1-Household Electrical Consumption Indonesia PDF

Renewable and Sustainable Energy Reviews 57 (2016) 11601173
Contents lists available at ScienceDirect
Renewable and Sustainable Energy Reviews

journal homepage: www.elsevier.com/locate/rser
Characteristics of urban households' electrical energy consumption

in Indonesia and its saving potentials
Hakimul Batih a,b,n, Chumnong Sorapipatana a,b
a
b
The Joint Graduate School of Energy and Environment, KMUTT, Bangkok, Thailand
Center of Excellence on Energy Technology and Environment, Ministry of Education, Thailand
art ic l e i nf o
a b s t r a c t
Article history:
Received 28 November 2013
Received in revised form
11 September 2015
Accepted 17 December 2015
Available online 8 January 2016
An attempt aims to nd out an alternative to avoid expanding existing fossil power plants to cope with a
high growth demand of electricity in Indonesia could be done through electrical energy efciency
improvement. Provided that in-depth knowledge of end-use characteristics of electricity consumption is
known, in this study, a primary survey was conducted between Nov. 2011Jan. 2012 to investigate the
characteristics of urban household electrical energy consumption in Indonesia. Data were collected from
600 respondents living in seven major cities in Indonesia, only 474 (79%) of respondents were considered
to be valid.
The result shows that, lamps, television sets, refrigerators and air conditioning units are appliances
that have the greatest potentials for electrical energy saving. The information from the survey was then
adopted to initiate energy saving programs to avoid construction of new power plants.
It was found that if energy efciency improvement programs are initiated and implemented for the
four types of appliances mentioned above from now to 2030, the costs of saved energy (CSE) of these
energy saving programs are only a fraction of the long run marginal cost (LRMC) of new power plant
constructions, ranging from 5.1% for lamps, 21.6% for television sets, 80.9% for refrigerators, and 19.5% for
air conditioning units, respectively. In addition, energy efciency improvement program can also help to
reduce total amount of electricity generation and CO2 emission. In 2030, this alternative can reduce
amounts electricity generation and CO2 emission attributed for household sector by 21.4% and 21.6%,
respectively, as compared to the conventional supply side of expanding new power plants.
& 2015 Elsevier Ltd. All rights reserved.
Keywords:
Urban household
Electrical energy consumption
Energy efciency
Contents
1.
2.
3.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Data collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.1.
Selection of survey sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2.
Attribution of questionnaire . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.3.
The pilot survey and sample size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4.
Actual survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.5.
Income class determination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Methods for estimating electrical energy saving potentials and economic assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.1.
Households' electrical energy consumption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2.
The ratio of electrical energy expenditure to income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3.
Projection of households' electrical energy consumption under the business as usual case (BAU). . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.4.
Electrical energy consumption by urban households under energy efciency improvement case (EE case) . . . . . . . . . . . . . . . . . . . . .
3.4.1.
Efcient lighting technology program (EE_lighting) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.4.2.
Efcient television technology program (EE_TV) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.4.3.
Efcient space cooling system program (EE_Refrig and EE_AC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Corresponding author at: The Joint Graduate School of Energy and Environment, KMUTT, Bangkok, Thailand. Tel.: 662 4708309 10; fax: 662 8726978.
E-mail address: hakimulbatih@yahoo.com (H. Batih).
http://dx.doi.org/10.1016/j.rser.2015.12.132
1364-0321/& 2015 Elsevier Ltd. All rights reserved.
1161
1161
1162
1162
1162
1163
1163
1164
1164
1164
1164
1165
1165
1165
1165
H. Batih, C. Sorapipatana / Renewable and Sustainable Energy Reviews 57 (2016) 11601173
1161
3.5.
Estimation of electrical energy saving potential of EE programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1166
3.6.
Economic evaluation of EE programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1166
3.7.
Estimation of avoided CO2 emission by electrical energy efciency improvement of the household sector . . . . . . . . . . . . . . . . . . . . . 1166
4. Result . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1166
4.1.
Urban households' electricity consumption in Indonesia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1166
4.2.
Total electrical energy saving potential and total CO2 emission reduction of the proposed energy efciency improvement (EE)
programs in the household sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1169
4.3.
Comparison between the investment cost of electrical energy savings and the long-run marginal cost of new
power plant development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1169
4.4.
Benet of EE programs to households' electricity expenditures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1170
5. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1170
5.1.
Characteristics of electrical energy consumption and potentials of saving in urban areas of Indonesia . . . . . . . . . . . . . . . . . . . . . . . . . . 1170
5.2.
Economic feasibility of electrical energy saving programs as compared to new power plants development. . . . . . . . . . . . . . . . . . . . . . 1171
5.3.
Electrical energy price and subsidy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1171
5.4.
Environmental Impact and Sustainability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1171
6. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1171
6.1.
Policy recommendation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1172
Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1172
Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1172
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1173
1. Introduction
Indonesian electricity demand has experienced a relatively
high growth rate for the past few years. Total electricity demand
was 107.03 TW h in 2005 and 157.99 TW h in 2011, amounting to a
growth rate of 6.7% per year [1]. In order to meet this high
demand, the Government of Indonesia (GOI) directs the National
Electricity Company (Perusahaan Listrik Negara, PLN) to develop a
medium-term power development plan covering a 10-year time
horizon period ofcially called the Electrical Power Supply Business Plan (Rencana Umum Penyediaan Tenaga Listrik, RUPTL). This
plan species the number and type of power plants for its
expansion plan, but does not embrace any energy conservation
measures to reduce electricity demands in the long term [2]. At
present (2013), a total installed capacity of power plants was
50,990 MW. Out of that amount 41,044 MW or 80.5% was the fossil
power plant while 9,945 MW or 19.5% was renewable power plant.
Those installed capacity generate a total amount of 216,185 GW h
of electricity. Where 189,56 GW h or 87.7% was generated from
fossil energy sources (oil 9.6%, coal 51.1%, and gas 27.0%) and the
percent share of electricity generated by renewable energy was
12.3% (hydro 7.8%, geothermal 4.4%, and biomass 0.1%) or a total of
26,618 GW h [3]. This plan intends to meet the projected demand
of electricity at the least cost for those fossil power plants. In 2030,
by the RUPTL plan, a total installed capacity will be 209,928 MW
corresponds to 204,828 GW h of electricity production comprises
of 890,212 GW h or 86.1% generated from fossil energy sources (oil
1.1%, gas 20.0%, and coal 64.9%) and 144,059 GW h or 13.9% (hydro
3.0%, geothermal 10.9%, and biomass 0.1%) generated by renewable
energy sources.
This is conventional supply side approach in most of developing countries and it generally results in exploiting more fossil
resources and emits more pollutants, which in turn it accelerates
to resources depletion and deteriorating environment. This leads
to question of sustainability due to fast economic development. On
the other hand, the utilization of electrical energy in these countries is frequently inefcient and there is still plenty of room to
improve it, particular for the consumption at end-users. Consequently, an attempt to improve efciency of electrical consumption by the demand side management is worth to investigate
whether it could avoided or delay construction of new power
plants or not. In addition, this approach would be able to abate
environmental impacts and help to extend depletion of nite fossil
resource in a long term. However, such approach to improve

electrical energy efciency, it entails with data in details and
understanding of the characteristics of electrical energy consumption of end users, which are necessary for assessing energy
saving potentials and economic feasibility. Unfortunately, such
information is scarce or even non-existent in most of developing
countries like Indonesia. To resolve this problem, this study was
initiated and conducted a primary survey to investigate characteristics of electrical energy consumption of the household
sector in urban areas in Indonesia. Its results were then used to
identify potential areas for electrical energy savings and to evaluate their economic viability.
By objectives of this study it intents to delineate the electrical
energy consumption characteristics of household sector in urban
areas in Indonesia and secondly, it is to assess the potential for
electrical energy saving of the Indonesian household sector in
urban areas. The assessment of the electrical energy saving
potential is done for the period of 20122030. The results were
then adopted to assess its investment costs of energy conservation
approach and compare with the option of new power plants
construction.
The overview of contents is as follow: Section 2 of this paper
provides the data collection method which covers survey sites,
sample size of respondents and the method for determining the
income class of the respondents. Section 3 describes methods for
calculating households electrical energy consumption and the
projection of electrical energy demand of the household sector for
Business as Usual (BAU) case and for Energy Efciency (EE) case,
estimation of energy saving potentials and the economic evaluation of targeted Energy Efciency (EE) programs, and analysis of
the impact of EE programs to the electricity planning particularly
for household sector. Sections 4 and 5 present results and discussion of this study, respectively. Finally, Section 6 provides the
conclusion and policy recommendations.
2. Data collection
Indonesia is the fourth most populous nation in the world. Its
population spreads over a large archipelago of more than 6000
inhabited islands. The archipelago consists of 17,508 islands located between latitudes 6N and 11S, and longitudes 95E and 141E
[4]. As reported in the 2010 census, conducted by Statistics
1162
Indonesia (Badan Pusat Statistik, BPS), the population of Indonesia

