Energy Policy 39 (2011) 6683–6688

Contents lists available at ScienceDirect

Energy Policy
journal homepage: www.elsevier.com/locate/enpol

Communication

Risk assessment of river-type hydropower plants using fuzzy logic approach

Serhat Kucukali n
- ankaya University, Balgat 06530, Ankara, Turkey
Civil Engineering Department, C

a r t i c l e i n f o abstract

Article history: In this paper, a fuzzy rating tool was developed for river-type hydropower plant projects, and risk
Received 12 January 2011 assessment and expert judgments were utilized instead of probabilistic reasoning. The methodology is
Accepted 30 June 2011 a multi-criteria decision analysis, which provides a flexible and easily understood way to analyze
Available online 23 July 2011
project risks. The external risks, which are partly under the control of companies, were considered in
Keywords: the model. A total of eleven classes of risk factors were determined based on the expert interviews, field
Hydropower studies and literature review as follows: site geology, land use, environmental issues, grid connection,
Risk Analysis social acceptance, macroeconomic, natural hazards, change of laws and regulations, terrorism, access to
Fuzzy Logic infrastructure and revenue. The relative importance of risk factors was determined from the survey
results. The survey was conducted with the experts that have experience in the construction of river-
type hydropower schemes. The survey results revealed that the site geology and environmental issues
were considered as the most important risks. The new risk assessment method enabled a Risk Index
(R) value to be calculated, establishing a 4-grade evaluation system. The proposed risk analysis will give
investors a more rational basis to make decisions and it can prevent cost and schedule overruns.
& 2011 Elsevier Ltd. All rights reserved.

1. Introduction In the literature there are several studies considering risk

Renewable energy projects are subjected to different types
of risks throughout their life cycle each of which will cause
additional cost and schedule overrun or project failure. The
investors and financial institutions are interested in projects that
can provide secure cash flows with minimal risk. The surveys
conducted by Gronbrekk et al. (2010) and Komendantova et al.
(in press) identified the highest risk as political and regulatory
changes for renewable energy projects in developing countries.
Similarly, Ernst and Young (2010) identified the most important
business risk for 2010 as regulation and compliance.
Construction of river-type hydropower plants involves uncer-
tainties because of various external factors such as site geology, grid
connection and environmental issues. These factors increase the
construction costs and duration. For example, in one of the river-
type hydropower plants, Kulp IV in Turkey, the cost of civil works
increased twofold because of unpredicted geologic structure at the
tunneling site. In another example, the judges have ruled against
hydroelectric power plants in 33 completed cases in Turkey, issuing
a stay of execution decision or canceling the construction altogether
because of the environmental issues. Besides, a hydropower scheme
on the Malagarasi River in Tanzania was initially approved for grant
funding, but was subsequently rejected because of the identifica-
tion of potentially significant biodiversity impacts, which had not
been adequately addressed in Environmental Impact Assessment
(EIA) report (Hovland, 2010).
n
Tel.: þ90 312 2844500.
E-mail address: kucukali@cankaya.edu.tr
analysis in construction projects (Zavadskas et al., 2010). However,
risk analysis in renewable energy projects, especially for hydro-
power plants, is very limited. In classical project risk analysis
techniques, risk rating values are calculated by multiplying impact
and probability values, but direct analysis of these linguistic factors
is often neglected (Dikmen et al., 2007). Most existing risk analysis
models, such as Monte Carlo simulation and tornado chart, are
based on quantitative techniques, which require numerical data.
Kangari and Riggs (1989) note that probabilistic models suffer from
detailed quantitative information, which is not normally available
in the real construction world. However, much of the information
related to risk analysis is not numerical (Mustafa and Al-Bahar,
1991). Rather, this information is expressed as words or sentences
in a natural language. These conceptual factors can be expressed in
linguistic terms, so-called fuzzy information (Kucukali and Baris,
2010). Uncertainty factors such as ‘‘poor geology’’ or ‘‘unstable
policy’’ fall into this category. The aim of this paper is to introduce a
new approach for hydropower projects risk assessment through the
fuzzy logic concept. The methodology addressed the decision
circumstances at a preliminary phase of an investor that investi-
gates the risk of alternative potential hydro projects in order to
decide which ones to undertake.
2. Methodology
In the context of this study, a total of eleven classes of risk
factors were determined based on the expert interviews, field
studies and literature review (Table 1). The risk factors are

0301-4215/$ - see front matter & 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.enpol.2011.06.067
6684 S. Kucukali / Energy Policy 39 (2011) 6683–6688

Table 1
Answers of each of the 15 experts for the importance of the risk factors. The survey was done based on a qualitative basis: Low¼ 1, Medium¼ 2, High¼3 and Very High¼ 4.

