decision makers; 2008.

[11]

Kaltschmitt M, Hartmann H, Hofbauer H. Energie aus Biomasse-

Grundlagen, Techniken und Verfahren. 2009. Heidelberg Dordrecht

London NewYork: Springer. ISBN 978-3-540-85095-3.

[12]

Esrey SA, Andersson I, Hillers A, Sawyer R, et al. Closing the loop:

ecological sanitation for food security. 2001. Water Resources No.

18. Sweden/Mexico: Swedish International Development Cooper-

ation Agency (SIDA). Water Resources No. 18. Sweden/Mexico:

Swedish International Development Cooperation Agency (SIDA).

[13]

Graeub BE, Chappell MJ, Wittman H, Ledermann S, Kerr RB,

Gemmill-Herren B. The state of family farms in the world. World

Development. 2016;87:1–15. doi:10.1016/j.worlddev.2015.05.012.

[14]

URoT U. National Sample Census of Agriculture 2007/2008. Volume

VI: Regional Report: Mbeya Region Tanzania: National Bureau of

Statistics. 2012;pp. 1–326.

[15]

Ogola S. Land and natural resources conflicts in trans-boundary

agro-ecosystem management project, Kagera basin. Food and Agri-

culture Organization (FAO) of the United Nations Retrieved January.

2013;8:2014.

[16]

Rugalema G, Okting’Ati A, Johnsen F. The homegarden agro-

forestry system of Bukoba district, North-Western Tanzania. 1.

Farming system analysis. Agroforestry Systems. 1994;26(1):53–64.

doi:10.1007/BF00705152.

[17]

Krause A, Kaupenjohann M, George E, Koeppel J. Nutrient recy-

cling from sanitation and energy systems to the agroecosystem-

Ecological research on case studies in Karagwe, Tanzania.

African Journal of Agricultural Research. 2015;10(43):4039–4052.

doi:10.5897/AJAR2015.10102.

[18]

Excerpts from the self-conception of facilitating initiatives under-line

the importance of sustainability and sustainable community devel-

opment for our partner organizations. According to MAVUNO [88]

the objective of the organisation is “(. . . ) to empower rural people

of Karagwe by (. . . ) creating sustainable alternatives”. Moreover,

“projects are always planned and implemented in collaboration with

community members in order to achieve sustainability and long term

socio-economic and ecological improvement”. CHEMA [89] envi-

sions “(. . . ) to increase awareness (. . . ) of the relationship between

environmental protection, sustainable agriculture and sustainable

development” and “(. . . ) to identify surrounding opportunities and

put them into use in a sustainable manner”.

[19]

Krause A, Rotter VS. Linking energy-sanitation-agriculture: In-

tersectional resource management in smallholder households in

Tanzania. Science of the Total Environment. 2017;590:514–530.

doi:10.1016/j.scitotenv.2017.02.205.

[20]

Krause A, Nehls T, George E, Kaupenjohann M. Organic wastes

from bioenergy and ecological sanitation as a soil fertility improver:

a field experiment in a tropical Andosol. Soil. 2016;2(2):147–162.

doi:10.5194/soil-2-147-2016.

[21]

Krause A, Rotter VS. Recycling improves soil fertility manage-

ment in smallholdings in Tanzania. Agriculture. 2018;8(3):31.

doi:10.3390/agriculture8030031.

[22]

According to [90], Sustainability Assessment is defined as a method

“to provide decision-makers with an evaluation of global to local inte-

grated nature-society systems in short and long term perspectives

in order to assist them to determine which actions should or should

not be taken in an attempt to make society sustainable”.

[23]

Christian R. Feasibility assessment tool for urban anaerobic diges-

tion in developing countries, a participatory multi-criteria assessment

from a sustainability perspective applied in Bahir Dar, Ethiopia. MSc

Thesis Environmental Science (Major in Environmental Technology),

Wageningen University Netherlands, Final 24 January 2012 Google

Scholar; 2012.