was 237.5 million with 80.5% of the population living in Java,
Sumatra, and Bali islands [5,6]. 54.1% of Indonesian households are
in urban areas. The urbanization levels vary among the islands.
Population in urban areas of Sumatra, Java, and Bali islands, which
are the three most populated islands, accounted for 45.2%, 69.4%,
and 64.8% of each islands total population, respectively [6].
As mentioned in the previous section, to develop an energy
conservation scheme in household sector, detailed data of electricity consumption characteristics is needed. This data will help in
identifying in what appliances the energy efciency improvement
should be prioritized. By knowing this data, one can calculate
energy saving, additional investment, and benet of the proposed
scheme. Unfortunately, to the best of our knowledge, the Government of Indonesia (GOI) does not have that kind of data.
Therefore, a primary survey to collect the data on characteristics of
household electricity consumption is needed.
Generally, a demographic survey for techno-economic census it
consumes a large amount of manpower, budget and time consuming. The stratied random sampling technique is commonly
used in this kind of survey. By concept, this method involves with
classication of population strata with similar socio-economic
attribution, and determination of an appropriate sampling size
(amounts of collecting data versus reliability and accuracy of a
results) for a survey. In practice, attribution, relevant variables and
their variation of population in a target group are unknown, a pilot
survey is necessary to be taken in order to identify these mentioned factors and to estimate minimum sampling size for the
actual survey. After knowing all necessary information, the second
round of survey (the actual one) is then undertaken. Details of the
survey method will be elaborately discussed as below:
2.1. Selection of survey sites
Since more than 80% of Indonesian population live in Java,
Sumatra, and Bali Island [6],we decided to conduct a survey of
electrical energy consumption by urban households on these
islands. Due to budget and time limitation, only seven highly
populated cities i.e. Denpasar, Surabaya, Yogyakarta, Jakarta, Bandung, Bandar Lampung, and Palembang, were selected for the
survey. Population of these cities is almost a half of the total
population living in urban areas of these three islands. Fig. 1
depicts the location of the surveyed cities. Table 1 presents the
population in urban areas of Java, Sumatra, and Bali, and the
population of each city.
then chosen into questionnaire for a pilot survey. Evidences from

the result of pilot survey were then used to justify whether a
variable should be included, dropped, or even added into questionnaire for the second round of the survey. A new version of
revised questionnaire was then used for the nal (actual) survey.
Types of appliances in the questionnaire of the actual survey were
thus able to identity by obtaining from the result of the pilot
survey.
2.3. The pilot survey and sample size
The method of stratied random sampling was used in this
survey. In stratied random sampling, all members of population
in each stratum are categorized into a homogenous group based
on their similar characteristics dened by the purpose of samplings. Then samples were randomly selected from each stratum.
There are certain parameters that need to be considered for the
determination of sample size i.e. types of variables to be measured,
error estimation and estimation of variance. Cochran [12]
addressed two factors for error estimation: (a) level of precision
and (b) level of condence. In this study, 5% level of precision of
estimated values was used, and a level of condence of 95% was
selected for sample sizes determination in this study. Cochran [12]
showed that sampling size for discrete random variables (categorical variables such as income classes, types of equipment, etc.) can
be represented by
ncategorical
2.2. Attribution of questionnaire
Bandung
Z 2 s2
Ue Ue2
where m is the true mean of a population, and s is the estimated

standard deviation of population.
A critical issue of sample size determination is variance estimation of the primary variable of interest. The more
Palembang
Badar Lampung
Jakarta
where Z is a standard normalized value which corresponds to the

selected condence level. p, is an estimated proportion of an
attributed variable presents in the population and q is 1 p. e is a
specied level of precision of the estimated true mean value which
is normally specied as a percentage of the true mean of a
population. Researchers often use a value of 0.5 as the estimated
proportion value for p to estimate a dichotomous (proportional)
variable [13]. Since a proportion of 0.5 indicates the maximum
possible variability in a dichotomous population; this proportion
will result in the maximization of variance which also produces
the maximum sample size. In this study, we adopted the proportion of 0.5 for our estimation to derive an appropriate sampling
size of categorical random variables in Eq. (1) for our survey.
For continuous random variables, a sampling size, ncontinuous, to
estimate a true mean value of a population can be given by
ncontinuous
List of candidate variables to be included in the questionnaire

was rstly determined through literature and by observation of
common practices in Indonesian households. List of variables was
Z 2 pq
;
e2
Surabaya
Yogyakarta
Denpasar
Fig. 1. Survey sites.
1163
Table 1
Population of seven surveyed cities [711].
Islands
Sumatra
Population Population in urban area Population in urban area as % of the total

population
(Million)
(Million)
(%)
49.89
22.54
45.18
Java
137.78
95.60
69.38
Bali
Total
3.63
191.30
2.36
120.49
64.80
62.98
Surveyed cities on each islands Population in each city

(Million)
Bandar Lampung
Palembang
Greater Jakarta
Surabaya Metropolitan
Yogyakarta
Greater Bandung
Denpasar
0.88
1.46
28.02
9.12
0.39
7.89
0.79
48.54
Remark:
The total population of Sumatra, Java, and Bali is 80.55% of the total population of Indonesia.
The urban population of the seven surveyed cities accounted for 40.28 % of the total urban population of Sumatra, Java, and Bali.
II
III
IV
: mean
: standard deviation
Income per capita
Fig. 2. Classication of income groups in this study.
heterogeneous a population, the larger the sample size is needed

to achieve a given level of precision. Generally, the variance of a
variable of interest is unknown. Dealing with this problem, we
took sampling in two steps. The variance of the rst samplings is
used to determine a sample size for the second run of samplings in
order to obtain a more accurate result.
A pilot survey of 29 households was conducted in Java and
Sumatra. The expenditure on electrical energy consumption1 was
chosen as a basic variable to calculate sample size because its
mean and standard deviation give the largest sample size as
compared to other continuous variables. The mean and standard
deviation of this variable were found to be 162,758 IDR or USD
17.13 and 134,984 IDR or USD 14.21 (USD 1 9500 IDR, 2012
currency exchange rate), respectively. This standard deviation was
then assumed to be the estimated standard deviation of populations, s. By using Eq. (2), the sample size for 95% condence level
(Z 1.96) and the level of precision 5% (e 0.05) is
ncontinuous
Z 2 s2
U e U e
2
1:96 U 134; 984
Z 2 s2
264:
Because of some random variables are discrete, a check of

sample size by using Eq. (1) is needed, the sample size for the
discrete random variable at the same level of condence and
precision and assuming that maximum variability exists in the
population (i.e. p 0.5), is
ncategorical
The actual survey was conducted between November, 2011 and

January, 2012. The number of sample size calculated in the previous section (384 households) is the suggested minimum number
of respondents needed to meet the above mentioned level of
condence and precision (95% and 5%, respectively). Therefore,
due to the potential of invalid response, a larger number of
respondents are needed. We decided to collect 600 respondents to
make allowance for that mentioned error. The numbers of
respondent samplings were proportionally allocated for each surveyed city in accordance with the number of population in each
city. The survey was conducted by interviewing respondents in
their home. This is to help respondents and surveyors to ll the
questionnaire correctly by directly checking to the appliances. Out
of 600 respondents, only 474 of them, or 79% of respondents, were
considered to be acceptable.
2 U Ue2
2 U 162; 758 U 0:052
2.4. Actual survey
Z 2 pq 1:962 U0:5 U 0:5
384:
e2
0:052
Since the sample size for categorical (discrete) variables is larger than the sample size of continuous variables, 384 is adopted as
the minimum suggested sample size in this study.
1
The other variables were expenditure for: transportation; food drinks, and
tobacco; personal care; dressing and foot wear; housing; health; and education.
2.5. Income class determination

Income is one of the strongest inuencing parameter on
household's electrical energy consumption. As incomes rise, there
is a decline in the percentages of consumers who consume traditional fuels [13]. Therefore, income classes were chosen as the key
parameter to stratify consumer classes. Unfortunately, there are no
ofcial statistical data of Indonesian households based on income
classes. This research used qualitative judgment to determine
income levels. The criteria for income classes of respondents
include oor-space of houses, vehicle ownerships, and dweller
areas. The distribution of incomes was assumed to be a normal
distribution. Incomes of respondents were then categorized into
four classes of income groups. Each class was segregated by one
standard deviation. Fig. 2 illustrates the method to classify
households in each income group.
1164
Table 2
Projection of population, household size, number of household, and percentage of households in urban areas.
Year
Population
(Thousands)
Household size
(Person/hh)
Household
(Thousands)
% household in urban areas

(%)
Urban households
(Thousands)
Rural households
(Thousands)
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
240,730
243,860
247,030
250,241
253,495
256,790
260,128
263,510
266,936
270,406
273,921
277,482
281,089
284,743
288,445
292,195
295,993
299,841
303,739
307,688
4.00
3.98
3.95
3.93
3.91
3.89
3.86
3.84
3.82
3.80
3.77
3.75
3.73
3.71
3.69
3.67
3.64
3.62
3.60
3.58
60,182
61,321
62,480
63,662
64,866
66,092
67,342
68,615
69,913
71,235
72,582
73,954
75,353
76,778
78,229
79,709
81,216
82,752
84,317
85,911
55
56
57
58
59
59
60
61
62
63
64
64
65
66
66
67
68
68
69
70
33,013
34,226
35,470
36,746
38,001
39,287
40,604
41,954
43,335
44,698
46,094
47,521
48,981
50,474
51,951
53,461
55,004
56,581
58,192
59,837
27,169
27,095
27,010
26,916
26,865
26,805
26,737
26,662
26,578
26,536
26,488
26,433
26,372
26,304
26,278
26,248
26,212
26,171
26,125
26,074
3. Methods for estimating electrical energy saving potentials

and economic assessment
3.1. Households' electrical energy consumption
3.2. The ratio of electrical energy expenditure to income
Electrical energy in households is generally consumed for

lighting, space cooling (air conditioning, refrigerating, and electric
fan), cooking, entertainment (television, computer and laptop),
laundry, water supply, and other purposes. A total electrical energy
consumption of all households, EH, is dened as [14]
EH
n X
m
X
Ni;j UP i;j UM i;j U I i;j ;
i1j1
where N is a number of households, P is a number of appliances

per household (unit/household), M is a length of time in using an
appliance (h), and I is a rated power of an appliance (W), i is a type
of equipment, and j is an income class.
Because of on/off cycle operations and kinds of electric thermostat controls for some types of household appliances such as air
conditionings and refrigerators, the rated W h of appliances is not
always equal to the actual W h in real use [15]. Therefore, Eq. (3)
can be rewritten as
EH
n X
m
X
Ni;j UP i;j UM i;j U I i;j URi;j ;
study. For refrigerators and other appliances R is assumed to be