Site Land Environ. Grid Social Macro Natural Change of Terrorism Access to Revenue
geology use issues connection acceptance econ. hazards laws infrastructure

Exp. 1 2 1 2 1 3 2 2 2 3 2 3
Exp. 2 3 2 3 3 3 1 1 2 2 1 1
Exp. 3 3 2 3 3 3 3 2 3 1 2 3
Exp. 4 4 2 3 2 2 1 1 1 2 1 2
Exp. 5 3 3 3 2 3 2 2 2 2 2 3
Exp. 6 3 1 2 1 2 2 2 1 1 2 2
Exp. 7 3 4 3 2 3 4 3 3 1 2 4
Exp. 8 3 4 3 4 2 4 3 3 2 3 4
Exp. 9 2 4 3 2 2 2 2 3 1 2 4
Exp. 10 3 3 4 2 4 2 4 1 2 3 1
Exp. 11 3 3 3 1 2 1 2 2 1 1 4
Exp. 12 4 3 3 3 2 2 3 2 1 2 2
Exp. 13 3 3 3 1 2 2 1 1 1 1 1
Exp. 14 3 3 4 3 3 2 2 2 2 2 2
Exp. 15 2 3 4 3 4 3 2 3 3 2 3
Total Score 44 41 46 33 40 33 32 31 25 28 39
Share of total 11.2 10.4 11.8 8.5 10.1 8.5 8.2 7.9 6.3 7.1 9.9
score (%)

determined as follows: site geology (geotechnical properties of 2008). Eq. (2) shows the fuzzy grading matrix (FG) used in this
the construction site), land use and permits (right to use of the study:
land for the construction of hydropower scheme), environmental 2 3
1 1 0:4 0 0
issues (impact of the scheme on ecosystem), grid connection
266 0:2 1 0:2 0 7
7
(connection to the power system), social acceptance (impact FG ¼ Score 6 7 ð2Þ
of the scheme on local community who use the river or the 4
3 0 0:2 1 0:2 5
surrounding lands), macroeconomic (inflation and interest rate), 4 0 0 0:4 1
natural hazards (earthquake, flooding, storm and landslide), The fuzzy assessment matrix (FA) was obtained by multiplying
change of laws and regulations (level of political stability), input matrices (I) with fuzzy grading matrix (FG) of the para-
terrorism (human-made disasters), access to infrastructure (road) meter,
and revenue (cash flow). These risk factors and their evaluation
criteria are listed in Table 2. For land use and permits evaluation FAj ¼ Ij  FGj ðj ¼ 1 to 11Þ ð3Þ
was done based on the time needed to obtain all permits required. where j is the row number of the fuzzy assessment matrices. The
Estimated average time needed to obtain the permit in each case membership degree matrix (MD) was obtained by multiplying
is as follows: Forest ¼1 month, Property of Treasury ¼2 months, weight of parameters (w) with fuzzy assessment matrix (FA) and
Agricultural land¼6 months and Residential area ¼8 months. summing the columns resulting in a one-row matrix;
Economic risks are divided according to gross national income
(GNI) per capita, calculated using the World Bank Atlas method. MD ¼ w  FA ð4Þ
The natural hazards that are considered in this study are as In order to evaluate the unit investment cost of hydropower
follows: earthquake, flooding, storm and landslide. The probabil- plants the Risk Index R is defined using the membership degree
ity of each natural hazard is evaluated. The highest risk level of curve as
the natural hazards is assessed in the model.
1  A12 þ2  A23 þ 3  A34
In order to determine the relative importance (impact) of the R¼ ð5Þ
risk factors, a survey was conducted with the experts working for AT
banks and companies that have experience in the construction Aij refers to the area between the curve and the horizontal axis
of river-type hydropower schemes. A total of 15 experts were between i and j features. ‘‘I’’ and ‘‘j’’ features follow each other
participated to the survey. The survey was done based on a (i¼ 1, 2, 3, 4 and j ¼iþ1). ‘‘AT’’, on the other hand, refers to the
qualitative basis (Table 1). The linguistic terms that were used total area between the membership rating qualifications graphic
in the ranking of risk factors are: low, medium, high and very and its horizontal axis. This enabled a Risk Index (R) value to be
high. The experts were asked to mark the level of risk factors via calculated, establishing a 4-grade evaluation system: low risk
e-mail. Then the average score value of each risk factor has been having R values between 1.2 and 1.6; medium risk, between
used to determine their relative importance in the model. The 1.6 and 2; high risk, between 2 and 2.4; very high risk, between
experts ranked environmental issues and site geology as the most 2.4 and 2.8. The risk scale index represents the minimum and
important risks for river-type hydropower plants (Fig.1). maximum values calculated by Eq. (5). Risk index is linked to the
For each of the 11 parameter, an 1  4 input matrix was specific investment cost of hydropower plants presented in
developed, each column corresponding scores from 1 to 4. If the Table 3.
score for a parameter is 3 then the input matrix (I) for the
parameter is
  3. Investment costs of hydropower schemes
I¼ 0 0 1 0 ð1Þ

