An-Najah National University 

Faculty of Graduate Studies 

 

 

 

Developing Non-Industrial Buildings Energy 

Performance Indicators in Nablus city 

 

 

By 

Razan Abed Assaf 

 

Supervisor 

Dr. Mohammed Al-Sayed 

 

 

 

This Thesis is Submitted in Partial Fulfillment of the Requirements for 

the Degree of Master of Engineering Management, Faculty of 

Graduate Studies, An-Najah National University, Nablus-Palestine. 

2019 





III 

 

Dedication 

To my father PROF. ABED ASSAF who taught me how to never stop 

asking for knowledge and to be always ambitious. To my beloved mother 

the one she never gave up on my ability, she was always there for me.  

To my brother eng. Basel Assaf (May his soul rest in peace) the only one 

encourage me to continue  

All my family, my brothers, my husband and my friends thank you for 

everything 

 

 

 

 

 

 

 

 

 

 



IV 

 

Acknowledgment 

The first and final gratitude is to Allah who gave me the strength and 

determination to continue this work, who gave me the ability to endure all 

the difficulties and challenges to successfully complete this work.  

This work would not be completed if not for the help and directions from 

my department professors and my supervisor Dr. Mohammad Alsayed, 

who I would like to thank for his patience, understanding and precious 

time. A special thank for my family and friends whom without their help 

and support I would not have been where I am now. Also 

I would like to thank AlNajah University, the staff and everyone who 

taught me how to make this happen 

Finally, I would like to thank my mother, the greatest mother on earth, my 

father, who never spares me any of expertise and help. Thank you to my 

brothers for their effort, support, and standing by my side. Thanks to my 

husband and daughter for their patience and support. 

 

 

 

 



V 

 

 اإلقرار

 رسالة تحت عنوان:ال ةأدناه مقدم  ةالموقعانا 

 

Developing Non-Industrial Buildings Energy Performance Indicators 

in Nablus city 

 

ليو إشارة باستثنـاء ما تمت اإل ,أقر بأن ما اشتممت عميــو ىذه الرسـالة ىو نتـاج جيـدي الخاص
و بحثي أي بحث عممي أو أ ,ي درجة عمميةأنيل وأن ىذه الرسالة لم تقدم من قبل ل ,حيثما ورد

 خــرى. أو بحثية أي مؤسسة تعميمية أل

Declaration 

The work provided in this thesis, unless otherwise referenced, is the 

researcher's own work, and has not been submitted elsewhere for any other 

degree of qualification. 

 

Student's Name: :اسم الطالب 

Signature: :التوقيع 

Date: :التاريخ 

 

 



VI 

 

Table of Contents 

No Contents Pages 

 Dedication III 

 Acknowledgment IV 

 Declaration V 

 List of Tables VIII 

 List of Figures X 

 List of Abbreviations XI 

 Abstract XII 

 Chapter One: Introduction 1 

1.1 Chapter overview 1 

1.2 Nablus and non-industrial buildings 3 

1.3 research problem 6 

1.4 research objective and expected outcome 7 

1.5 research question 8 

1.6 significance of the study 9 

1.7 Thesis structure 9 

 Chapter Two: Literature Review 11 

2.1 Chapter overview 11 

2.2 Energy indicator EUI and ECI and their importance 13 

2.3 Overview of two energy management indicators 14 

2.4 Energy intensive is a good indicator for non-industrial 

building 

17 

2.5 Energy audit and energy management 18 

2.6 Some of non-industrial building in Nablus and their 

Energy Profile 

20 

2.7 Analysis background 22 

 Chapter Three: Research Methodology 25 

3.1 Introduction 25 

3.2 Research type 25 

3.3 Research methodology 25 

3.4 Sample size 27 

3.5 Research sample size 29 

3.6 Questionnaire 30 

3.7 Pilot study 31 

3.8 Data analysis approach 32 

3.9 T test & ANOVA test 34 

 Chapter four: Results and discussion 35 

4.1 Introduction 35 

4.2 Questionnaire analysis 35 

4.3 Interview analysis 41 

4.4 Quantitative analysis and discussion 42 



VII 

 

 Chapter Five: Conclusion and Recommendations 85 

5.5 Conclusions 85 

5.2 Limitations 87 

5.3 Recommendations 88 

 References 89 

 Appendices 95 

 ب انمهخض 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



VIII 

 

List of Tables 
No Title Pages 

Table 1 Classification of members of the general authority by 

sector 

4 

Table 2 Number of establishment of different sectors in the city 

of Nablus 

5 

Table 3 Distribution of the sample among different categories 28 

Table 4 Expert and arbitrator who reviewed the questionnaire 35 

Table 5 Questions for each category 33 

Table 6 ECU and EUI for all categories 36 

Table 7 Questions related to each category 42 

Table 8 descriptive statistics for Companies EUI 44 

Table 9 Tolerance interval for Companies EUI 45 

Table 10 Regression for EUI vs. Conditioned area 46 

Table 11 one way ANOVA EUI for Company 46 

Table 12 Standard Error 46 

Table 13 Predicted EUI VS Real measures 47 

Table 14 Descriptive statistics for Companies ECI 49 

Table 15 Tolerance  probability for Companies ECI 50 

Table 16 Regression ECI and Conditioned area 50 

Table 17 ANOVA for Companies ECI 55 

Table 18 Standard error 55 

Table 19 Actual ECI VS predicted ECI 55 

Table 20 Energy VS condition area 54 

Table 21 ANOVA Energy for company 54 

Table 22 Standard error 54 

Table 23 Predicted and actual energy 54 

Table 24 Descriptive Statistics for Hospitals EUI 56 

Table 25 Probability for Hospital EUI 56 

Table 26 Regression Statistics 57 

Table 27 ANOVA EUI for Hospital 57 

Table 28 Standard errors 57 

Table 29 Residual Output 57 

Table 30 Descriptive Statistics for Hospitals ECI 59 

Table 31 Regression for ECI V.S Conditioned Area for hospitals 60 

Table 32 ANOVA ECI for hospital 60 

Table 33 Standard Error 60 

Table 34 RESIDUAL OUTPUT 60 

Table 35 Regression Statistics 62 

Table 36 Standard error 63 

Table 37 RESIDUAL OUTPUT 63 

Table 38 Descriptive statistics for Restaurants and Hotels EUI 64 



IX 

 

Table 39 Probability Restaurants and Hotels EUI 65 

Table 40 Regression 65 

Table 41 ANOVA EUI for Restaurants and Hotels 65 

Table 42 Standard error 65 

Table 43 Predicted EUI VS Real measures 66 

Table 44 Descriptive Statistics for Restaurants and Hotels ECI 68 

Table 45 Probability for Restaurants and Hotels 69 

Table 46 Regression Statistics 69 

Table 47 ANOVAECI for Restaurants and Hotels 69 

Table 48 Standard error 70 

Table 49 RESIDUAL OUTPUT 70 

Table 50 Regression statistics 72 

Table 51 ANOVA Energy for Restaurants and Hotels 72 

Table 52 Standard error 72 

Table 53 RESIDUAL OUTPUT 73 

Table 54 Descriptive statistics for Schools EUI 74 

Table 55 Probability for Schools EUI 75 

Table 56 Regression Statistics 75 

Table 57 ANOVA EUI For School 75 

Table 58 Standard Error 75 

Table 59 RESIDUAL OUTPUT 76 

Table 60 Descriptive Statistics for Schools ECI 78 

Table 61 Probability for Schools ECI 79 

Table 62 Regression 79 

Table 63 ANOVA ECI for School 79 

Table 64 Standard Error 80 

Table 65 Predicted ECI VS Real measures 80 

Table 66 Regression statistics 83 

Table 67 ANOVA Energy  for School 83 

Table 68 Standard Error 83 

Table 69 RESIDUAL OUTPUT 83 

 

 

 

 



X 

 