0.36 and 1 [15], respectively.
i1j1
where R is a ratio of the actual W h consumed to a nominal rated

W h of an appliance. Utama and Gheewala [16] used a dynamic
building analysis software ECOTECT to estimate the electrical
energy consumption for air conditioning system units in one-oor
houses in Indonesia. The result showed that for those common
commercial air conditioning appliances with cooling capacity of
70009000 Btu/h having the coefcient of performance (COP) of
3.2 (equivalent to 732.5 W), operating time of 9 h/day, and temperature setting around 24 C, the electrical energy consumption
was found to be 181.73 kWh/month. Technically, the maximum
possible monthly electricity consumption of that type of AC is
197.77 kWh/month ( 732.5 W 9 hours/day 30 day/month ). By
dividing actual electrical energy consumption with the maximum
possible monthly electrical energy consumption, R, shown in the
Eq. (4) for AC can be calculated. In this case, it was found that R for
AC equal to 0.92. This gure was then adopted for R of AC in this
A ratio of electricity expenditure to income can be seen as an

indicator to measure vulnerability of households to the electrical
energy price shock, and also to measure behavior of consumers in
electrical energy consumption. The higher the ratio of electrical
energy expenditure to income, the more vulnerable to a price
shock that household group is exposed to. In addition, a relatively
more expensive electricity price is likely to pressure consumers to
conserve more electrical energy. In this study, ratios of electrical
energy expenditure by appliance types to incomes of each income
class were also calculated to see what kinds of appliances
accounted for the biggest share of energy expenditure.
3.3. Projection of households' electrical energy consumption under
the business as usual case (BAU)
The number of households (N) and number of appliances per
household (P), in Eq. (4) were projected for each year. Electric
rated power per appliance (I) was assumed to be constant. Then, a
total electricity consumption of all households under the business
as usual (BAU) case was calculated. The projected number of
households in each year was achieved by taking a projected
population divides by a predicted average size of households in the
same year. Historical data of Indonesian population from 1990 to
2008 showed an increase in population with a growth rate of 1.3%
per year and the size of household declined at 0.58% per year
[17,18]. This study adopted 1.3% for the population growth rate and
0.58% declining rate per year of the household size for projecting
number of households. A projected number of urban household
was calculated by multiplying projected total household in the
country with the percent share of households in urban areas in the
same year. This study used ofcial data of the percent share of
households in urban areas from Statistics Indonesia [6]. Table 2
showed the projected population, household sizes, numbers of
households, and percent shares of households in urban areas from
2011 to 2030.
3.4. Electrical energy consumption by urban households under

energy efciency improvement case (EE case)
100%
Observed data
90%
Logistic function
80%
The energy efciency scenario was assumed to be the introduction of more-efcient, end-use technologies to replace conventional ones. Selection of specic EE programs would depend on
results of the survey. Selective EE programs would then focus on
the improvement of those end-use appliances that have relatively
high electrical energy consumption. The result of the survey
revealed that appliances for lighting (lamps), entertainment (television), and space cooling (refrigerator and air conditioning systems) are the appliances that consume most of electrical energy in
the household sector (see Section 4.1). Therefore, this study
recommends the adoption of high efciency lamps, television sets
(TVs), refrigerators, and air conditioning systems (ACs). We
assumed that the number of appliances per household (P) for each
type in Eq. (4) are constant throughout the planning period
horizons.
3.4.1. Efcient lighting technology program (EE_lighting)
In this program, called EE_lighting, we proposed the utilization
of Compact Fluorescent Lamps (CFL) to replace incandescent
lamps, Light Emitting Diode (LED) lamps to replace CFLs, and T5
Fluorescent Lamps (FL) to replace T9 and T12 orescent lamps.
Table 3 gives the initial investment cost and life time for conventional and high efcient lighting technology for comparison.
Percent shares of accumulated high efciency appliances of
annual total stock implemented from 2012 to 2030 are assumed to
follow a S-curve. The percent share of implemented high efciency
appliances in households varies as a function of time (year) and is
denoted by PH (t) as shown in Eq. (5). Since Indonesia has no prior
experiences in implementing any kinds of new high efciency
appliance household replacement programs, the penetration rate
is unknown. This study adopted Thailand's experience for the
penetration rate of new high efciency appliances substituting for
traditional inefcient appliances in residential sector, as reported
in [22], for the rate of PH(t).
A logistic function, PH(t) in Eq. (5), was tted to actual data of
the penetration rate of high efciency appliances for energy saving
programs implemented in Thailands residential sector,
PHt
1
1aUebUt
where t is time (years), while a and b are constant parameters

representing the characteristics of logistics function. By using nonlinear regression technique, it was found that a 46.51 and
Table 3
Comparison between conventional and high efcient lighting technology [1921].
Measures
Replace
Replace
Replace
Replace
Replace
Replace
Replace
Replace
Replace
Replace
Incandescent 25 W by CFL 8 W
T9 FL 18 W by T5 FL 14 W
CFL 8 W by LED 3 W
CFL 11 W by LED 5 W
CFL 18 W by LED 7 W
Initial investment cost (USD)
Life time (h)
Con
Eff
Con
Eff
0.47
0.58
0.63
1.37
1.37
1.68
1.68
2.74
2.84
3.47
2.74
2.84
3.47
3.49
3.49
4.05
4.05
4.58n
7.33n
11.00n
1500
1500
1500
10,000
10,000
10,000
10,000
8000
8000
8000
8000
8000
8000
25,000
25,000
25,000
25,000
50,000
50,000
50,000
Remark:
Con: conventional.
Eff: efcient.
n
Initial investment cost for LED lamps is a forecasted price in 2020.
1165
PH (t ) =
70%
60%
1
1 + 46.51 e 0.31t
50%
40%
30%
20%
10%
0%
1
9 10 11 12 13 14 15 16 17 18 19 20
Year
Fig. 3. Thailand's average penetration rate of high efciency appliances in the

residential sector [21].
Table 4
Comparison between conventional and high efciency TV [23].
Measure
Approximate screen
size (in.)
Average
power
Input (W)
Conn
Replace CRT
by LED
Replace LCD
by LED
Replace PPD
by LED
Initial investment cost

(USD)
Eff Con
Eff
Con Eff
189.47
21
86.05 30
10
30
32
97.02 60 300.00 338.95 20
30
42
132.96 85
124.21
Life
time
(years)
468.42
615.79
15
30
Remark:
Con: conventional.
Eff: high efciency.
n
Data obtained from survey.
b0.31; it generated PH(t) that nearly duplicates Thailand's

penetration rates of new high-efcient appliances implementation
for the residential sector (coefcient of determination, R2 0.996)
(see Fig. 3).
3.4.2. Efcient television technology program (EE_TV)
In this program, called EE_TV, Light Emitting Diode (LED) TVs
are proposed to replace Cathode Ray Tube (CRT), Liquid Crystal
Display (LCD), and Plasma Display Panel (PPD) TVs. Table 4 provides gures for initial investment costs and life times of conventional and electricity efcient LED television technology for
comparison.
The yearly percent share of high efciency appliances in the
total stock during the period of implementation is also assumed to
follow a S-curve as shown in Fig. 3.
3.4.3. Efcient space cooling system program (EE_Refrig and EE_AC)
There are two types of appliances considered in this program
i.e. appliances for food preservation (refrigerator) and appliances
for comfort space cooling (air conditioning system). In the case of
efcient food preservation program, called EE_Refrig, higher efcient refrigerators are proposed to replace conventional inefcient
ones. Table 5 compares the initial investment costs and the power
consumption of various refrigerator capacities between conventional and high efciency ones. Additionally, it is assumed that the
expected life of both refrigerator types is the same at 13 years. [24]
In the case of electrical energy improvement program for space
cooling for comfortable living, called EE_AC, we distinguish Air
Conditioning systems (AC) into two groups, i.e. conventional and
high efciency group. Each group is categorized by its Coefcient
1166
where ALCC is the annualized initial investment cost of appliances