Each parameter has an identical membership grading matrix. Hydropower is the backbone of carbon dioxide free energy
The fuzzy grading matrices were developed considering the generation; about 22% of the world’s electricity production comes
degree of error a scoring observer may cause due to subjectivity from hydropower installations (Boyle, 2004). Hydropower plants
and bias in the assessment process (Ergin et al., 2006; Sahin, can be classified into two categories: storage and river-type.
 Table 2
Evaluation criteria of risk factors in river-type hydropower plants.

Risk factor Low risk (Score ¼ 1) Medium risk (Score ¼ 2) High risk (Score ¼3) Very high risk (Score ¼4)

Site geology Rock mass quality is very good-good: RQD ¼%60– Rock mass quality is fair: RQD ¼ %40–60 Rock mass quality is poor–very poor: Soil with high ground water level
100 RQD ¼%0–50
Land use and Forest Property of Treasury Private property: agricultural land Private property: residential area
permits
Environmental Project has a detailed Environmental Impact Project has an Environmental Impact Assessment Report Project has no Environmental Impact Project is in an ecological sensitive area.
issues Assessment Report (Biodiversity issues are Assessment Report
addressed)
Grid The distance of the facility to the power system is The distance of the facility to the power system is between 1–5 km The distance of the facility to the Connection to the power system has
connection less than 1 km power system is higher than 5 km some limitations
Social A robust public consultation process has been A public consultation process has been carried out. The locally The facility reduces local community The facility stops or limits local
acceptance carried out. No major objections from local affected community has been notified and adequate mitigation use of either the river or the communities ability to utilize the river
communities were raised measures have been taken. surrounding lands to provide a livelihood
Macroeconomic GNI per capita412,196 $ GNI per capita¼ 3946 $- 12,195 $ GNI per capita¼ 996 $- 3945 $ GNI per capita o 996 $
Natural hazards Less than 1% probability of occurring Likely to occur between 1% and 10% Likely to occur between 10% and 50% Greater than 50% probability of
occurring
Change of laws Country has robust legal context for renewables Country has legal context for renewables including, hydropower Country has legal context for other Country has no legal context for