[24]

Belton V, Stewart T. Multiple criteria decision analysis: an integrated

approach. Springer Science & Business Media; 2002.

[25]

According to [24], MCDA is “an umbrella term to describe a col-

lection of formal approaches which seek to take explicit account of

multiple criteria in helping individuals or groups explore decisions

that matter”.

[26]

We use the abbreviation MCDA for referring to Multi-Criteria Deci-

sion Analysis. Other established names for the method include multi

criteria decision aid, multi attribute decision making, multi objective

decision making, or multi criteria decision making [91].

[27]

According to [29], a Multi-Criteria Analysis (MCA) “establishes pref-

erences between options by reference to an explicit set of objectives

that the decision making body has identified”.

[28]

The abbreviation MCA refers also to Multi-Criteria Assessment [32].

[29]

Dodgson JS, Spackman M, Pearman A, Phillips LD. Multi-criteria

analysis: a manual. 2009;.

[30]

oksalan M, Wallenius J, Zionts S. An early history of multiple

criteria decision making. Journal of Multi-Criteria Decision Analysis.

2013;20(1-2):87–94. doi:10.1002/mcda.1481.

[31]

Munda G. Multiple criteria decision analysis and sustainable de-

velopment. In: Multiple criteria decision analysis: State of the art

surveys. Springer; 2005. pp. 953–986. doi:10.1007/0-387-23081-

5 23.

[32]

Gerber JF. Guide on multicriteria evaluation for Environmental Jus-

tice Organisations. 2013.

[33]

Kiker GA, Bridges TS, Varghese A, Seager TP, Linkov I. Appli-

cation of multicriteria decision analysis in environmental decision

making. Integrated environmental assessment and management.

2005;1(2):95–108. doi:10.1897/IEAM 2004a-015.1.

[34]

Mendoza GA, Martins H. Multi-criteria decision analysis in natural re-

source management: a critical review of methods and new modelling

paradigms. Forest ecology and management. 2006;230(1-3):1–22.

doi:10.1016/j.foreco.2006.03.023.

[35]

Huang IB, Keisler J, Linkov I. Multi-criteria decision analy-

sis in environmental sciences: ten years of applications and

trends. Science of the total environment. 2011;409(19):3578–3594.

doi:j.scitotenv.2011.06.022.

[36]

Mucunguzi DB. Sustainability assessment of ecological sanitation

using multi criteria analysis: A Case Study of Kabale, Uganda;.

Dar es Salaam, Tanzania: University of Dar es Salaam; Masters

Dissertation; 2010.

[37]

Rajabu H, Ndilanha A. Improved Cook Stoves Assessment and

Testing. Report of the ICS Taskforce;. Dar Es Salaam, Tanzania:

University of Dar Es Salaam, SNV; 2013.

[38]

Schindler J, Graef F, K

onig HJ, Mchau D, Saidia P, Sieber S. Sustain-

ability impact assessment to improve food security of smallholders in

Tanzania. Environmental Impact Assessment Review. 2016;60:52–

63. doi:10.1016/j.eiar.2016.04.006.

[39]

Multi-criteria decision analysis. In: Lecture notes from Interdisci-

plinary PhD training week ’Is small sustainable?’.

[40]

Wilkens I. Multikriterielle Analyse zur Nachhaltigkeitsbewertung

von Energiesystemen-Von der Theorie zur praktischen Anwendung.

2012;.

[41]

Liu KFR, Ko CY, Fan C, Chen CW. Combining risk assessment, life

cycle assessment, and multi-criteria decision analysis to estimate

environmental aspects in environmental management system. The

International Journal of Life Cycle Assessment. 2012;17(7):845–862.

doi:10.1007/s11367-012-0407-x.

[42]

Nordstr

om EM. Integrating multiple criteria decision analysis into

participatory forest planning. vol. 2010; 2010.

[43]

Stahl CH, Cimorelli AJ, Chow AH. A new approach to environmental

decision analysis: Multi-criteria integrated resource assessment

(MIRA). Bulletin of Science, Technology & Society. 2002;22(6):443–

459. doi:10.1177/0270467602238888.