(USD/year), and ELSAV is the annual electrical energy savings
(kW h/year), thus the unit of CSE is USD/kWh.
Without electrical energy saving program, construction of new
power plants would be necessary to meet the expected rising
demands. The cost of supply side electrical power generation
expansion is normally calculated by a long run marginal cost
(LRMC) of newly developed power plants. However, because of the
complexities involved in the calculation of LRMC, Shresta [27]
suggested the use of Long Run Average Cost (LRAC) instead of
LRMC for simplicity. In this study we adopted LRAC to estimate the
cost of electricity generation expansion, as given by Eq. (8),
Table 5
Comparison between conventional and high efciency refrigerator [25].
Type
Approximate
capacity (l)
1 door
160
2 doors
210
Side by side doors 530
Average power
input (W)
Initial investment
cost (USD)
Conn
Eff
Con
Eff
85.83
124.23
184.89
63
97
110
142.11
265.26
842.11
157.89
294.74
1052.63
Remark:
Con: conventional
Eff:high efciency
n
Data obtained from survey.
LRAC
Cooling
capacity
(Btu/h)
Conventional AC
Average
power
input (W)
Initial investment cost

(USD)
Average
power
input (W)
Initial investment cost (USD)
5000
7000
9000
12,000
18,000
523
732
942
1256
1884
254.74
289.47
305.26
395.79
561.05
386
540
694
925
1388
277.89
335.79
341.05
456.84
631.58
High-efciency AC
of Performance (COP). Conventional AC systems have COP around

2.8, while high-efcient AC systems have COP around 3.8. The
result of this survey revealed that the share of air conditioning
units in Indonesia consists of about 80% conventional and only 20%
high-efciency AC systems. These gures are used for the beginning period of the implementation of the EE_AC program. Table 6
compares the initial investment cost and electricity power consumption between conventional and high-efcient AC systems. In
addition, we assumed that the expected life of both conventional
and high-efcient one are the same at 12 years. [24]
Similar to the appliances in the EE_lighting and EE_TV program, the percent share of accumulated high efciency appliances
in annual total stock is assumed to follow a S-curve during the
period of implementation from 2012 to 2030. A share of 20% for
the high efciency ACs was assumed to be the initial gure of the
high efciency ACs in the stock at the beginning of the EE_AC
program implementation for PH(t) in Fig. 3.
3.5. Estimation of electrical energy saving potential of EE programs
Electrical energy saving potential for households (ELSAV) was
calculated as
n X
m
X
N i;j UP i;j UM i;j U PH i;j UI BAU
i;j I EE i;j
i1j1
where PH is the percent share of accumulated high efciency

appliances in households, IBAU and IEE are the rates of electricity
power consumption for conventional and high efciency appliances, respectively.
3.6. Economic evaluation of EE programs
An annual investment cost for each energy saving programs
can be determined by the term Cost of Saved Energy (CSE) [14]
which can be calculated by
CSE
ALCC EE ALCC BAU

;
ELSAV
t1
Table 6
Comparison between conventional and high efciency AC systems [26].
ELSAV
TC
;
T

P
Et =1 rt
where TC is the present value of a total power generation cost of

new power plants for the planning horizon, Et is generated electrical energy in year t, r is the discount rate, and T is the year of
planning horizon. Data for LRAC calculation were obtained from
PLN's and Ministry of Energy and Mineral Resources (MEMR)
ofcial publications [2,28,29]. 12% discount rate was used in this
study (about 2% greater than the long term average of the government bonds interest rate)[30,31].
The value of CSE derived from the saving programs to reduce
electrical demands was then compared to the value of LRAC
(LRMC) of the electrical power supply-side expansion. Value of CSE
lower than LRMC means that EE program is more cost-effective
than the supply-side option.
3.7. Estimation of avoided CO2 emission by electrical energy efciency improvement of the household sector
Based on information obtaining from RUPTL plan, the primary
energy mix, emission factors and total generated electricity can be
gured out. In addition, by knowing amounts and types of primary
energy and their emission factors, one can estimate a total CO2
emission. Then an overall average of CO2 emission per kWh of
generated electricity can also been calculated. Since amounts of
electrical energy savings by efciency improvement in the
household sector can be estimated by the method given in Section
3.5, amounts of avoided CO2 emission by electrical energy efciency improvement programs can be estimated. Obtained result
is given in Section 4.2.
4. Result
4.1. Urban households' electricity consumption in Indonesia
Out of 600 respondents, only 474 were considered to be valid.
Table 7 showed number of respondents, average per capita
monthly income, and the corresponding standard deviation of the
average per capita monthly income for each city.
By using the data in Table 7, the income classes were classied
by applying the scheme shown in Fig. 2, and the result is displayed
in Table 8.
This survey's questionnaire included types of appliances, rates
of electric power consumption (W), length of operating hours
(h) and unit of appliances in each type per household (units per
household) for various income classes; the results are presented in
the Appendix. By using the survey's results as shown in the
Appendix, the average monthly households' electrical energy
consumption for each income class was then calculated and the
result is presented in Fig. 4.
Table 7
Number of respondents in each city.
City
Denpasar
Surabaya
Yogyakarta
Bandung
Jakarta
Bandar
Lampung
Palembang
Total
Average
Standard
deviation
()
(Household)
(Household)
Monthly
average per
capita
income (l)
(Rp)
30
100
95
116
123
66
29
78
70
87
106
51
989,696
963,350
928,801
966,211
1,174,109
659,617
569,862
401,258
656,177
668,530
668,530
317,979
70
600
53
474
990,521
459,145
977,874
565,523
Number of
respondents
Valid respondents' survey

data
(Rp)
kWh/month per household
Appliances
Lighting
Space cooling
Air conditioning
(AC)
Refrigerating
Electric fan
Cooking
Television (TV)
Income class Monthly per capita

income (Rp)
o 412,351
412,351977,873
977,8741,543,397
41,543,397
I
II
III
IV
Table 9
Monthly urban household's estimated electrical energy consumption in each
income class and the corresponding percent share of each type of appliances to the
total electrical energy consumption.
Entertainment
Table 8
Classication of income classes.
Percentage of households in each
income class (%)
18.2
28.2
42.7
10.8
Personal computer
(PC) and Laptop
Laundry
Electric iron
Washing machine
Water supply
Other
Total
300
248.1
1167
Income class
I
(kW h)
II
(kW h)
III
(kW h)
IV
(kW h)
20.22
(19.98%)
22.80
(22.53%)
2.67 (2.64%)
27.07
(16.66%)
57.31
(35.28%)
33.64
(20.71%)
15.53
(9.56%)
8.14 (5.01%)
17.70
(10.89%)
37.38
(23.00%)
23.55
(14.49%)
13.83
(8.51%)
13.67
(8.41%)
9.02 (5.55%)
4.65 (2.86%)
3.48 (2.14%)
5.84 (3.60%)
162.46
(100.00%)
27.55
(13.83%)
72.96
(41.66%)
58.51
(29.38%)
18.42
(9.25%)
6.03 (3.03%)
23.29
(11.70%)
35.79
(17.97)
22.18
(11.14%)
13.61
(6.83%)
13.93 (6.99)
46.56
(18.77%)
108.32
(43.66)
79.46
(32.03%)
19.13 (7.71%)
12.53
(12.38%)
7.60 (7.51%)
11.78
(11.64%)
23.89
(23.60%)
17.58
(17.37%)
6.31 (6.24%)
11.28
(11.15%)
8.37 (8.27%)
2.91 (2.87%)
4.53 (4.48%)
6.70 (6.62%)
101.21
(100.00%)
7.71 (3.87%)
6.22 (3.12%)
6.62 3.33%
8.99 4.51%
199.14
(100.00%)
9.73 (3.92%)
18.45
(7.44%)
42.85
(17.28%)
32.69
(13.18%)
10.16 (4.10%)
18.35
(7.40%)
9.94 (4.01%)
8.41 (3.39%)
6.39 (2.58%)
7.16 (2.89%)
248.07
(100.00%)
250
199.1
200
162.5
150
101.2
100
50
0
I
II
III
IV
Income class
Fig. 4. Average monthly household electrical energy consumption by income class.
As expected, average monthly electrical energy consumption