List of Figures 
No Title Pages 

Figure 1 Classification of member of general authority by 

sector 

4 

Figure 2 Change in primary energy intensity in selected 

countries and regions 

52 

Figure 3 Primary energy use in U.S. commercial building in 

2010 

22 

Figure 4 Normal probability plot for Companies EUI 45 

Figure 5 EUI and conditioned area 47 

Figure 6 New EUI and conditioned area in companies 49 

Figure 7 Normal Probability plot for Companies ECI 50 

Figure 8 Actual ECI VS predicted ECI 52 

Figure 9 The relationship between actual energy and the 

predicted energy 

53 

Figure 10 Predicted EUI VS Real measures 58 

Figure 11 New EUI and energy in Hospitals 58 

Figure 12 Normal Probability plot for Hospitals 59 

Figure 13 The relationship Predicted ECI VS Real measures 65 

Figure 14 New ECI and energy in Hospitals 65 

Figure 15 Predicted Energy VS Real measures 63 

Figure 16 Normal Probability plot for Restaurants and Hotels 

EUI 

64 

Figure 17 Predicted EUI VS Real measures 66 

Figure 18 New EUI and Energy 67 

Figure 19 Normal probability plot for Restaurants and Hotels 68 

Figure 20 Predicted ECI VS Real measures 70 

Figure 21 New ECI and conditioned area 75 

Figure 22 The relationship between: Predicted Energy VS Real 

measures 

73 

Figure 23 Normal Probability Plot for Schools EUI 74 

Figure 24 Predicted EUI VS Real measures 77 

Figure 25 New EUI and conditioned area 77 

Figure 26 Normal probability plot for Schools ECI 79 

Figure 27 Predicted ECI VS Real measures 85 

Figure 28 New ECI and conditioned area 82 

Figure 29 The relationship between Predicted Energy VS Real 

measures 

84 

 



XI 

 

List of Abbreviations 

CCBFC Canadian Commission on Building and Fire Codes 

CI 
 

Confidence Interval 

CT Computerized Tomography 

ECI Energy Cost Index 

EIM Energy Institutions Managements 

EM Energy Management 

EUI Energy Utilization Index 

GDP Gross Domestic Product 

MRI Magnetic Resonance Imaging  

PI Probability Interval 

OECD Organisation for Economic Co-operation and Development 

SDPD Seattle Department of Planning and Development 

TI Tolerance Interval 

US United State 

 

 

 

 

 

 



XII 

 

Developing Non-Industrial Buildings Energy Performance Indicators 

in Nablus city 

By 

Razan Abed Assaf 

Supervisor 

Dr. Mohammed Al-sayed 

Abstract 

Scientific evidence indicates the seriousness of energy loss. Given the 

importance of energy, the global direction is toward energy conservation. 

Studies showed the amount of energy loss and big energy consumption in 

Palestine. From that point, this study was conducted, in order to adapt and 

develop energy performance indicators, study the main variables affecting 

those indicators. The study was conducted in the city of Nablus because it 

is the economic capital of Palestine.  

Energy performance indicators, ECI and EUI were calculated. ECI is the 

average annual consumption of energy divided by conditioned area, while 

EUI is the average annual payment divided by size.  After calculating both 

indicators, variables affecting those indicators were studied with special 

focus on size multiplied by occupancy rare, which was termed conditioned 

area.   

The methodology for this study was to collect data through questionnaires 

and interviews; the sample was 78 facilities, distributed into categories 

including companies, hospitals, hotels, restaurants, and schools. After data 

collection, statistical analysis including descriptive and differential analysis 

was conducted, descriptive statistics of the questionnaire showed that 

multiple variables could affect energy indicators, which should be studied 



XIII 

 

in the future, those variables include an age of the facility and used 

equipment.  

Results indicated a direct relation between conditioned area and energy 

(Hotels p=.005, schools p=.008, companies p=.0005) this relationship 

between energy and conditioned area reflect the relation with energy 

indicators. The researcher mainly recommends studying every category 

separately taking into consideration all variables affecting energy 

indicators, and the extent of their effect on both ECI and EUI. As a 

secondary recommendation, the researcher encourages facilities to register 

and save their energy data, and following energy conservation strategies. 

 

 

 

 



1 

 

Chapter One 

Introduction 

1.1 Chapter overview 

This chapter gives a general idea about this research; it includes an 

introduction, definition and objectives, the hypothesis and research 

questions, the importance of the study and finally the research structure. 

In order to deal with Global Energy challenges and its corresponding 

negative environmental impacts, so what is energy.  Simply, energy can be 

defined as the ability to do work. According to the Organisation for 

Economic Co-operation and Development (OECD)) 2012), as an economy 

grows, demands for energy increases rapidly, thus energy consumption 

increases. Forecasts show that by 2050 the world economy will become 

four times larger than today and it is projected to use 80% more energy 

(OECD, 2012). So we need a good quality of statistics to understand the 

underlying trends in energy consumption.  

Since energy use has a big impact on our environment, which can be 

negative or positive, many civilized governments support sustainable 

development in their societies which will benefit current and future 

generations. 



2 

 

One of the most basic steps to counter this situation is to start developing, 

monitoring, and improving energy performance indicators. energy indicator 

such as energy utilization index (EUI) and energy cost index (ECI), the first 

one talks about energy used per one unit of area. Energy index will help to 

further study many parameters like energy used per worker; energy used 

occupancy energy effect by age buildings, the used equipment and the type 

of used equipment’s. The second type is energy cost; here will talk about 

the cost aspect of energy 

An accurate and a holistic understanding of Energy Indicators (EnPIs), 

which are easily understood, quantitative measure of performance will 

increase efficiency, decrease intensity, increase understanding of 

improvement, define abnormal situations, implement projects that reduce 

energy consumption, increase production output, and capture cumulative 

impact of all projects by statistically isolating various influences on energy 

use.    

The proposed research aims to develop Palestinian local community energy 

consumption as Energy Utilization Index (EUI) and Energy Cost Index 

(ECI) and its direct and indirect effects on their local environment and 

economy. Moreover, aims to calculate and analyse the average of EUI and 

ECI for Nablus city, in such a way, these indicators could be assumed as a 

baseline for determining potential energy savings opportunities for various 

buildings types, which could help experts, engineers, accountants, and 

policy makers in defining better solutions and developing appropriate 

plans. 



3 

 

1.2 Nablus and non-industrial buildings 

Nablus is a Palestinian city, with an area of 12700 acres, it is 

considered one of the economically active cities, it plays an important role 

in the economy despite occupation and siege. In fact, despite those 

challenges the city of Nablus is considered the economical capital of 

Palestine, it was a center for Arab and Palestinian banks in the west bank 

before most of them were transferred to Ramallah with the establishment of 

the Palestinian Authority there, Nablus is also famous for the construction 

of its markets, especially in the Kasbah district of the Old City. It is also 

famous for its sweets. 

The City of Nablus occupies the first place among Palestinian cities 

in terms of non-industrial facility population According to 2016 statistics; 

Nablus had a population of 153,061 people with a population density of 

540 per Km2. (Nablus Chamber of Commerce, 2011). 

According to the 2009 records of the Palestinian Central Bureau of 

Statistics, the number of industrial and commercial facilities in Nablus 

District was about 13,742 distributed among different sectors (private, 

public and civil as shown in (Table 1) (Figure 1). 

 

 



4 

 

Nablus Chamber of Industry and Trade 

Table 1: Classification of members of the general authority by sector 

Sectors Number of members 

Commercial sector 3621 

Industrial sector 969 

Services 383 

Careers 151 

Construction 183 

 

 

Figure 1: Classification of member of general authority by sector. 

In addition, the overall number of workers reached 38,542 between 

employers and workers. Recently, a new sector emerged in Nablus 

pertained to services including hotels, restaurants, stock-exchange, and 

communications, being managed with total of 4000 workers distributed 

among 900 services’ utilities. (Nablus Chamber of Commerce, 2011) 

 

 

commercial

sector
industrial

sector
services

careers



5 

 

Table 2: Number of establishment of different sectors in the city of 

Nablus 

 

Economic indicators showed a daily loss of Nablus District during 2002-

2008 estimated at $ 1.2 million per year of which industrial sector share 

was 42%, services sector was 30%, the agriculture sector was 20%, and 

tourism sector was 2% (Nablus Chamber of Commerce, 2011). 