S. Kucukali / Energy Policy 39 (2011) 6683–6688

and regulations including hydropower renewables but not for hydropower renewables
Terrorisma Terror risk index of country is low Terror risk index of country is medium Terror risk index of country is high Terror risk index of country is extreme
Access to Quality of the nearby road network is high Quality of the nearby road network is medium Quality of the nearby road network is There is no road network to the
infrastructure low construction site
Revenue Design discharge occurrence probability is more Design discharge occurrence probability is between 60–80% Design discharge occurrence Design discharge occurrence probability
than 80% probability is between 40–60% is less than 40%

a
Terrorism risk index is based on Maplecroft (2010).
poor geology and the technical demands by TEIAS, respectively.
grid connection for the project increased twofold because of the
Kulp IV hydropower plant. The actual cost of civil works and the
length¼1885 m. Table 5 presents the investment cost analysis of
20 m3/s, output¼12.68 MW, energy¼36.64 GWh/year and tunnel
of the project are as follows: gross head¼77 m, discharge¼
structed on Dicle River in East Anatolia (Fig. 3). The characteristics
real-time hydropower project namely Kulp IV, which was con-

4. Application of the proposed methodology

and grid connection (Table 5).

financial, civil works, electro and hydro mechanical equipment
classified as follows: project design, land use and permits,
Based on the investment costs a hydropower plant can be
installed capacity decreases (Gordon, 1983; Aggidis et al., 2010).
that the specific investment cost increases as the head and
head and installed capacity. It is well known from the literature
investment cost of a hydropower plant is the function of the net
overrun as the rate of inflation and site geology. The specific
Gordon (1983) identified the main factors, which may lead to cost
hydropower plants among other renewable energies (Table 4).
specific; hence there is a wide range of investment costs of
for 65–75% of the capital cost. Each hydropower project is site-
2000–4000 $/kW. Also, they reported that the civil works account
capacity) of river-type hydropower plants in USA in the range of
(total investment cost of a project divided by the installed

2009).
greatly with time of the year or the weather (Kucukali and Baris,
flow rate and can present problems of reliability if the flow varies
River-type hydropower plants are dependent on the prevailing
hydropower (SHP) plants are mostly included in this category.
portion of river through a channel or tunnel (Fig. 2). Small
flow in a reservoir. A river-type hydropower plant diverts a
In storage type hydropower plants, dams are used to retain river

the survey results.

Fig. 1. The importance of risk factors of river-type hydropower plants based on
The developed risk assessment technique was applied to a

Hall et al. (2003) determined the specific investment cost

Risk index and its relation with unit investment cost.

Table 3
Very high: R¼ 2.4–2.8
High: R ¼2–2.4
Medium: R¼ 1.6–2
Low: R¼ 1.2–1.6

Risk index
5500–7000
4000–5500
2500–4000
1000–2500

cost ($/kW)
Unit investment

6685
6686 S. Kucukali / Energy Policy 39 (2011) 6683–6688

DIVERSION CHANNEL

Y
E

BA
AC W A T ER

RE
L R INTAK
E
AI

FO
T
RIVER
ENVIRONMENTAL
H FLOW
BYPASSED REAC FLOW
WEIR

Fig. 2. Site plan and components of a river-type hydropower plant.

Table 4
Comparison of investment and energy production costs for some renewable.
data source: IEA, 2008.

Type of renewable Plant size (MW) Typical current Typical energy

specific investment production
cost ($/kW) cost ($ct/kWh)

Large hydro 10–18,000 1,000–5,500 3–12

Small hydro 1–10 2,500–7,000 6–14
Onshore wind 1–3 1,200–1,700 1–14
Geothermal 1–100 1,700–5,700 3–10
Solar PV 1–10 5,000–6,500 20–80

Table 5
Analysis of the investment cost of Kulp IV hydropower plant (P ¼12.7 MW, unit investment cost ¼4163 $/kW).

Description Estimated cost ($) Actual cost ($) Rate of Share of total Reason
Increase (%) Cost (%)

Project design 1,090,000 1,180,000 8.3 2.2 Additional project designs

Civil works 12,500,000 25,700,000 105.6 48.6 Poor geology (serpentine) at the tunneling site
EMEa 6,790,000 7,490,000 10.3 14.2 Under estimated costs of Technical equipment demand
HMEb 8,900,000 4,800,000  46.1 9.1 The prices of DSI are very high compared to the market.
Grid connection 2,600,000 5,400,000 107.7 10.2 Technical demands by TEIAS and length of the power supply line
was increased
Land use and permits 2,200,000 2,600,000 18.2 4.9 The cost of the forest usage permit was not taken into account
Financial 3,100,000 5,700,000 83.9 10.8 Increase in interest rates because of the financial crisis

a
Electro mechanical equipment.
b
Hydro mechanical equipment.