was found to be higher for a higher income group. The uses of
electrical energy are for: lighting, space cooling (air conditioning,
refrigerating, and electric fan), cooking, entertainment (television,
computer and laptop), laundry, water supply, and others. The
details are shown in Table 9.
Appliances for lighting, space cooling and entertainment are
the top three equipment that consume the largest electrical energy
for all income classes. But the ranking of consumption of electrical
energy by types of equipment in each income class differs. For
income class I (the lowest income group), lighting (lamps) contributes the largest share of electrical energy consumption, follow
by entertainment (TVs), and space cooling (refrigerators). While
for the electricity consumption of income class II, III and IV (the
highest income group), space cooling for living comfort, lighting
and entertainment are the top three utilizations of electrical
energy consumption. The largest consumption is contributed by
air conditioning systems (ACs), lamps, and TVs, respectively. This is
because low-income household cannot afford the high initial costs
of air conditioning systems and the subsequent high electrical bill.
Indonesia's daily electrical power demand characteristics is as
follows: the lowest off-peak demand occurs between 22.00 p.m.
and 09.00 a.m.; the intermediate demand occurs between 09.00 a.
m. to 17.00 p.m.; and the peak demand occurs between 17.00 p.m.
to 22.00 p.m. (see Fig. 5) [32].This is the time when most people
return home from work and turn on lamps, air conditioning
systems and TVs. Based on this argument, lamps, ACs, TVs and
refrigerators are likely the appliances that contribute the most to
the peak demand of a day. Table 10 shows the contribution of the
estimated power load demands of the mentioned appliances by
urban households in Indonesia as a percentage of the maximum
peak demand of the year 2011.
By multiplying the estimated monthly electrical energy consumption for each type of appliances as shown in Table 9 with an
average electrical energy tariff for each income class, one can
calculate monthly electricity expenses by types of appliances for
each income class. Table 11 shows percentages of relative electrical
energy expense to income for each income class. It was found that
percentage of electrical energy expense to income tends to be
smaller for a higher income class. Even though this is not the case
for income class I and II where it is lower for class I, it is, however,
not statistically signicant by t test at 95% condent level. But for
every other income class comparisons, this trend holds true.
Based on the data from Table 9 through 11, in order to reduce
electrical energy usage and peak power demand most effectively,
energy efciency improvement programs to replace current
appliances with high energy efcient lighting (lamps), entertainment (TV), and space cooling (refrigerator and AC) appliances
should be initiated as this would lead to the largest potential
saving of electrical energy.
By comparing the average electrical generation cost per unit, as
reported in [33] and the retail electrical selling prices to urban
households, as reported in [34], the amount of subsidy per kW h of
electrical energy for each income class could be estimated. This
study found that subsidy per kW h of electricity sold to income
class I, II, III, and IV are USD 0.056, 0.056, 0.053, and 0.049 (USD
1 9500 IDR, 2012 currency exchange rate) per kW h, respectively. Although the subsidy per kWh are generally smaller for
higher income classes, they actually receive a larger total subsidy
since they consumes more units of electrical energy per household
(see Fig. 6).
1168
Lowest
demand
Peak demand
Intermediate demand
Lowest demand
Fig. 5. A typical daily load demand of JavaBali electric system in Indonesia, on 16 November, 2011[31].
Income
class
I
II
III
IV
Total
Lamps
ACs
TVs
Refrigerator
Total
(MW)
(MW)
(MW)
(MW)
(MW)
647.84
(2.01%)
1363.41
(4.24%)
2005.76
(6.24%)
763.21
(2.37%)
4780.22
(14.87%)
113.48
(0.35%)
1686.90
(5.25%)
4998.57
(15.55%)
1926.17
(5.99%)
8725.13
(27.14%)
576.63
(1.79%)
1,099.47
(3.42%)
1936.25
(6.02%)
686.97
(2.14%)
4299.32
(13.37%)
387.26 (1.2%)
1725.21
(5.37%)
743.92 (2.31%)
4893.70
(15.22%)
1335.92 (4.16%) 10,276.50
(31.96%)
350.88 (1.09%)
3727.24
(11.59%)
2817.98 (8.76%) 20,622.65
(64.14%)
Remark:
The maximum peak load in year 2011 is 32151.5 MW. The total number of urban
households in Indonesia in 2011 is 33,013 thousand households.
Table 11
Percentage of electrical energy expense to estimated income in each income class,
by types of appliances.
Appliances
Lighting
Space cooling
Air conditioning
Refrigerating
Electric fan
Cooking
Entertainment
Television
PC and laptop
Laundry
Electric iron
Washing machine
Water supply
Other
Total
Income class
I
II
III
IV
0.68%
0.77%
0.09%
0.42%
0.25%
0.40%
0.80%
0.59%
0.21%
0.38%
0.28%
0.10%
0.15%
0.22%
3.40%
0.60%
1.28%
0.75%
0.35%
0.18%
0.39%
0.83%
0.52%
0.31%
0.30%
0.20%
0.10%
0.08%
0.13%
3.61%
0.41%
1.22%
0.86%
0.27%
0.09%
0.34%
0.53%
0.33%
0.20%
0.21%
0.11%
0.09%
0.10%
0.13%
2.93%
0.42%
0.99%
0.72%
0.17%
0.09%
0.17%
0.39%
0.30%
0.09%
0.17%
0.09%
0.08%
0.06%
0.07%
2.26%
Remarks:
Estimated monthly household income (IDR)
Income class I: 2,046,443
Income class III: 4,841,743
Income class II: 3,051,792
Income class IV: 8,220,800
35.00
USD/month per household
Table 10
Estimated daily power demands of selected appliances of households in urban
areas and their percent shares of the maximum peak demand in 2011.
30.00
12.19
25.00
10.57
20.00
9.20
15.00
10.00
5.00
5.65
19.58
14.94
11.61
7.32
0.00
I
II
III
IV
Incom e class
Electricity expense
Subsidy
Fig. 6. Monthly estimated electrical energy expenses as paid per household and the
subsidy for each income class.
Appendix shows types of lighting technologies which are

commonly used in urban households in Indonesia, they are:
Compact Fluorescent Lamp (CFL), Incandescent Lamp, T9 FL, and
T12 FL. More units of appliances per household are generally
deployed for higher income classes. Incandescent lamps are still
commonly used in all Indonesian urban households. Low rated
powers of incandescent lamps (5 W and 10 W) are commonly
found in all income groups; they are generally used for small
compartments' lighting, such as in toilets, storages, etc while more
units of appliances with higher rated powers of incandescent
lamps (15, 25, 40 W) are commonly found in higher income
groups.
The type of air conditioning systems (AC), commonly found in
urban households in Indonesia, is the split type with the capacity
of 0.5, 0.75, and 1 hp (1 hp0.746 kW), as shown in the Appendix.
Unlike lighting, air conditioning systems are not commonly found
in the lowest income group (income class I). Units of the air conditioning systems with a capacity of 0.5 hp is the most common
type found for income classes II, III, and IV.
Interestingly, it is discovered that water pumps are ubiquitous
in all urban households regardless of income classes, but they
consume only a small share of electrical energy ranging from 2.1%
to 4.5% of the total electricity consumption per household and
higher income households tends to deploy more water pumps.
1169
Table12
Annual electrical energy saving potential (per household) (a) and (b) CO2 emission by implementing EE programs for household sector in 2020 and 2030.
Income class
I
II
III
IV
Total (Weighted average)
CO2 emission (million ton)

CO2 emission reduction (million
ton)
EE_Lighting
GWh (kW h/hh)
EE_TV
GWh (kW h/hh)
EE_Refrig
GWh (kW h/hh)
2020
170.66
(20.98)
262.20
(20.80)
531.64 (27.85)
2030
988.02 (90.72)
1583.48
(93.84)
3202.75
(125.35)
211.86 (43.89) 1311.37
(202.92)
1176.36
7085.63
(26.32)
(118.41)
2020
295.31
(36.30)
512.08 (40.62)
2030
1350.84
(124.04)
2,342.36
(138.81)
710.90 (37.25) 3251.82
(127.27)
252.74
1156.08
(52.35)
(178.89)
1771.02
8101.11
(39.62)
(135.39)
2020
105.31
(12.95)
206.87
(16.41)
389.51
(20.41)
99.23
(20.55)
800.91
(17.92)
Scenario
2020
2030
BAU
EE
80.44
74.45
5.98
102.00
84.10
17.90
This is due to the limited and poor municipal water supply infrastructure in most urban areas in Indonesia. This could potentially
be a serious problem in the future should the surface water
become contaminated.
2030
2020
481.72 (44.23) 53.00 (6.51)
2030
92.26 (8.47)
946.26 (56.08) 1,033.01 (81.95) 1798.42

(106.58)
1,781.72
2,720.30
4735.92
(69.73)
(142.53)
(185.36)
453.88 (70.23) 934.29 (193.54) 1626.56
(251.69)
3663.58
4740.60
8253.16
(61.23)
(106.06)
(137.93)
30,000
GWh
4.2. Total electrical energy saving potential and total CO2 emission
reduction of the proposed energy efciency improvement (EE) programs in the household sector
EE_AC
GWh (kW h/hh)
25,000
13.5%
20,000
26.1%
EE_Refrig
EE_Lighting
15,000
EE_TV
29.9%
EE_AC
10,000
Electricity saving potential was calculated by using Eq. (6).

Annual electricity savings depends on how large a number of high
efciency appliances accumulated for each type of appliances in
the stock in each year. The percent share of the accumulated high
efciency appliances in the stock of that type of equipment generally increases as a function of time (year) after implementing the
EE program and can be assumed to follow the S-curve function. As
mentioned before, it is assumed to be similar to that of Thailands
experiences (as shown in Fig. 3). Table 12 shows the total amount
of electrical energy saving potential (kW h) and the saving per
household (kW h/hh) for each income class in 2020 and in 2030. In
2030, EE program for the household sector can save electricity for
about 27.1 TW h or a 21.4% reduction as compared to that of BAU
case. It was also found that if none of EE programs is implemented,
CO2 emission per kW h will be 0.807 kg/kW h in 2030, The
implementation of EE program in the household sector will reduce
CO2 emission intensity to 0.805 kg/kW h, or a total CO2 emission is
reduced from 102.22 to 79.98 million ton CO2 in 2030 for this
sector, or it can avoid CO2 emission of about 22.02 million ton CO2
or a 21.6% reduction from the BAU case.
Clearly, as shown in Table 12, the electrical energy saving
potential, as well as CO2 reduction, becomes higher as the accumulated units of new high efcient appliances in the stock become
larger in later years. Since higher income classes generally possess
more units of equipment per household and with higher rated
power equipment, this leads to higher potentials of electric energy
saving per household for the higher income groups than the
lower one.
For the lowest income group, the potential of savings is from
those appliances which are larger in numbers, i.e. lamps and TVs.
The lowest income class I has higher potential of saving from
improving in the area of lighting (lamps) and entertainment (television sets) than that of space cooling (refrigerating and air
conditioning). In contrast, the higher income classes II, III and IV
have highest electrical energy saving potential from improving
5,000
30.5%
0
2020
2030
Fig. 7. Electrical energy savings due to appliance efciency improvement of each

type in 2020 and 2030.
space cooling (air conditioning systems) than from lighting