    Today, and upon the reinforcement of the security, it was clearly noticed 

that a number of new service facilities were opened including restaurants, 

public parks, and different commercial centres. Nablus enjoys the high 

potential for being the leader of Palestinian industry and commercial 

activities even in the hardship and worst situations. It is, with no doubts, 

the business incubator and the economic capital of Palestine. 

All this data indicates the importance of this study especially in Nablus, 

given that it is a non-industrial city with a large number of non-industrial 

buildings. Also, the importance of energy conservation in Nablus which is 

essential for the provision of environmental conservation and development 



6 

 

in the future. This study is concerned specifically with the city of Nablus 

hoping it might have a clear, positive and useful impact on its economy, 

given the financial and economic deficits the city has been suffering from 

in the last few years, hopefully, the conservation of energy and finding the 

causes of energy loss will have a role in improving the economy in the 

future.  

1.3 Research problem 

In the field of energy, the Palestinian economy and especially energy 

situation is a bit more complicated because Local environmental legislation 

is limited and the lack of real data describing the current situation. Thus, 

Palestinian decision-makers do not have a clear vision regarding the 

performance of their energy consuming systems. Having a clear and 

controllable vision regarding energy consumption will enhance and 

improve the economy which will benefit the environment which will also 

benefit the society as it removes obstacles and maintains sustainability  

 Palestinian decision-makers do not have a reference point to know 

how much they consume and how their lifestyle sustainability and its 

consequences on their environment and local economy.  In addition, the 

proposed project aims to facilitate business owners’ energy conservation 

decisions. However, an effective energy conservation and management 

program should begin from understanding current energy consumption 

situation. Based on that, comparison with average indicators gives a good 

understanding regarding the current situation and potential saving that 



7 

 

could be achieved. In this context, EUI is one of the most powerful 

indicators. It sums up that all building energy consumption per year 

(electrical, gas, oil fuel …etc.) then dividing over conditioned space area, 

in such a way, a kJ/m2 or kWh/m2 indicator then be calculated. In addition, 

the ECI which is the same as EUI except that its numerator contains the 

annual energy cost, thus, the indicator will have the unit $/m2. Both 

indicators will be calculated based on the better comparison between 

buildings, assessing energy management programs and estimating possible 

saving opportunities.  

In order to do so, a statistical survey is conducted. An interview-

based Questionnaire is used, which aims to accomplish the main objectives 

of the study which is targeting different building structures to calculate its 

EUI and ECI. To do it effectively, the US building in the same context 

classification which includes at least the following subdivisions: all 

buildings, education buildings, food sales, food services, health, lodging, 

retail, office, assembly, safety, mosques and churches, service, warehouse, 

and others will be used as a guide to establish our targeted building 

subdivisions 

1.4 Research objective and outcome 

The main objectives of this research proposal are: 

1. Calculating the average value for EUI and ECI for Nablus city  

2. Define general trends building energy consumption in Nablus. 



8 

 

3. Establishing a reliable data base of information to be used by buildings 

owners as a reference point to estimate possible savings 

4. Identifying factors that have a direct effect on energy indicators  

This research has been started so as to reach the following outcomes: 

1. Estimated EUI ECI for non-industrial building in Nablus  

2. Estimated CI and TI with 95% level of significance   

3. Description of the energy consumption trends and behaviour for non-

industrial building. 

4. Energy indicators with tables and figures displaying the variations in 

order to use it in the future  

1.5 Research questions 

To reach the research outcome, the researcher needs to find answers for the 

following questions: 

1. How much energy consumption in EUI and ECI? 

2. How much the variance between non-industrial buildings with respect to 

the two indicators? 

3. What are the significant difference between categories? or can we sum it 

in under one category called building  



9 

 

4. How can we define a specific relationship between energy indicators, 

area, size, occupancy rate and number of employee? 

1.6 Significance of the study 

Through the development of energy indicators, energy will be 

conserved and opportunities to decrease cost will be available, thus 

increasing competition and profits of the company. Also, energy audit 

plays a key role for the benefit of a company, because it provides 

opportunities to learn what needs to be done to develop and improve 

energy conservation rates. The first step for any company in energy 

auditing is to calculate basic indicators that will be studied here such as 

EUI and ECI. The outcome of these indicators includes tables and 

statistical values that would be a reference for any company or organization 

in order to know their place in terms of energy conservation. doing useful 

and is there any way to increase energy conservation. 

1.7 Thesis structure 

This thesis consists of five chapters. An outline of each chapter is given: 

Chapter 1 gives a general idea about this research; includes an overview of 

Palestinian situation and particular focus on energy use in Palestine, and 

the encompass of Nablus and non-industrial buildings as the main problem 

and objectives, and research questions, the importance of the study and 

finally the research structure. 



10 

 

chapter 2 represents, previous research work including an overview of 

previous studies about energy indicators. This chapter covers various 

aspects of building energy performance and defines the two indicators EUI 

and ECI, from the relationships of energy use and different building 

characteristics, also focused on the energy audit and energy levels to 

energy conservation and retrofitting in building finally talked about the 

sector that we will talk about  

Chapter 3describes the research type by defining population of sample size 

which would be 78 and data analysis approach by using conducting 

quantitative interviews and questionnaire for data collection to get the 

average for ECI and EUI for non-industrial building. 

Chapter 4 presents and discusses our results; each result is presented 

separately with a comprehensive discussion.  

Chapter 5 talks about the conclusions and recommendations and some of 

the limitations by defining difficulties from gathering data and information  

 

 

 

 



11 

 

Chapter Two 

Literature Review 

This chapter reviews the previous studies regarding energy saving in 

buildings, energy management and energy audit and how can we develop 

indicators to control the energy consumption by benchmarking between 

each categories in the non-industrial buildings 

2.1 Chapter overview 

This chapter reviews the previous studies regarding energy 

consumption and their analysis of empirical and theoretical data in order to 

demonstrate the importance of energy saving in nonindustrial buildings. 

Also, by studying two indicators the EUI and ECI we can know if we want 

to reduce energy consumption which leads to the first step in an energy 

audit, how much we consume, are we above average and out of limit 

energy. 

Many authors discussed the positive relationship between the 

management and efficiency of energy consumption. Bloom et al. (2010) 

were able to find a positive relationship between good management 

practices and productivity/energy efficiency which was later defined as the 

positive impact in EM (Energy Management), suggesting that well-run 

firms use energy more efficiently. Also, Cowan et al. (2010) further noted 

that especially energy consumption and conservation qualifies for setting 

measurable targets which are the basic line for energy management 

practice. Thus, good management results in more efficient energy 



12 

 

conservation. One of the first steps of good management is calculating 

quantities and comparing them with indicators that we will study and know 

what can be done, developed and adjusted for proper conservation, which is 

within the levels of energy.  

This was also noticed later on the global report (energy efficiency 

services limited, 2017) which reported an improvement in energy 

efficiency by 13% between 2000 and 2016. Changes in the global primary 

energy intensity are influenced by improvements in energy efficiency as 

well as changes in economic structure, such as the movement of economic 

activity away from energy-intensive industry towards less intensive service 

sectors Thus can be noted that countries that have developed the proper 

management of energy conservation have achieved a significant increase in 

energy efficiency and this shows that countries adopt energy management 

to improve the efficiency of consumption, as shown in (Figure 2) 

 

Figure 2: Change in primary energy intensity in selected countries and regions. 

 



13 

 

2.2 Energy indicator EUI and ECI and their importance 

The indicator called Energy Utilization Index (EUI) is one of the 

most powerful indicators. sums up all building energy consumption per 

year (electrical, gas, oil fuel …etc.) then dividing over conditioned space 

area, in such a way, a kJ/m2 or kWh/m2 indicator will be calculated. In 

addition, the Energy Cost Index (ECI) which is the same as EUI except that 

its numerator contains the annual energy costs, the benefits from the two 

indicators are to know how much we consume, is there a lot of energy loss 

in buildings, what is the solution and hopefully working on raising 

awareness of global environmental challenges in the future by developing 

and adopting a number of environmental legislation.  