Fig. 3. River basins of Turkey and location of the Kulp IV hydropower plant.
 S. Kucukali / Energy Policy 39 (2011) 6683–6688 6687

Table 6
Fuzzy risk assessment rating tool application for Kulp IV Hydropower Plant.

Risk assessment

No. Risk factor Score Relative Input matrix (I) Fuzzy logic evaluation
importance
(W)
Fuzzy grading matrix (FG) Fuzzy assessment matrix (FA)
Membership degree

1 2 3 4

1 Geology 3 0.112 0 0 1 0 FG¼ f(I) 0.00 0.20 1.00 0.20 FA¼ FG*W 0.000 0.022 0.112 0.022
2 Land rent 4 0.104 0 0 0 1 0.00 0.00 0.40 1.00 0.000 0.000 0.042 0.104
3 Environment 3 0.118 0 0 1 0 0.00 0.20 1.00 0.20 0.000 0.024 0.118 0.024
4 Grid connection 3 0.085 0 0 1 0 0.00 0.20 1.00 0.20 0.000 0.017 0.085 0.017
5 Social acceptance 2 0.101 0 1 0 0 0.20 1.00 0.20 0.00 0.020 0.101 0.020 0.000
6 Macroeconomic 2 0.085 0 1 0 0 0.20 1.00 0.20 0.00 0.017 0.085 0.017 0.000
7 Natural hazard 3 0.082 0 0 1 0 0.00 0.20 1.00 0.20 0.000 0.016 0.082 0.016
8 Law changes 3 0.079 0 0 1 0 0.00 0.20 1.00 0.20 0.000 0.016 0.079 0.016
9 Terorism 4 0.063 0 0 0 1 0.00 0.00 0.40 1.00 0.000 0.000 0.025 0.063
10 Access to insfrastructure 3 0.071 0 0 1 0 0.00 0.20 1.00 0.20 0.000 0.014 0.071 0.014
11 Revenue 2 0.099 0 1 0 0 0.20 1.00 0.20 0.00 0.020 0.099 0.020 0.000
Membership degree matrix (MD) 0.057 0.395 0.672 0.277
0.23 0.53 0.47 1.23
A12 A23 A34 AT
Decision parameter (R) R ¼ 2.20 High risk

Table 6 shows the application of the risk assessment to the Kulp
IV hydropower plant. In the assessment, the fuzzy grading matrix
provides more room for the justification of relationships between
variables on the basis of fuzzy words.
The project risk evaluation was done based on the criteria
presented in Table 2. For example, Turkey has been ranked as
extreme for Terrorism Risk Index by Maplecroft (2010). Therefore,
the risk factor of terrorism has a score of 4 for the project. Yet,
project has no Environmental impact report, which yields the
score of environmental issues as 4. By applying this method to
other risk factors, the Kulp IV hydropower project Risk index was
calculated as 2.20, which means that the project involves high
risk. Based on the risk analysis the project unit investment cost
has been predicted in the range of 4000–5500 $/kW, which is in
agreement with the observed unit investment cost value of the
project as 4163 $/kW.
The influence of fuzzy grading matrix on the results is below
5% and it has very low impact on the calculation of Risk Index
value. However, the input matrix and weights of the parameters
have a considerable influence on the calculation of Risk Index
value. A sensitivity analysis has been done considering four
different scenarios as follows: (i) particularities of the potential
site: site geology, environmental issues, grid connection and
access to infrastructure; (ii) project: site geology, environmental
issues and social acceptance; (iii) stochastic nature: natural
hazards and revenue; (iv) central policy: macroeconomic, change
of laws and terrorism. In each scenario the relevant risk factors
weights have been increased by a factor of two in the model. Then
the risk index value was increased 8% for project scenario, 6% for
stochastic nature scenario and 5% for particularities of the
potential site and the central policy scenarios (Table 7).
5. Conclusions
In this research, a new methodology is proposed for risk rating
of river-type hydropower plant projects. The relative importance
of the risk factors was determined from the expert judgments.
The survey results showed that the most concerned risks are
environmental issues and site geology. Applicability of the pro-
posed methodology was tested on a real case. Findings of the case
study demonstrated that the proposed methodology can easily be
applied by the professionals to quantify risk ratings. The advan-
tage of the proposed methodology is that it will give investors a
more rational basis to make decisions and it can prevent cost and
schedule overruns. Forecasting the measure of risk of a river-type
hydropower plant can be made by any decision maker with the
help of the fuzzy rating tool described in this paper.
Acknowledgment
The author acknowledges the valuable comments of the
reviewer.