[44]

Here, the term trifecta is used to describe an integrated system

that consists of three sub-systems. Hence, “trifecta” was used syn-

onymously to “triple”; http://www.merriam-webster.com/dictionary/

trifecta;accessed27/09/2016. The PDF has later been adapted to

‘Intersectional Resource Management’ [19].

[45]

Brown J. A resource guide for hosting conversations that matter

at the World Caf

e. Available from: http://wwwinscenach/fileadmin/

user upload/inscena/pdf/worldcafe.pdf. 2002.

[46]

Jakobsen H. Focus groups and methodological rigour outside the

minority world: Making the method work to its strengths in Tanzania.

Qualitative Research. 2012;12(2):111–130.

[47]

Hellstr

om D, Jeppsson U, K

arrman E. A framework for systems

analysis of sustainable urban water management. Environmental im-

pact assessment review. 2000;20(3):311–321. doi:10.1016/S0195-

9255(00)00043-3.

[48]

Van de Klundert A, Ansch

utz J, Scheinberg A. Integrated sustainable

waste management: the concept. Tools for decision-makers. expe-

riences from the urban waste expertise programme (1995-2001).

WASTE; 2001.

[49]

UN. Progress towards the Sustainable Development Goals. The

Secretary-General. Report E/2016/75; 2016.

[50]

Kubanza NS. Urban environmental problems: social and envi-

ronmental injustices in solid waste management in Kinshasa, the

Democratic Republic of Congo; 2016.

[51]

Baccini P, Brunner PH. Metabolism of the anthroposphere: analysis,

evaluation, design. MIT Press; 2012.

[52]

Stoorvogel JJ, Smaling EMA. Assessment of soil nutrient depletion

in Sub-Saharan Africa: 1983-2000. Vol. 2: Nutrient balances per

crop and per land use systems. ISRIC; 1990.

[53]

Santillo D. Reclaiming the Definition of Sustainability (7

pp). Environmental Science and Pollution Research-International.

2007;14(1):60–66. doi:10.1065/espr2007.01.375.

[54]

Cumming G, Fidler F, Vaux DL. Error bars in experimen-

tal biology. The Journal of cell biology. 2007;177(1):7–11.

doi:10.1083/jcb.200611141.

[55]

Friederici K. Everyday understanding of energy and energy services

in the context of sustainable implementation of micro-energy sys-

tems in Tanzania;. Berlin, Germany: Micro-energy system Research

Program, TU Berlin; PhD project in progress.

[56]

Yiougo LS, Koanda H, Wethe J, Yonkeu S, Da E. Multi-Criteria

Decision Analysis and choice of sanitation technology: challenge

in urban water quality of Pouytenga, Burkina Faso. Urban waters:

Resource or Risk?;p. 42.

[57]

Rupf GV, Bahri PA, de Boer K, McHenry MP. Barriers and op-

portunities of biogas dissemination in Sub-Saharan Africa and

lessons learned from Rwanda, Tanzania, China, India, and Nepal.

Renewable and Sustainable Energy Reviews. 2015;52:468–476.

doi:10.1016/j.rser.2015.07.107.

[58]

Chaggu EJ. Sustainable environmental protection using modified

pit-latrines; 2004.

[59]

Richert A, Gensch R, J

onsson H, Stenstr

om TA, Dagerskog L. Prac-

tical guidance on the use of urine in crop production. SEI; 2010.

[60]

Loetscher T, Keller J. A decision support system for selecting sani-

tation systems in developing countries. Socio-Economic Planning

Sciences. 2002;36(4):267–290. doi:10.1016/S0038-0121(02)00007-

[61]

Bond A, Morrison-Saunders A, Pope J. Sustainability assessment:

the state of the art. Impact Assessment and Project Appraisal.

2012;30(1):53–62.

[62]

Joyce LA. Improving the flow of scientific information across the

interface of forest science and policy. Forest Policy and Economics.