However, local environmental legislation is limited, and 

environmental problems are not assumed with high priority in comparison 

with any political issue either internally or externally. In addition, there is a 

lack of real data describing the current situation. Thus, Palestinian 

decision-makers do not have a clear vision regarding their lifestyle 

sustainability, and its consequences on their environment and the local 

economy.  

EUI is a very important energy performance indicator, many studies 

discussed the effects of establishing the EUI in buildings and how after 

knowing the average EUI for each commercial building can save energy 



14 

 

2.3 Overview of two energy management indicators: 

There are several studies connecting EUI with saving environment energy 

performance and ECI with economic performance (Allcott and Greenstone, 

2012; Bloom et al., 2010; Cetindamar and Husoy, 2007; Cowan et al., 

2010; Enderle and Tavis, 1998; Porter and Linde, 1995).  

In their paper, Enderle and Tavis (1998) suggested a balanced concept for 

organizations, combining economic, social and environmental 

responsibilities. Savings in energy consumptions, in particular, are found to 

make both economic and environmental sense. They further argued that 

environmental and economic responsibilities partly overlap when comes to 

the use of energy, since energy savings can be justified from a purely 

economic as well as a noneconomic point of view. While Enderle and 

Tavis (1998) remained cautious and reported the overlap being only partial, 

other authors such as Porter, and Lindeand  Cetindam (1995) argued more 

strongly and suggested a ‘win-win’ situation, They suggested that well 

placed environmental standards and goals in organizations will trigger 

innovations that will subsequently lower overall costs. Similarly, 

Cetindamar and Husoy (2007) noted that environmentally sound measures 

are often at the same time economically sound and have the potential to 

result in higher profits in the long run. This theoretical assumption has been 

tested and confirmed by Al-Tuwaijri et al. (2004) who found a significant 

positive relationship between environmental and economic performance. 



15 

 

Al-Tuwaijri et al. (2004) further suggest the environmental performance 

and economic performance are closely linked to management quality in 

energy management. Indeed, management seems to have realized the 

importance of tackling energy usage and energy efficiency. Bloom et al. 

(2010) were able to find a positive relationship between good management 

practices and productivity/energy efficiency which later defined as the 

positive impact in EM, suggesting that well-run firms use energy more 

efficiently. Similarly, Montabon et al. (2007) argued that environmental 

management practices are positively related to firm performance. In order 

to reach environmental and other organizational goals, Waggoner et al. 

(1999) and Caldelli and Parmigiani (2004) perceived exclusively financial 

performance measures as not sufficient. by supporting the inclusion of ECI 

performance indicators in organizational performance measurement energy 

systems. Authors have found support for the notion that environmental 

accounting can positively influence an organization’s ability to estimate 

and control environmental costs (Buhr, 1998; Caldelli and Parmigiani, 

2004; Li and McConomy, 1999).  

Dingwerth and Eichinger (2010) argued that collecting data will raise 

awareness and could set internal processes within the organization in 

motion. Caldelli and Parmigiani (2004) likewise suggested that the 

inclusion of environmental goals in a performance measurement system 

will help manage these as well as to report on transition. They stated that 

environmental accounting is used to improve the organizational oversight 



16 

 

over impact and effects the organization’s activities have on the 

environment and with respect to energy. 

Cowan et al. (2010) further noted, that especially energy consumption and 

conservation qualifies for setting measurable targets which are the basic 

line for energy management and it is practice, on the other hand, EUI is 

very important performance energy indicator,  lots of studies talked about it 

after establishing the EUI in building and how after knowing the average 

EUI for each commercial building thought that now we can save energy by 

knowing the EUI for each commercial building and compared with control 

chart to be in a control limit so can control the use of energy, but In theory, 

the approach has merit, but most previous attempts to use simple targets for 

commercial buildings have failed (Goldstein and Eley 2014). Two major 

drawbacks to an EUI target approach are difficulty in setting an appropriate 

and fair target and difficulty in having a reliable prediction of building 

energy use. Setting fair and appropriate targets can be a substantial 

challenge. EUI targets can be developed based on the actual energy use of 

typical existing buildings or by using prototype building models 

normalized for the climate. Unfortunately, few buildings are typical. Even 

simple buildings vary in function, number, and frequency of occupants, 

plug and process loads, hours of operations, and other energy services. That 

makes fixed targets either too easy or too difficult to meet (Goldstein and 

Eley 2014)   



17 

 

2.4 Energy intensive is a good indicator for non-industrial building 

Commercial consumers are defined to be all non-residential consumers, 

with the exception of users classified as a large industry.  Studies in many 

countries revealed that non industrial buildings are one of the most energy-

intensive building categories as mentioned in (global energy efficiency 

2017) the movement of economic activity away from energy-intensive 

industry towards less intensive service sectors, Changes in global primary 

energy intensity are influenced by improvements in energy efficiency as 

well as changes in economic structure 

Energy efficiency indicators, or more generally energy performance 

indicators, give the links between energy use and some relevant monetary 

or physical indicators measuring the demand for energy services. 

Generally, energy efficiency indicators are intensities, presented as a ratio 

between energy consumption (measured in energy units) and activity data 

(measured in physical units) (IEA, 2014), as shown in the following 

equation:  

𝐸𝑛𝑒𝑟𝑔𝑦𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦𝐼𝑛𝑑𝑖𝑐𝑎𝑡𝑜𝑟= 𝐸𝑛𝑒𝑟𝑔𝑦𝑐𝑜𝑛𝑠𝑢𝑚𝑝𝑡𝑖𝑜𝑛/𝐴𝑐𝑡𝑖𝑣𝑖𝑡𝑦𝑑𝑎𝑡𝑎 

Through this study, we will find the correlation between variables like the 

number of workers and occupancy rate with EUI, which will provide the 

opportunity to develop energy indicators to calculate the national GDP and 

to develop further indicators  

 



18 

 

2.5 Energy audit and energy management 

The main point of choosing a level of an energy audit is the right level from 

the first time to reduce effort and cost. Based on the American Society of 

Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) 

(ASHRAE, 2015)) defines three levels of audits. Each audit level builds on 

the previous level. As audit complexity increases, so do thoroughness of 

the site assessment, the amount of data collected and the details provided in 

the final audit report. This effort can translate into higher energy savings. 

Each level depends on the next; some companies cannot identify its place 

between levels of an energy audit, despite the availability of money, time 

and resources. In this study we will attempt to identify those levels by 

defining them according to the American society of energy engineers, and 

identifying energy indicators which will make it easier for business owner 

to find out if there is an excessive use of energy, why there is an excess and 

what is the easiest and comprehensible plan for energy audit. Those levels 

are:  

 Level I: Site Assessment or Preliminary Audits identify no-cost and low-

cost energy saving opportunities (ASHRAE,2015), only a month to month 

comparison of the general energy use will be done including electric bills 

and gas bills, in order to know a general energy, use of the company, gives 

a general view of potential capital improvements. Activities include an 

assessment of energy bills and a brief site inspection of your building.  Is 

there a reasonable reason for energy loss in a specific month, which is an 



19 

 

easy and inexpensive process? Through this study, by identifying the 

energy indicators (EUI, ECI) we can compare a company general energy 

use with other companies from the same field which in turn gives us a 

broader indicator for energy consumption and if we have energy loss or 

not. 

Level II: Energy Engineering Analysis Audits identify energy lost cost and 

low-cost opportunities, and also provide EEM table and what are the 

recommendations in line with financial plans and potential capital-intensive 

energy savings opportunities based on (ASHRAE,2015). Level II audits 

include an in-depth analysis of energy costs, energy usage and building 

characteristics and a more refined survey of how energy is used in a 

building. This process is very important but also very expensive, where an 

external team would be assigned to do the audit, yet it is a big step in the 

right direction for big companies. 

Level III: Detailed Analysis of Capital-Intensive Modification Audits 

provides solid recommendations and financial analysis for major capital 

investments(ASHRAE,2015). In addition to Level, I and Level II activities, 

Level III audits include monitoring, data collection, and engineering 

analysis.  It is a very expensive process that could last for months. It is very 

comprehensible, it takes a detailed look at the company energy loss plans, 

and it looks for every reason for energy loss and what can be done to 

conserve energy. Engineers will help in setting the appropriate budget and 

levels of energy. 