Table 7
Sensitivity analysis results for four different scenarios. In each scenario the relevant risk factors
weights have been increased by a factor of two in the model.

Scenario Risk index value

Normal conditions 2.20

Particularities of the potential site: site geology, environmental 2.301
issues, grid connection, access to infrastructure
Project: site geology, environmental issues, social acceptance 2.378
Stochastic nature: natural hazards, revenue 2.331
Central policy: macroeconomic, change of laws, terrorism 2.313
6688 S. Kucukali / Energy Policy 39 (2011) 6683–6688
References
Aggidis, G.A., Luchinskaya, E., Rothschhild, R., Howard, D.C., 2010. The costs of
small-scale hydropower production: impact on the development of existing
potential. Renewable Energy 35, 2632–2638.
Boyle, G., 2004. Renewable Energy: Power for a Sustainable Future. Oxford
University Press.
Dikmen, I., Birgonul, M.T., Han, S., 2007. Using fuzzy risk assessment to rate cost
overrun risk in international construction projects. International Journal of
Project Management 25, 494–505.
Ergin, A., Williams, A.T., Micallef, A., 2006. Coastal scenery: appreciation and
evaluation. Journal of Coastal Research 22 (4), 958–964.
Ernst and Young, 2010. Business Risk Report. /www.ey.comS.
Gordon, J.L., 1983. Hydropower costs estimates. Journal of Water Power Dam
Construction 35, 30–37.
Gronbrekk, W., Barton, H., Khoury, R.H., 2010. International sustainability tools for
hydropower role, relevance and industry reporting trends. In: Proceedings of
the Hydro 2010 Conference, Lisbon, Portugal.
Hall, D.G., Hunt, R.T., Reeves, K.S., Carroll, G.R., 2003. Estimation of Economic
Parameters of US Hydropower Resources. Idaho National Engineering and
Environmental Laboratory.
Hovland, V.L., 2010. Addressing biodiversity issues early in hydropower project
development cyle—case study from Malagarasi, Tanzania. In: Proceedings of
the Hydro 2010 Conference, Lisbon, Portugal.
IEA-International Energy Agency, 2008. Deploying Renewables: Principles for
Effective Policies. OECD.
Kangari, R., Riggs, L.S., 1989. Construction risk assessment by linguistics. IEEE
Transactions on Engineering Management 36, 126–131.
Komendantova, N., Patt, A., Barras, K., Battaglini, A., in press. Perception of risks in
renewable energy projects: the case of concentrated solar power in North
Africa. Energy Policy doi:10.1016/j.enpol.2009.
Kucukali, S., Baris, K., 2009. Assessment of small hydropower (SHP) development in
Turkey: laws, regulations and EU policy perspective. Energy Policy 37, 3872–3879.
Kucukali, S., Baris, K., 2010. Turkey’s short-term gross electricity demand forecast
by fuzzy logic approach. Energy Policy 38 (5), 2438–2445.
Maplecroft, 2010. Terrorism Risk Index. /www.maplecroft.comS.
Mustafa, M.A., Al-Bahar, J.F., 1991. Project risk assessment using the analytic
hierarchy process. IEEE Transactions on Engineering Management 38, 46–52.
Sahin, F., 2008. Scenic Evaluation of the Western Black Sea Coasts Using Fuzzy
Logic, Msc. Thesis, ZKU, Graduate School of Natural and Applied Science,
Deptartment of Civil Engineering, Zonguldak, Turkey (in Turkish).
Zavadskas, E.K., Turskis, Z., Tamosaitiene, J., 2010. Risk assessment of construction
projects. Journal of Civil Engineering and Management 16, 33–46.