(lamps) and entertainment (TV). This is because the percent share
of electrical energy consumption for lighting (lamps) and entertainment (TV) are generally a smaller fraction in the high income
group comparing to the share of electrical energy consumption
from air conditioning systems.
Although, as per household, the electrical energy saving potential
is largest for the income class IV, as shown in Table 12; the largest
electrical energy saving potential among all groups is from the
income class III, due to a combination of a relatively high electricity
consumption per household (second only to income class IV) and
the largest number of households in this income class which in turn
leads to the highest saving potential (see Tables 8 and 12).
In 2030, total electrical energy savings from lighting (lamps),
entertainment (TVs), and space cooling (refrigerators and ACs) are
expected to be about 27,103 GW h, or a reduction of 21.4% from the
BAU case with the improvement of air conditioning systems' efciency accounting for 30.5%, the improvement of, TVs, lamps, and
refrigerators' efciency accounting for 29.9%, 26.1%, and 13.5% of
the total saving mentioned above, respectively (see Fig. 7).
4.3. Comparison between the investment cost of electrical energy
savings and the long-run marginal cost of new power plant
development
Eqs. (7) and (8) were used to calculate CSE and LRAC. As mentioned earlier, for simplicity, LRAC is frequently used as a proxy for
1170
0.080
Table 13
Percentage of electrical energy expenditure to household income for each income
class in 2011 and 2030.
0.070
0.060
USD/kWh
Income class Percentage of electricity expenses to income

0.050
LRMC
0.040
EE_lighting
0.030
Before EE programs
(2011)
At the end of implementation of EE

program (2030)
3.40%
3.61%
2.93%
2.26%
2.77%
2.96%
2.34%
1.75%
EE_TV
EE_AC
0.020
EE_Fridge
0.010
I
II
III
IV
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
0.000
Year
Fig. 8. Comparisons between CSE of energy saving programs and LRMC of power
plant expansions.
LRMC for the cost of supply-side electricity power plant expansion.

Fig. 8 shows that, the implemention of electrical energy saving
programs to reduce the electrical energy demand are more cost
effective than the supply-side expansion, because their CSEs are
much lower than LRMC of the supply-side electrical energy
expansion.
CSEs of the EE_lighting program ranges from 0.0072 USD/kWh
equal to 9.7% of LRMC, at the beginning of the implementation
program, to 0.004 USD/kW h equal to 5.1% of LRMC, at the end of
the implementation program. In the cases of CSEs for EE_TV,
EE_Refrig, and EE_AC program, it was found that their annual
values are relatively constant at about 0.016, 0.06, and 0.014 USD/
kWh equal to 21.6%, 80.9%, and 19.5% of LRMC during the entire
periods of implementation, respectively.
As the shares of accumulated stock of high efciency appliances
increase, the investment cost and the accumulated saving of
electrical energy from the EE programs also increase. But the rate
of growth in electrical energy savings of EE programs grows faster
than the rate of increase of the investment cost in each year. As a
result, the investment for electrical energy saving of the appliances
mentioned above is more effective than the investment of power
plant expansion.
As shown in Fig. 8, CSE of EE_lighting is the cheapest and CSE of
EE_Refrig is the most expensive, while CSE of EE_AC and CSE of TV
are in between these two values. Thus, the EE_Lighting should be
given rst priority for implementation due to its economic
attractiveness. It should be noted that the CSE value of EE_Lighting
will dramatically dropped after year 2015, as shown in Fig. 8. This
is because of the introduction of LED lamps in the saving program
at that time point; its CSE is substantially lower than CSE of T5 and
CFL which are assumed to be implemented before year 2015.
4.4. Benet of EE programs to households' electricity expenditures
Although EE programs entail additional initial costs of investments, gains from their savings are substantially larger and can
recoup the cost of the initial investments which in turn can help to
eventually lessen expenditure on electricity as a few percentages
of household income for all income groups (see Table 13).
5. Discussion
5.1. Characteristics of electrical energy consumption and potentials
of saving in urban areas of Indonesia
The survey reveals that electrical energy consumption is related
with income earning of households. As income rises, households
tend to purchase more varieties and units of appliances. Electrical

energy consumptions of urban households in Indonesia are mainly
for lighting, entertainment, and space cooling. Since lighting is
indispensable for life after sunset, lamps are the most common
appliances found in all households followed by televisions and
space cooling appliances for food preservation and living comfort.
The ownership of these electric appliances is inuenced by their
initial acquisition prices. The acquisition prices of television sets
are cheaper than those of refrigerators and air conditioning units.
As a result, for the low income class with low purchasing power,
priority is given to lighting and entertainment (televisions) before
refrigerators and air conditioning systems. And because Indonesia
is a tropical country where the ambient temperature is relatively
hot all year round, as family's income increases, appliances for
increased living comfort are likely to be purchased additionally.
Hence, the percent share of electrical energy consumption for
space cooling (refrigerators and air conditioning systems) becomes
larger as the familys income increases.
The fact that the amounts of electrical energy consumed by the
household sector accounts for 41.2% of the total electrical energy
consumption of the country and the daily peak demand occurs
between 17.00 p.m. to 22.00 p.m. (see Fig. 5) underline the
importance of the household sector in urban areas as the prime
driver of electricity consumption of the country both in term of
electrical energy (GW h) and power load demand (GW). Usually,
peoples return home after work and turn on lamps for lighting,
television set for entertainment, and air conditioning system for
spacing cooling comfort during the evening hours before going to
bed. In addition, appliances for space cooling for food preservation
(refrigerators) are usually operated at high loads during this time
of a day because refrigerators are opened and closed more frequently than at other times to remove preserved food for cooking.
This utilization pattern means that lamps, TVs, ACs, and refrigerators are the type of appliances that most likely drive the daily
peak power demand of electricity in Indonesia. The total electric
power demand by these four types of appliances was about 64% of
the maximum daily peak demands in 2011(see Table 10).
Since power rates, numbers of appliances and their operating
hours are the three factors that determine electrical energy consumption, any attempt to improve electrical energy efciency of
those appliances operating during peak power demand period
(lamps, TVs, ACs, and refrigerators) will plays an important role to
decrease the peak demand and total electrical energy consumption. The saving potential for each type of equipment can also be
estimated from their corresponding power rates, numbers in use,
and operating hours. It was found that ACs has the largest electrical energy consumption, followed by lamps, TVs, and refrigerators respectively. Although the operating hours for ACs, lamps,
and TVs are roughly similar at about 6 h per day, and refrigerators
are operated for an entire day, it was found that ACs normally
consume the largest amount of electrical energy because the rated
power of each unit of ACs is substantially much higher than other
types of appliances. As a result, the saving potential of ACs is the

largest. Since lamps are used in all households, their electrical
energy consumption is the second largest after ACs due to their
sheer large numbers. TVs are more commonly used in households
than refrigerators; therefore, urban households electrical energy
consumption for TVs is higher than for refrigerators. This indicates
that in the long term, the Indonesian government should give
priority to improving the energy efciency of these four types of
appliances, particularly for ACs because of their high rated power
consumption and because with ongoing economic development
pushing higher the standard of living, more and more units of
appliances will be deployed; this is likely to place a burden on the
government to keep up with the rising electrical energy demand.
The situation will worsen if poor efciency appliances continue to
prevail in the country.
5.2. Economic feasibility of electrical energy saving programs as
compared to new power plants development
The result of this study shows that if every initiative of the
energy efciency improvement programs is implemented from
2012 to 2030 to reduce the demand, it will be more cost-effective
than the alternative of supply-side power plant expansion as the
costs of electrical energy saving programs are much lower than the
cost of LRMC of the supply side expansion. The cost of energy
saving program for lighting is the lowest among all energy saving
programs. Moreover, the total saving cost of this program will
becomes even lower when an accumulated share of high efciency
lamps in the total stock becomes signicant. Particularly, when the
most advanced lighting technology LED lamps are introduced into
the program. Despite a high initial investment cost of LED lamps
when comparing to the conventional incandescent and CFL lamps,
the combination effect of low rated electric power consumption
and long expected life of LED lamps allow ones to save more
energy for lighting than any other types of lamps. In the near
future, we expect that the cost benets of LED lamps' implementation will be likely to increase further because of a declining
trend of LED prices. The costs of other electrical energy saving
programs are found to be roughly constant throughout the period
of implementation. This suggests that the Government of Indonesia should focus on the implementation of LED lamps as the rst
priority.
5.3. Electrical energy price and subsidy
Presently, the Indonesian government directly subsidizes the
price of electrical energy; this study found that households in
urban areas generally pay only a fraction of the actual electrical
energy cost. The lower income households pay 56% of the actual
average unit cost, while the high income households pay 62% of
the actual average unit cost. Although per unit subsidy in Rps per
kWh is larger for the lower income class, the higher income class
actually receives more total subsidy per household in Rps per
month since a high income class normally consumes much higher
monthly electrical energy than a lower income class. As one can
see in Table 13, the relative electrical expense of a higher income
class as a percentage of income is lower than that of a lower
income class. This is an inappropriate electric pricing policy for the
power sector in Indonesia for two reasons: (i) It gives subsidy
directly to the consumed electrical energy which lowers the actual
price. Thus, this policy encourages consumers to consume more
electricity, and will place increasing burden on the Indonesian
government in the long term, (ii) It is an unfair allocation of
government's nancial resources because the higher income
groups command larger total subsidy.
1171
The result of this study conrms that the implementation of