20 

 

By identifying energy audit levels, we should note that the development of 

performance indicators which is the main purpose of this study will serve 

the interest of auditing. The first level of auditing depends mainly on 

consumption and comparison, and through this study, we will make it 

easier and give everyone the opportunity to start the audit process and give 

them an idea about the amount of consumption  

2.6 Some of non-industrial building in Nablus and their Energy Profile 

by chose a classification according to the US classification, did not take it 

fully but relied on the same basis for the selection process 

Restaurant’s: According to data available from the U.S. Department of 

Energy (DOE), food service buildings are the most energy-intensive of all 

commercial buildings Estimates of energy consumption attribute 50% of an 

average restaurant’s energy use to the preparation and storage of food 

(Clark Energy Cooperative, 2002). Both full service and fast food 

restaurants pose significant challenges to HVAC systems because the 

presence of people and the activities of food preparation and presentation 

creates high heat gain; ventilation air requirements are high; kitchen doors 

must be isolated; and there are typically a large influxes of outdoor air 

through entrances during peak hours Therefore, considered to include 

restaurants and hotels with conditioning and heating systems and 

refrigerators.  The sample was taken from the city of Nablus randomly and 

here is worth mentioning that the hotels have been added because the hotels 



21 

 

here are only for accommodation not for entertainment, hotels include 

rooms and restaurants. 

Hospitals: The electric loads of hospitals are largely lighting, HVAC, 

elevators, medical equipment (such as CT scan and MRI), computers, 

copiers and other office equipment. Also, gas and diesel use in the hospital 

is higher compared with other establishments, thus, the sample collected 

from hospitals is higher and more holistic. Hospitals also include health 

centers like eye centers, dialysis centers, fertility centres, and outpatient 

clinics. Taking into consideration that hospitals as a building and 

equipment’s are the same; the sample was taken from hospitals in Nablus. 

Office Buildings (Schools, Universities, office, company,) 

In the U.S., commercial buildings share 19% of the primary energy use 

among all sectors in 2010 (U.S. Energy Information Administration, 2011; 

2012). Figure (3) below shows the breakdown of primary energy use in 

commercial buildings by end-use services. The figure demonstrates that 

five energy services accounted for 79% of primary energy use in buildings 

in 2010. These are thermal comfort space conditioning that includes space 

heating, cooling and ventilation (39%), illumination (20%), sanitation and 

hygiene, including water heating (4%), communication and entertainment 

electronics including televisions, computers, and office equipment (8%), 

and provision of food, refrigeration and cooking (8%) (U.S. Department of 

Energy, 2008). The remaining 21% includes service station equipment, 

telecommunications 



22 

 

 

Figure 3:  Primary energy use in U.S. commercial building in 2010. 

By studying the establishments in Nablus city the following took in 

the inclusion criteria: The fact that the commercial sector represents the 

economy of the city, a great deal of attention was on including companies 

from the city taking into account the higher percentage of energy use 

among non-industrial buildings in Nablus.  

But in our case divide the schools alone because of the number of 

schools in the city. The rest of the classification was according to the USA 

classification which includes buildings of university, companies and others, 

some buildings were not included like banks because Nablus does not have 

main bank only branches.  

2.7 Analysis background 

An analysis was carried out to suit the required data. Using Minitab 

analysis is done and created descriptive data and the analysis was 



23 

 

conducted. Tolerance intervals are very useful when predict has wanted a 

range of likely outcomes based on sampled data, on the other hand, the 

confidence interval is an interval estimate for a parameter value. 

constructed in a way so that, in the long run, a given proportion of these 

intervals will include the unknown true parameter value. The proportion is 

given by the "level of confidence". For instance, you can expect that at 

least 90% of (a large series of) 90% confidence intervals will include the 

unknown true values of the parameters. 

The differences between CI &TI  

The width of a confidence interval depends entirely on sampling error. The 

closer the sample comes to including the entire population, the smaller the 

width of the confidence interval, until it approaches zero, but a tolerance 

interval's width is based not only on sampling error but also variance in the 

population. As the sample size approaches the entire population, the 

sampling error diminishes and the estimated percentiles approach the true 

population percentiles. 

According to (Gelo,2014) confidence interval is an Interval that we are (1- 

α)% confident covers μ, gets smaller as n increases, Provides a range for 

the practical interpretation of the mean and Tells us nothing about 

individuals in the population 

According to Olsson, (2013) tolerance interval like a confidence interval 

for individuals can cover a certain proportion of the population with a 



24 

 

certain degree of confidence, for example, a 99%/95% tolerance interval 

will include 99% of the population with 95% confidence.  

The prediction interval will always be longer than the confidence interval 

for because there is more variability associated with the prediction error 

than with the error of estimation. This is easy to see because the prediction 

error is the difference between two random variables, and the estimation 

error in the CI is the difference between one random variable and a 

constant. As n gets larger, the length of the CI decreases to zero So as n 

increases, the uncertainty in estimating goes to zero, although there will 

always be uncertainty about the future value even when there is no need to 

estimate any of the distribution parameters. Practical Interpretation: Notice 

that the prediction interval is considerably longer than the CI. This is 

because of the CI an estimate of a parameter, while the PI is an interval 

estimate of a single future observation 

 

 

 

 

 



25 

 

Chapter Three 

Research Methodology 

3.1 Introduction 

In this chapter will describe the research type and the sector that 

want to study by defining population of sample size and data analysis 

approach   

3.2 Research type 

Depending on the research goals, a researcher can find the type of the 

research (Bhattacherjee:2012) and this research is calculating the indicators 

of energy in order to explain the relationship between variables which is 

consumption in energy number of worker so, in this study quantitative 

method approach was used to collect quantitative data. The main point 

from quantitative data to find the average of EUI and ECI and after that to 

investigate the coloration between variables which is represented by the 

figures and charts (Creswell,2012) and to explore and understand the 

circumstances caused the calculated EUI & ECI, although the data must be 

understandable and give a solution to the research problem (Creswell,2014) 

3.3 Research methodology steps  

First, have been developed a strategy for the study methodology. 

have enumerated the number of non-industrial commercial buildings 

located in the city of Nablus according to the 2015 statistics of the 

Palestinian Central Bureau of Statistics.  



26 

 

Second, chose a suitable classification of buildings according to the 

Palestinian needs, the availability of information and the nature of the 

Palestinian environment itself. The classification includes schools, 

hospitals, restaurants, hotels, and companies. The sample was chosen based 

on a planned strategy. From the non-industrial buildings, we chose our 

sample based on size and occupancy rate, also, first began to do a pre 

studying in hospitals in order to find the variables that could affect this 

study, based on these criteria collected our sample from hospitals, 

restaurants, hotels, universities, schools, supermarkets, and shopping malls. 

For example, schools were chosen based on size, number of laboratories 

and the existent of energy saving programs. Hotels were chosen based on 

the number of rooms and occupancy rate. Nablus is a small city but it has a 

great economic status and lots of landmarks which makes small hotels 

compete with big hotels that why included hotels with more than 15 rooms 

in this study. Restaurants were also chosen based on size, occupancy rate 

and available equipment’s    

Third, the sample size was calculated based on a specific 

mechanism. The number of total operating establishments and the area of 

the same establishment were taken into account. Based on those criteria the 

establishments with small areas have been cancelled and the large ones 

with high occupancy rate were included 



27 

 

Fourth, by designed a questionnaire and distributed it, the quantitative 

questionnaire was designed based on a previous study with the same 

scientific purpose, the questionnaire was presented and checked by a panel 

of experts, and then the questionnaire was distributed to the targeted 

sample. Data were collected through a questionnaire (shown in appendix 1) 

distributed and collected by the researcher. An also quantitative interview 

was also conducted to collect more data 

Fifth: Purification and filtering the data; after data collection, the data was 

purified for information, knowing what will be the benefits from this study, 

and pointing to some important things for this study and other studies that 

might build on our results  

Sixth: Statistical analysis: a Minitab analysis program was used, based on 

descriptive analysis of data 

3.4 Sample size 

Nablus is the economic capital of Palestine. In 2009, the number of 

economic facilities in the city of Nablus was 13742(Nablus Chamber of 

Commerce, 2009) as shown in Table 2. facilities belong to social services 

with 980, facility 235 were educational facilities, 490 health, and social 

work, 484 were real state and commercial, 103 Transport, storage, and 

communications, 681 restaurants and hotels and 125Financial 

intermediations(Nablus Chamber of Commerce, 2011). According to the 

Industrial Analysis Report from the Chamber of Commerce and Industry, 



28 

 