2003;5(4):339–347. doi:10.1016/S1389-9341(03)00033-9.

[63]

Millstone E, Van Zwanenberg P, Marshall F. Monitoring and evaluat-

ing agricultural science and technology projects: theories, practices

and problems. IDS Bulletin. 2010;41(6):75–87. doi:10.1111/j.1759-

5436.2010.00185.x.

[64]

Department of Mechanical and Industrial Engineering, College of

Engineering and Technology, University of Dar es Salaam, Tanzania

(TZ).

[65]

School of Environmental Science and Technology, Ardhi University,

Dar es Salaam, TZ.

[66]

Department of Soil Science, Sokoine University of Agriculture, Mo-

rogoro, TZ.

[67]

Agricultural Research Institute of the Ministry of Agriculture, Food,

and Cooperatives, Uyole, TZ.

[68]

Swiss Federal Institute of Aquatic Science and Technology (Eawag),

Department of Sanitation, Water and Solid Waste for Development

(Sandec), D¨

ubendorf, Switzerland.

[69]

Centre for Research in Energy and Energy Conservation (CREEC),

College of Engineering, Design, Art and Technology, Makerere Uni-

versity, Kampala, Uganda.

[70]

Department of Agricultural Production, Makerere University, Kam-

pala, Uganda.

[71]

School of Agricultural Science, College of Agriculture and Environ-

mental Sciences, Makerere University, Kampala, Uganda.

[72]

Caritas Development Office, national implementing partner of the

TDBP, Roman Catholic Church Mbeya Region, TZ.

[73]

Centre for Agricultural Mechanisation and Rural Technology (CA-

MARTEC), national implementation agency of the TDBP, Arusha,

TZ.

[74]

Section Sanitary Engineering and section Integral Design and

Management, Department of water management, faculty of Civil

Engineering and Geosciences, University of Technology, Delft, the

Netherlands.

[75] Awamu Biomass Energy Limited, Kampala, Uganda.

[76] Jellitone Suppliers Ltd., Kampala, Uganda.

[77] Joint energy and Environment Projects, Kampala, Uganda.

[78]

Bremen Overseas Research and Development Association, Dar Es

Salaam, TZ.

[79] Norges Vel East adn Southern Africa, Dar Es Salaam, TZ.

[80] SNV Netherland Development Organiszation, Arusha, TZ.

[81]

Department of Mechanical Engineering, Mbeya University of Sci-

ence and Technology, Mbeya, TZ.

[82]

“Is small sustainable? Decentralizing Infrastructures and Utility

Systems in East Africa”, PhD summerschool of TU Berlin and TU

Darmstadt, Kurasini Training & Conference Centre, Dar es Salaam,

TZ.

[83]

The conceptual and analytical work was supported by Dr. L.

Scholten.

[84]

‘Stakeholder’ is defined as: “actors who have a stake, an interest

in the issue under consideration; who are affected by it, or who

-because of their position - have or could have an active or passive

influence on the decision making and implementation processes”.

[85]

‘Power’ is defined as: “the extent to which they (i.e. the participants)

are able to persuade or coerce others into making certain a decision

or following certain courses of action”.

[86]

‘Interest’ is defined as: “the extent to which a certain issue is given

priority”.

[87]

Microenergy systems Research Group, Postgraduate program at

Center for Technology and Society, TU Berlin, Germany.

[88]

MAVUNO. About the organisation and its objectives;. Available from:

http://www.mavunoproject.org/objectives/.

[89]

CHEMA. About the organisation and its vision and goals;. Available

from: http://chematanzania.org/?page id=19.

[90]

Ness B, Urbel-Piirsalu E, Anderberg S, Olsson L. Categoris-

ing tools for sustainability assessment. Ecological economics.

2007;60(3):498–508. doi:10.1016/j.ecolecon.2006.07.023.

[91]

Omann I. Multi-criteria decision aid as an approach for sustainable

development analysis and implementation. na; 2004.