energy efciency improvement programs not only will give benets in term of the avoided cost of new power plant construction
(which in turn can help to mitigate greenhouse gases emission),
but will also help in relieving the Indonesian governments
nancial burdens by reducing the amounts of subsidy for PLN. In
addition, the energy efciency improvement programs will also
help to alleviate households' nancial burden by reducing their
electricity expenses. Since the acquisition prices of high efciency
appliances are higher than the low efciency ones, indirect subsidy of high efcient appliances is much more appropriate if the
government intends to continue subsidizing households electricity expenses since this will spur more peoples to buy new, high
efcient appliance to replace old traditional, low efciency ones.
Thus, more electrical energy can be conserved by this kind of
policy which in turn could attenuate potential political unrest if
there is any future energy price shock.
5.4. Environmental Impact and Sustainability
As mentioned earlier in Section 3.7, the EE program is expected
to not only give benet in term of electricity saving and avoided
cost of power plant expansion but it will also help to avoid the
total CO2 emission (see Section 4.2).
Decrease of fossil energy consumption eventually helps to
support the economic development be more sustainable since it
helps to extend the depletion of nite fossil energy resources and
help to mitigate greenhouse gases affecting to the global warming
impact. EE program will strengthen the deployment of renewable
energy in the national electrical energy planning, if both RE and EE
are implemented together. This is because the cost of electricity
generated by RE is generally much more expensive than EE. The
implementation of EE helps to lower the overall average unit cost
of electricity.
So we suggest that during the transition period of changing
from the utilization of fossil resources to non fossil resources, the
implementation of renewable energy (RE) and Energy Efciency
(EE) should be worked out together. This is consistent with the
IEAs analysis that gured out the potential of world abatement of
CO2 emission in the 450 ppm scenario in 2030 is mainly contributed by EE for 57%, while RE can contribute only 20% of the
total potential [35]. IEA also pointed out how RE and EE would be
implemented together to contribute to energy security, economic
development and environmental protection goals [36].
6. Conclusion
This study revealed that the electrical energy consumption in
urban households in Indonesia is mainly for lighting, entertainment, and space cooling. The electrical energy consumption is
determined by the number, rated power and type of appliances,
which in turn is highly inuenced by households incomes. Lamps,
TVs, and refrigerator are the top three main contributors of electricity consumption for the lowest income class, while for other
higher income classes they are ACs, Lamps, and TVs.
Energy efciency improvement programs for lighting (lamps),
entertainment (TVs) and space cooling (ACs and refrigerators) are
found to be more cost-effective than supply side capacity expansion by construction of new power plants. Data analysis from this
study showed that the investment cost of saving programs is much
lower than the cost of LRMC for the expansion of the power
supply. In the short-term, energy efciency improvement for
lighting should rst be prioritized, as it can save electricity at the
lowest cost. Despite a high initial investment cost, LED lamps
allow ones to save more energy for lighting than any other types of
1172
lamps. In the near future, it is expected that the cost benets of

LED lamps' implementation will be likely to increase further
because of a declining trend of LED prices. This suggests that the
Government of Indonesia should focus on the implementation of
LED lamps as the rst priority. While, in a longer term, due to
economic development pushing higher the standard of living,
Government should focus on the implementation of high
efciency ACs.
Instead of giving direct electricity price subsidy, the Indonesian
government should instead give indirect subsidy to lower the
initial high acquisition prices of high efciency appliances. This
new policy not only would reduce electricity consumption in the
household sector, but would also alleviate the Indonesian governments nancial burden in subsidizing PLN since it can obviate
the need to develop new power plants which in turn can mitigate
greenhouse gases emission. This will trigger more peoples to buy
new, high efcient appliance to replace old traditional, low efciency ones. Thus, more electrical energy can be conserved which
in turn could attenuate potential political unrest if there is any
future energy price shock.
Finally, GOI should implement EE program as counterpart of
the renewable energy application instead of developing fossil and
RE power plants together, while ignorance of energy conservation.
6.1. Policy recommendation
As mentioned in Section 5, we recommended GOI to adopt the
following policies:
(1) The government should give priority to improve the efciency
of lamps, TVs, refrigerators, and ACs, particularly for ACs
because of their high rated power consumption and higher
standard of living of Indonesia peoples in a long term.
(2) In the short term, the government should focus on the
implementation of LED lamps as the rst priority for low
income households, as their investment cost is the most

effective and due to a large amount of deployment among
poor peoples.
(3) The implementation of efciency improvement of appliances
can give benet better than new power plants development.
If the government intends to continue subsidy for households
electricity expenses, indirect subsidy of high efcient appliances is much more appropriate.
(4) Electrical energy efciency improvement should be implemented together with the development plan of renewable
energy as a part of the national power supply business plan
(RUPTL) instead of focusing on utilization of fossil and
renewable energy. This will help to extend the depletion of
fossil energy resources, reducing unit cost of electric generation, and reducing intensity of CO2 emission per kW h of
generated electricity. It will also help the development of the
country to be more sustainable.
Acknowledgment
H.B. would like to thank the The Joint Graduate School of
Energy and Environment (JGSEE), King Mongkut's University of
Technology Thonburi (KMUTT) Thailand, and Center of Excellence
on Energy Technology and Environment, the Ministry of Education, Thailand, for providing a scholarship and research fund to
pursuit his degree at JGSEE. Authors are in debt to Dr. Chumrurn
Sorapipat for his valuable assistance in polishing English of this
article.
Appendix
(See appendix Table A1).
Table A1
Result of urban household's electricity consumption survey in Indonesia.
Income class I
Appliances
Lighting
Incandescent
Incandescent
Incandescent
Incandescent
Incandescent
Incandescent
FL
FL
FL
FL
FL
CFL
CFL
CFL
CFL
CFL
Air-conditioning
Split AC, 0.5 hp
Split AC, 0.75 hp
Split AC, 1 hp
Split AC, 1.5 hp
Split AC, 2 hp
Income class II
Income class III
Income class IV
I (W)
P (unit/household)
M (h/day)
P (unit/household)
M (h/day)
P (unit/household)
M (h/day)
P (unit/household)
M (h/day)
5
10
15
25
40
60
10
18
20
36
40
5
8
11
18
25
0.73
0.64
0.45
0.45
0.13
0.01
0.02
0.79
0.38
0.04
0.02
0.10
1.14
1.67
0.67
0.38
6.8
6.4
6.4
8.9
5.1
1.5
8.3
5.3
5.3
4.0
6.0
7.5
7.4
7.6
4.9
4.2
0.79
0.40
0.51
0.52
0.25
0.04
0.01
0.63
0.63
0.04
0.22
0.09
1.10
1.16
1.57
0.85
6.11
6.77
5.02
5.39
5.65
3.00
6.00
4.97
7.44
5.70
4.91
9.69
5.87
7.37
6.74
6.50
0.38
0.16
0.67
0.48
0.31
0.02
0.02
0.61
0.42
0.06
0.09
0.01
1.52
1.51
1.85
0.63
0.67
0.82
0.88
0.80
0.27
0.00
0.00
1.41
0.75
0.16
0.08
0.00
0.90
2.25
3.14
0.82
6.86
6.65
6.90
4.40
6.00
0.00
0.00
7.36
6.49
7.88
7.50
0.00
7.44
7.35
8.54
7.72
417.50
584.50
751.00
1002.00
1503.00
0.012
0.024
9.0
3.8
0.0
0.0
0.0
0.11
0.02
0.09
0.02
0.02
10.02
6.00
6.17
3.75
5.75
0.24
0.19
0.16
0.00
0.01
7.52
5.59
6.50
6.03
4.78
1.50
8.25
6.61
6.60
4.69
8.38
13.50
7.34
6.95
6.05
7.57
0.00
5.23
5.31
7.52
2.25
4.88
0.69
0.22
0.12
0.04
0.00
4.85
5.73
6.00
6.00
0.00
1173
Table A1 (continued )
Income class I
Appliances
Refrigerating
1 dor
2 dor
Big size
Electric cooking
Rice cooker (cooking)
Rice cooker (warming)
Magic jar
Water heater
Electric stove
Entertainment
CRT TV
LCD TV
Plasma TV
PC
Laptop
Laundry
Electric iron
Washing Machine
Fan
Electric fan
Water supply
Water pump
Others
Radio/Tape
DVD/VCD player
Microwave
Electric oven
Blender
Toaster
Vacuum cleaner
Dispenser
Income class II
Income class III
Income class IV
I (W)
P (unit/household)
M (h/day)
P (unit/household)
M (h/day)
P (unit/household)
M (h/day)
P (unit/household)
M (h/day)
84.34
123.41
0.38
0.26
18.0
18.0
0.34
0.40
18.00
18.00
0.36
0.38
0.09
0.39
0.41
0.08
18.00
18.00
18.00
359.11
47.08
48.89
312.50
0.54
0.42
0.21
0.05
0.7
9.0
6.6
0.4
0.75
0.68
0.09
0.05
0.01
0.72
10.61
11.59
0.44
0.01
0.79
0.76
0.12
0.06
0.02
18.00
18.00
18.00
0.00
0.98
11.09
9.18
0.31
1.50
0.76
0.61
0.10
0.22
0.74
10.74
8.70
0.68
80.65
91.64
110.00
244.17
55.87
0.98
0.13
0.05
0.21
0.38
6.4
5.3
3.9
2.8
3.1
0.88
0.24
0.14
0.30
0.55
6.42
6.51
7.73
3.77
3.96
0.77
0.41
0.19
0.37
0.57
5.31
5.42
5.52
3.50
2.77
0.69
0.67
0.41
0.31
0.98
6.45
6.02
4.65
1.82
3.36
314.05
261.57
0.75
0.33
1.2
1.1
0.89
0.48
0.77
0.59
1.13
1.37
0.96
1.12
1.06
185.00
0.55
4.7
0.0
1.5
0.43
0.51
1.10
1.28
0.00
2.87
0.00
2.06
0.96
0.75
55.51
1.13
1.22
0.00
4.90
0.00
1.64
0.65
5.34
0.00
1.79
45.47
60.00
700.00
243.25
325.00
0.29
0.04
0.00
0.01
0.33
0.02
2.3
1.5
0.0
0.0
0.5
0.1
120.95
0.08
14.7
0.25
0.04
0.06
0.02
0.49
0.05
0.02
0.05
1.95
0.98
0.64
0.75
0.42
0.32
0.38
11.73
0.36
0.07
0.04
0.00
0.43
0.08
0.03
0.08
1.10
1.05
0.25
0.02
0.50
0.27
0.47
13.76
0.31
0.08
0.16
0.04
0.59
0.20
0.04
0.08
2.11
0.97
0.29
0.02
0.32
0.26
0.47
13.88
Remark:
1 hp = 0.746 kW
I: electric power consumption
P: Number of appliance per household
M: length time of use
Monthly income per capita (Rp)