Nablus has a 56.3% share in the economic facilities which makes the 

number of facilities in Nablus 1748, by using the following equation: 

 

 

N=1748                   P=.5                        D=.05                           Z=1.96 

Based on the previous equation the desired sample would be 315, however, 

in this study only want to include large facilities, given that 25%of them 

are considered large facilities (Chamber of Commerce and Industry, 

Nablus). the sample size would be 78. Table (3) shows the distribution of 

the sample among different categories 

Table 3: Distribution of the sample among different categories 

Category  Number of 

establishments  

Percentage  Desired 

sample  

Collected 

sample 

Restaurants and 

hotels  

681 22% 16 6 restaurants 

and 6 hotels  

Transport, 

storage and 

communications 

103 3.3% 3 3 

Financial 

intermediation 

125 4.1% No main 

banks in 

Nablus  

-------- 

Not applicable  

social services 987 31% 24 ------not 

applicable  

health and social 

work 

490 16% 12 4  

Education 235 7.5% 6 22 schools and 

2 universities  

real state, 

renting, and 

business 

activities  

484 15.5% 12 30 companies 

 
     ppzdN

ppN
n






11

1
22



29 

 

Based on the Central Statistical Book of Palestinian Cities, relied on the 

number of establishments located in Palestine as in the table (2) to 

determine the classification of establishments and their number  

3.5 Research sample size 

The purpose of a survey is to capture the main characteristics of the 

population at any instant or monitor changes over time (Tan. 2004). Hence, 

proper design of the sampling process is very important to make the 

resulting sample representative of the population. According to Tan (2004), 

the trade-off between cost and precision in determining sample size may be 

derived using the Central Limit Theorem (The central limit theorem (CLT) 

The central limit theorem states that if you have a population with mean μ 

and standard deviation σ and take sufficiently large random samples from 

the population with replacement, then the distribution of the sample means 

will be approximately normally distributed, this theory could be applied in 

this study, were the sample taken from the population would represent the 

same data, with the same mean and standard deviation as the entire 

population of commercial facilities in Nablus (Chamber of Commerce and 

Industry, Nablus), but later noticed that data come from nonparametric and 

that because the sample was small and a Small sample  simply don't contain 

enough information to let you make reliable inferences about the shape of 

the distribution in the entire population (Motulsky,2013) 

about:blank


30 

 

3.6 Questionnaire 

The distributed questionnaire designed to get all the data needed for 

this research, it is designed by reviewing past studies with similar 

objectives based on ‘ENERGY PERFORMANCE OF HOTEL 

BUILDINGS questioner. It was acknowledged that a very long 

questionnaire with many details is likely to deter some building from being 

included in the survey. a very short one will inevitably fail to collect the 

necessary data. Therefore, the principle is to keep it succinct but still able 

to grasp the essentials. Ultimately, the questionnaire was finalized as a 

result of careful evaluation of these factors. The questionnaire was 

distributed manually by the researcher; extra questions were added based 

on the type of the facility (Appendix I) 

The questionnaire was divided to many sections which were: general 

information section which was designed in order to collect information 

about the establishment including its type, if there were any factors which 

increase energy use and bills for calculating energy use which will be used 

to calculate energy indicators ECI and EUI. 

The second section is the physical characteristics of the 

establishments which include information about the establishment itself in 

terms of size, number of workers, year of the establishment, number and 

type of equipment’s and period of use, taking into consideration the 

establishment type which has a great impact on those answers, this section 

also asks if an energy audit had been performed previously or not.  



31 

 

The last section was designed to collect information about the 

Operational characteristics of the facility including operating period, nature 

of work and occupancy rate  

3.7 Preliminary study 

A Preliminary testing is very important because by conducting a 

Preliminary study a lot of problem could be detected and what might face 

the respondent (bhattacherjee,2012) and then obtain validity and 

sustainability of the question (bhattacherjee,2012). A sample of hospitals 

were chosen; only private hospitals were included because governmental 

hospitals declined giving the data.  

then examined the questions in the questionnaire and a group of experts 

and arbitrator saw the questionnaire to be easy to answer. (Table 4) 

Table 4: Expert and arbitrator who reviewed the questionnaire 

Number Position 

4   Teaching staff at Al- Najah university 

2 Auditing of energy  ISO 



32 

 

3.8 Data analysis approach 

By using Conducting Quantitative Interviews for data collection. An 

average for ECI and EUI for non-industrial building will be calculated. An 

important step in conducting a survey is to check for data integrity. In the 

current study, this step was taken through site visit and conducting 

interview with the establishment engineers and accountants. 

Through site visits and questionnaire filling, the researcher focused 

on specific questions in order to collect correct information and clarify the 

differences between the chosen categories. For example, when collecting 

data from hotels, number of rooms was taken into consideration. However, 

when data were collected from schools, numbers of teachers, students and 

labs were taken into consideration because these factors affect energy use. 

In restaurants, asked about refrigeration rooms, fridges and stores. As for 

companies, asked about number of branches, stores, and warehouses, these 

questions were asked to explain expected results variability during data 

analysis.  

Conducting Quantitative Interviews 

Quantitative interviews are sometimes referred to as survey 

interviews because they resemble survey-style question-and-answer 

formats. They might also be called standardized interviews. The difference 

between surveys and standardized interviews is that questions and answer 

options are read to respondents rather than having respondents complete a 

questionnaire on their own. As with questionnaires, the questions posed in 



33 

 

a standardized interview tend to be closed ended. There are instances in 

which a quantitative interviewer might pose a few open-ended questions as 

well. This approach was adopted in order to be more clear and to find any 

other indicators (Rubin & Rubin. 1995). 

In quantitative interviews, an interview schedule is used to guide the 

researcher as he or she poses questions and answer options to respondents. 

An interview schedule is usually more rigid than an interview guide. It 

contains the list of questions and answer options that the researcher will 

read to respondents(Creswell, 1998) 

Table 5: Questions for each category 

Chapter four will present data analysis and results and discussion, 

which includes the effect of variables on energy indicators and we will talk 

CI &TI and t test and ANOVA test  

Name of the establishment  Specific questions  

Schools  Number of teachers  

Number of students 

Is there summer school 

Number of floors 

Number of scientific labs  

Companies  Number of stores 

Number of branches  

Numbers of reserved staff  

 

 

Hotels Number of single bed rooms and suites 

Number of floors 

Accessories from a wedding hall or 

conference room 

 



34 

 

3.9 T test & ANOVA test 

 Some conditions before performing the two tests are needed to be 

accomplished. For the T-test, population data should be normally 

distributed, and are comparing equal variances of the population. While for 

ANOVA tests, samples that are to be used are selected independently and 

randomly. should also assume that the population are taking the samples 

from is normal and have equal standard deviations. 

 

 

 

 

 

 

 

 

http://www.differencebetween.net/language/difference-between-data-and-information/


35 

 

Chapter four 

Results and discussion 

4.1 Introduction 

The questionnaire had been explained, data were analysed using SPSS 

programme. In order to spot the difference, the questionnaire part and the 

interview part were presented as followed  

4.2 Questionnaire analysis 

First section: general information 

In the first section we were only interested in the general information 

regarding the establishment including its type, nature of work, type of fuel 

used, electric and fuel bills. As seem in table 6 this information was used to 

calculate energy indicators (EUI, ECI). 

 

 

 

 

 

 

 



36 

 

Table 6: ECU and EUI for all categories 

No Name No Employee Size 

(m
2

) 

EUI 

Kj/ m
2

 

ECI 

Nis/ m
2

 

Occupancy 

rate % 

energy 

kj 

 19795.9732 %40.00 6.129542 31.27326 633 7 عانم انحُشُرت 1

2 awad center 3 500 48.70872 10.08 50.00% 24354.3624 

 44311.4094 %50.00 10.78824 52.13107 850 3 مذرضة االمهات 3

 34753.2886 %65.00 10.6875 54.30201 640 8 انبُث انطعُذ 4

 9852.88591 %20.00 13.59333 65.68591 150 4 انمركس انظُىٍ 5

 37671.9463 %55.00 13.85455 68.49445 550 5 طُذنُة طبرٌ  6

 19315.9732 %25.00 14.37037 71.54064 270 4 انحطاب نهمىبُهُا 7

 713923.862 %60.00 16.00441 98.47226 7250 33 شركة هضبة اندىُذٌ 8

 31853.9597 %40.00 21 106.1799 300 2 شركة انبُث انعظرٌ 9

2درَم مىل 10  5 120 110.9799 21.5 35.00% 13317.5839 

 35888.8591 %35.00 19.70313 112.1527 320 3 شركة طىقان 11

وىار طُذنُة اال 12    120 114.2013 22.16667 60.00% 13704.1611 

 121733.154 %55.00 24.84 121.7332 1000 9 اضىاق انظىانحٍ 13

 1275704.7 %70.00 26.4 127.5705 10000   خامعة انقذش  14

 116318.4 %50.00 27.26706 136.8452 850 10 درَم ضىبر ضحىر 15

 21535.5705 %40.00 28.73333 143.5705 150 7 اركاد نهعقار 16

 46357.878 %20.00 47.015 231.7894 200 15 وضانكى نهطباعه 17

 37061.8792 %30.00 37.52 247.0792 150 3 شركة انطهطان 18

19 Dand 4 100 296.3114 61.32 20.00% 29631.1409 

مدمىعة شركات ضامر  20

 شبُطه

6 250 729.1889 150.62 40.00% 182297.235 

 79440 %25.00 161.9 794.4 100 5 شركة فُروزٌ 21

 bci 5 220 824.1123 170.5455 40.00% 181304.698شركة  22



37 

 

 6389961.34 %75.00 202.7754 983.071 6500 44 بال عقار  23

 101496.483 %25.00 150.9 1014.965 100 4 شركة وقهُات انقىاضمٍ 24

 619242.462 %30.00 168.0083 1032.071 600 10 اضىاق ابى بكر انحدارَه 25

 1608270.46 %40.00 181.525 2010.338 800 16 ضىب ضحىر 26

 20295974.6 %50.00 25.62791 48323.75 420 15 شركة بططامٍ نالدوَه 27

 55048.5906 %60.00 27.6 137.6215  400 4 شركة انحح عهٍ 28

 4497784 %89 219.7119 1285.081 3500 153 مطحشفً االودُهٍ  36

 9658590.32 %90 190.9574 965.859 10000 450 مطحشفً انىداذ  37

 3751246 %75 173.5757 937.8115 4000 207 مطحشفً وابهص  38

 6716975.6 %80 303.1691 1865.827 3600 230 مطحشفً انعربٍ  39

 402927.709 %30 59.77673 335.7731 1200 22 االماكه 41

 1835809.92 %60 684.6069 5245.171 350 30 ورد 42

 202777.6 %20 98.84533 506.944 400 10 زادووا 43

 1725153.54 %70 934.1 6461.249 267 9 دومُىىز بُحسا 44

 2286177.73 %80 337.2421 2286.178 1000 70 انف نُهه ونُهه 45

 363408.738 %60 103.4057 519.1553 700 26 جشهُى  46

 1403221.39 %35.00 271.0555 467.7405 3000 32 انقظر  47

 443928 %70.00 35.866 177.5712 2500 12 ضهُم افىذٌ 48

 78912 %85.00 34.25 164.4 480 2 عطاف  49

 1200000 %70.00 82.5 480 2500 15 انقهعه 50

 73015 %50.00 17.14 85.9 850 6 االضراء 51

 459763.2 %75.00 88.3392 459.7632 1000 16 فىذق انُاضمُىه 52

 34522.8 %75 2.79686 16.67768 .2070 23.50 عراق انحاَه انثاوىَة 53

 52694.4 %70 2.130242 12.61537 4177 26.00 انُرمىك انثاوىبة 54

 35242.8 %75 1.443647 8.701926 4050 25.50 مذرضة انعائشُة انثاوىَة 55

نثاوىَةضمُر ضعذ انذَه ا 56  35.50 4091 12.68404 2.14747 70% 51890.4 

 42061.2 %80 1.656419 9.850398 4270 30.00 كمال خىبالط انثاوىَة 57



38 

 

 39040.8 %80 2.162663 12.75843 3060 23.50 انفاطمُة انثاوىَة 58

خمال عبذ انىاطر  59

 انثاوىَة

27.50 6445 10.00068 1.676631 85% 64454.4 

حُة انثاوىَة نهبىُهانظال 60  28.00 7430 18.2463 3.09776 75% 135570 

 32590.8 %70 1.469959 8.808324 3700 26.00 انظالحُة انثاوىَة نهبىات 61

عمرو به انعاص  62

 انثاوىَة

18.00 3900 10.24892 1.730638 70% 39970.8 

 53829.6 %60 2.651459 15.83224 3400 24.50 ضعذ طاَم االضاضُة 63

 17287.2 %65 2.88761 17.2872 1000 10.50 خمُهة بى حُرد االضاضُة 64

 33867.6 %70 5.029491 29.70842 1140 26.50 فهمٍ انظُفٍ 65

 17030.4 %70 3.510245 15.48218 1100 10.50 انىظامُة االضاضُة نهبىات 66

 24519.6 %70 2.731206 16.24891 1509 19.00 بالل به رباذ 67

 17088 %75 1.879568 11.18325 1528 13.00 انخىطاء االضاضُة 68

 18118.8 %75 0.60598 3.62376 5000 16.00 ابه ضُىا 69

 23810.4 %70 1.99854 11.9052 2000 21.50 فذوي طىقان 70

 8736 %75 2.086897 13.09745 667.00 12.00 طارق به زَاد 71

2200.0 16.50 عادل زعُحر االضاضُة 72

0 

17.17364 2.900482 70% 37782 

1577.0 18.50 عمر انمخحار االضاضُة 73

0 

13.86278 2.337514 75% 21861.6 

 

 

 



39 

 

After data collection, out of 78 questionnaires, 73 were collected and 

completed, 5 were not completed.  

Here, was noticeable the big variability in the data. So, after several 

trials including classifying by size, the results were not understandable yet, 

when was classified by category there was a big variability, this could be 

related to the wide differences even in the same category. For example, 

companies had big differences in a number of employee, branches, age 

equipment, and heating and conditioning system which gave the great 

variability. In the second section, the energy performance indicators were 

calculated.  

In the case of hospitals, the sample was small, but was noticeable 

that the nature of hospitals was the same, same equipment; same heating 

and conditioning system, all used the same energy sources (Gas, diesel, 

electricity). That’s why other factors like size and occupancy rate were 

obvious, and their effects on the energy indicators were clear which was 

seen in the statistical analysis.  

In cases of restaurant and hotels, clear differences were spotted, 

which is related to the differences in the same category, for example, some 

hotels target college students only for sleeping, like in Israa and Assaf 

hotels. However, AlQasr hotel despite its big size, other factors affected the 

hotel, like its old establishment, and the shortage of its visitors. As for 

restaurants, the nature of the work used equipment and the age of the 

building had an effect on the results. In cases of schools, only 



40 

 

governmental schools were selected, which makes it easier to compare and 

it would be difficult to get data from private schools.  

After data analysis noticed that the analysis should be based on the 

Conditioned Area (size*Occupancy rate). New EUI and New ECI were 

calculated based on the conditioned Area. We also took into consideration 

the effect of other factors on New EUI and New ECI. 

Second section: Physical characteristics of the establishment 

In the second section, we were focusing on questions like the age of the 

facility, size, number of employees, type of glass, number of equipment, 

number of work hours, and questions regarding energy conservation.  

Through these questions, it was expected what the factors and how they 

affected the energy performance indicators. But, the differences were clear, 

for example, hotels had big halls and equipment, restaurants had different 

working hours, type of glass, energy saving strategies, which gave very 

important results.  