Income class I : o 412,351
Income class IV :4 1,543,397
References
[1] Ministry of Energy and Mineral Resources (MEMR). Handbook of Energy &
Economic Statistics of Indonesia. 2012.
[2] PLN. The Electrical Power Supply Business Plan 20122021. 2012.
[3] Directorate Generale of Electricity Ministry of Energy and Mineral Resources.
Electricity Statistics 2014. 2014.
[4] Shrestha RM, Khummongkol P, Biswas WK, Timilsina GR, Sinbandchongjit S.
CO2 mitigation potential of efcient demand-side technologies: the case of
Thailand. Energy Sources 1998;20:30116.
[5] International Energy Agency (IEA). Energy Policy Review of Indonesia. OECD/
IEA; 2008.
[6] Statistics Indonesia (BPS). Indonesia Population Projection 20052025.
Jakarta 2008.
[7] Wijaya EM, Tezuka T. Electricity saving potential in indonesian households: a
techno-socio-economic analysis. In: Proceedings of the 4th international
conference on sustainable energy and environment (SEE 2011). Bangkok; 2011.
[8] Hasan MH, Mahlia TMI, Nur. H. A review on energy scenario and sustainable
energy in Indonesia. Renew Sustain Energy Rev 2012;16:231628.
[9] Prambudia Y, Nakano M. Scenario analysis of Indonesia's energy security
by using a system-dynamic approach. International Journal of Social, Behavioral, Educational, Economic. Business and Industrial Engineering 2010;4
(12):22838.
[10] Wikipedia. Gerbangkertosusila. 2012.
[11] Wikipedia. Bandung Metropolitan Area. 2012.
[12] Cochran WG. Sampling techniques. 3rd ed.. New York: John Wiley & Sons; 1977.
[13] Barnes DF, Krutilla K, Hyde W. The urban household energy transition: energy,
poverty, and the environment in the developing world 2004.
[14] Swisher JN, Jannuzzi GM, Redlinger RY. Tools and methods for integrated resource
planning: UNEP Collaroration Centre for Energy and Environment; 1997.
[15] Murata A, Koondou Y, Hailin M, Weisheng Z. Electricity demand in the Chinese
urban household-sector. Appl Energy 2008;85:111325.
1.27
Income class II : 412,351977,873

Income class III : 977,8741,543,397
[16] Utama A, Ghewala SH. Life cycle energy of single landed houses in Indonesia.
Energy Build 2008;40:19116.
[17] Ministry of Energy and Mineral Resources (MEMR). Handbook of Energy and
Economics Statistics of Indonesia 2005.
[18] Ministry of Energy and Mineral Resources (MEMR). Handbook of Energy and
Economics Statistics of Indonesia. 2009.
[19] Lindeteves Trade Center. LED lamp for household. 2012.
[20] Cahaya Electric. Daftar harga lampu LED. 2012.
[21] Glodok Electronik. Daftar harga lampu. 2012.
[22] JGSEE. Development of the 20-Year Energy Efciency Development Plan for
Thailand. Joint Graduate School of Energy and Environment; 2011.
[23] Tokotivi. Listing price of televisions. 2013.
[24] ATD Home Inspection. Average Life Span of Homes, Appliances, and
Mechanicals. ATD Home Inspection; 2009.
[25] Bursa Elektronik. Listing price and specication of air refrigerator. 2013.
[26] Tokobagus. Daftar harga ac split baru. 2012.
[27] Shresta RM, Shresta R, Bhattacharyya SC. Environmental and electricity
planning implications of carbon tax and technological constraints in a
developing country. Energy Policy 1998;26:52733.
[28] Ministry of Energy and Mineral Resources (MEMR). Electricity and Energy
Statistics 2008. Jakarta2009.
[29] State Electricity Company (PLN). The Electrical Supply Business Plan 20062015. PT. PLN (Persero); 2005.
[30] The Central Bank of Republic of Indonesia. Government bond coupon. 2013.
[31] Indonesian Central Securities Depository. Government bonds. 2013.
[32] Manoppo J. CFL in Indonesia Today. Jakarta: The Indonesian Electrical lighting
Industry Association (APERLINDO); 2011.
[33] PLN. PLN's Statistics 2012. Jakarta2012.
[34] President of Indonesia. Presidential decree No. 8/2011 on basic electricity
tariff. Jakarta 2011.
[35] International Energy Agency (IEA). World Energy Outlook. Paris 2009.
[36] Barnsley I, Blank A, Brown A. Enabling renewable energy and energy efciency
technol ogies. IEA 2015.

1-Household Electrical Consumption Indonesia PDF

Загружено:

Сведения о документе

Исходное описание:

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

1-Household Electrical Consumption Indonesia PDF

Загружено:

Авторское право:

Доступные форматы

Renewable and Sustainable Energy Reviews 57 (2016) 11601173

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

Characteristics of urban households' electrical energy consumption

H. Batih, C. Sorapipatana / Renewable and Sustainable Energy Reviews 57 (2016) 11601173

resource in a long term. However, such approach to improve

H. Batih, C. Sorapipatana / Renewable and Sustainable Energy Reviews 57 (2016) 11601173

Indonesia (Badan Pusat Statistik, BPS), the population of Indonesia

then chosen into questionnaire for a pilot survey. Evidences from

2.2. Attribution of questionnaire

where m is the true mean of a population, and s is the estimated

where Z is a standard normalized value which corresponds to the

List of candidate variables to be included in the questionnaire

Fig. 1. Survey sites.

H. Batih, C. Sorapipatana / Renewable and Sustainable Energy Reviews 57 (2016) 11601173

Population Population in urban area Population in urban area as % of the total

Surveyed cities on each islands Population in each city

Income per capita

Fig. 2. Classication of income groups in this study.

heterogeneous a population, the larger the sample size is needed

1:96 U 134; 984

Because of some random variables are discrete, a check of

The actual survey was conducted between November, 2011 and

2 U 162; 758 U 0:052

2.4. Actual survey

Z 2 pq 1:962 U0:5 U 0:5

2.5. Income class determination

H. Batih, C. Sorapipatana / Renewable and Sustainable Energy Reviews 57 (2016) 11601173

% household in urban areas

3. Methods for estimating electrical energy saving potentials

3.2. The ratio of electrical energy expenditure to income

Electrical energy in households is generally consumed for

Ni;j UP i;j UM i;j U I i;j ;

where N is a number of households, P is a number of appliances

Ni;j UP i;j UM i;j U I i;j URi;j ;

study. For refrigerators and other appliances R is assumed to be

where R is a ratio of the actual W  h consumed to a nominal rated

A ratio of electricity expenditure to income can be seen as an

H. Batih, C. Sorapipatana / Renewable and Sustainable Energy Reviews 57 (2016) 11601173

3.4. Electrical energy consumption by urban households under

where t is time (years), while a and b are constant parameters

Initial investment cost (USD)

Life time (h)

Initial investment cost for LED lamps is a forecasted price in 2020.

Fig. 3. Thailand's average penetration rate of high efciency appliances in the

Initial investment cost

97.02 60 300.00 338.95 20

Data obtained from survey.

b0.31; it generated PH(t) that nearly duplicates Thailand's

H. Batih, C. Sorapipatana / Renewable and Sustainable Energy Reviews 57 (2016) 11601173

where ALCC is the annualized initial investment cost of appliances

Data obtained from survey.

Initial investment cost

Initial investment cost (USD)

of Performance (COP). Conventional AC systems have COP around

N i;j UP i;j UM i;j U PH i;j UI BAU

where PH is the percent share of accumulated high efciency

ALCC EE ALCC BAU

where TC is the present value of a total power generation cost of

H. Batih, C. Sorapipatana / Renewable and Sustainable Energy Reviews 57 (2016) 11601173

Valid respondents' survey

kWh/month per household

where R is a ratio of the actual W h consumed to a nominal rated