Schools had a big size, applied many strategies to decrease its energy 

consumption according to the ministry of education plans, which also 

included changing the lighting to LED, and future plans for applying sun 

cells on the school’s roof. As for companies, it was noticed that some 

companies had the greatest number of employee, greatest size, but, their 

consumption was less which is related to their nature of work and applying 

strategies for reducing energy consumption.  



41 

 

Third section: Operational characteristics of the facility 

In the third section asked regarding the occupancy rate and number of 

workers in the first shift which is a very important for this study.  

4.3 Interview analysis 

Upon finishing the data collection and interview process, data 

analysis was performed in order to answer the research questions. To 

collect more data and to obtain more information about the research 

problem, by using a semi-structured interview analysis. In the first step, 

data were taken from people in charge; the questionnaire was analysed 

using a strategic process in which every category was analysed separately 

including related questions and output (Table 7) in order to reach our 

objectives. (See Appendix III) 

By correlating the interview questions with the overall energy 

consumption, data were analysed using SPSS. The analysis was separate 

for each category, with regards to the factors affecting their energy 

consumption. As seen in table (7). For example, a number of students, size 

and number of laboratory had a role in increasing energy consumption in 

schools. But, in companies, a number of branches and number of stores 

questions indicated high energy consumption. As for hotels, a number of 

rooms, accessory conference rooms, and halls had a roll in increasing 

energy consumption. 

 



42 

 

Table 7: Questions related to each category 

Name 

establishmen

t 

Specific questions Affect  

Significance α 

Schools  Number of teachers  

Number of students 

Is there summer school 

Number of floors 

Number of scientific labs  

No Effect.767 

No Effect.644 

No Effect.9 

Effect.022 

Effect .032 

Companies  Number of stores 

Number of branches  

Numbers of reserved staff  

Effect .014 

Effect .031 

No Effect.184 

Hotels Number of single bed rooms and suites 

Number of floors 

Accessories from a wedding hall or 

conference room 

Effect.042 

Effect .030 

Effect.019 

 

 

4.4 Quantitative analysis and discussion 

Introduction  

By calculating the energy indicators (EUI) which is energy 

consumption divided by the area of the establishment (condition area = 

actual area*average occupancy rate).  

 Using Mini tab programme, T-test analysis was performed; 

confidence interval and descriptive statistics were found. Regression 

analysis was also used but did not give satisfactory for some of the 

equations, this might have related to the high variability between variables. 

ANOVA test was performed to identify significance relationship. Finally, 

by calculated Tolerance Interval (TI) in order to identify the highest and the 

lowest values. Data were found to be not normally distributed 



43 

 

As shown below in the flow chart, the analysis process was as 

follows: using Minitab to analyzes and find the prediction equation for 

indicators, in case the equation failed we went to energy as a primary 

variable for the indicators, which can use to find the correlation between 

actual area and energy (actual area is the of area multiplied by occupancy 

rate), can also prove that actual area has an impact on energy indicators. 

 

EUI for companies 

Normal probability plot shows high variability which indicates that 

other immeasurable variables like nature of work and type of equipment 

could have an effect on EUI  

One-Sample T test: new EUI  

A confidence interval for companies EUI mean value was calculated 

using a t-test, the test assumes unknown variance and data not in normal 

distribution. Results are presented in table 8. It is shown that the average 



44 

 

EUI value equals to 1086 kJ/m
2
, with a 95% confidence interval varies 

between 506 to1630 kJ/m
2
. This value is equivalent to 30.2 litres of diesel 

per m2. But it should be kept on mind that the original consumption will 

exceed 90 Litre of diesel per m
2
since the efficiency of electricity generation 

is about 35%. 

Table 8: descriptive statistics for Companies EUI 

Variable N Mean SD SE Mean 95% CI 

EUI new 26 1068 1391 273 (506, 1630) 

Tolerance interval  

As shown in the normal probability plot figure (4), all values are 

confined between the highest and lowest value by a 95% ratio so the 

tolerance interval between (78.18, 5025.84) (Table 9) Upper tolerance 

interval and lower tolerance interval so far between each other 

Upper tolerance interval and lower tolerance interval so far between 

each other, which is related to the differences in the nature of work for each 

establishment, more specific, detailed and precise classification are needed 

for future study. That’s why the data shown in the graph was nonparametric  

Normality Probability Plot with Nonparametric data: 

 Nonparametric tests have less power than parametric tests, and the 

difference is noticeable with tiny samples. Unfortunately, normality tests 

have little power to detect whether or not a sample comes from a Gaussian 

population when the sample is tiny. Small samples simply don't contain 

enough information to let you make reliable inferences about the shape of 



45 

 

the distribution in the entire population (Motulsky,2013). Normality tests 

work well with large samples, but here we have a small sample which 

didn’t contain enough data to let us make reliable inferences about the 

shape of the distribution of the population from which the data were drawn. 

So normality tests don't answer the question. Normality tests ask the 

question of whether there is evidence that the distribution differs from 

Gaussian. But with huge samples, normality testing will detect tiny 

deviations from Gaussian, differences small enough so they shouldn't sway 

the decision about parametric vs. nonparametric testing(Motulsky,2013). 

That’s why the sample was distributed as shown in (Figure 4) 

 

 

 

Figure 4:  Normal probability plot for Companies EUI 

Table 9:  Tolerance interval for Companies EUI 

N Mean  SD Lower  Upper 

26 1067.61 1391.45 78.18 5025.84 

Regression for EUI V.S Conditioned Area for company  

Assumed the relationship to be linear and calculated the regression 

equation, the values were R square =.022 and P value =.4, the significance 

was higher than .05 which means that cannot consider conditioned area as a 

predictor for EUI, as shown in (Table 10) 



46 

 

Table 01:  Regression for EUI vs. Conditioned area 

Multiple R 0.151649 

R Square 0.022997 

Adjusted R Square -0.01608 

Standard Error 1381.444 

Observations 27 

One-way ANOVA: New EUI versus condition area  

After using ANOVA test to prove the effect of condition area which 

is all area multiplied by occupancy rate, p value was .4 which is more than 

p value in the regression and there is no significant relationship between 

conditioned area and EUI, as shown in (Tables11 and 12) 

Table 00: one way ANOVA EUI for Company 

ANOVA     

  SS MS F Significan

ce F 

Regression 1123030 112303 0.5884 0.450196 

Residual 4770971 19083   

Total 4883274       

Table 02: Standard Error 

 Coefficients Standard 

Error 

t Stat P-value Lower 95% Upper 95% 

Intercept 1133.435 290.8873 3.896476 0.000646 534.3418 1732.529 

conditione

d area 

-0.12026 0.156766 -0.76712 0.450196 -0.44312 0.202608 

After applying the regression equation, the percentage of error was very 

high which means that the linearity assumption is not valid. This indicates 

that the relationship was not linear and the equation is not useful. So, by 

drawing a graph (Figure 5) representing only the relationship between EUI 



47 

 

and conditioned area, the graph was binomial and we can use to predict 

future energy consumption.  

Variables other than conditioned area and occupancy rate which 

could have an impact, such as, number of employee load factor, working 

hours, building age, applying energy conservation measures caused 

variation and deviation from linearity assumption; table 13 shows the 

differences between the actual EUI and the predicted one. 

 

 

 

Figure 5: EUI kj/m
2

 and conditioned area m
2

 

Table 03:  Predicted EUI VS Real measures 

Observation Predicted new 

euikj/m
2
 

Residuals new eui (kj/m
2
 

1 1131.030164 350.5268833 1481.557047 

2 1130.428873 2047.171127 3177.6 

3 1130.428873 2929.430456 4059.859329 

4 1129.827583 -801.3980526 328.4295302 

5 1128.625002 30.32194765 1158.94695 

6 1128.384486 -811.299155 317.0853308 

7 1128.023712 -304.4263961 823.5973154 

8 1126.21984 -767.2936657 358.9261745 

9 1125.317905 -839.1553393 286.1625652 

10 1124.776743 -934.4411727 190.3355705 

11 1122.852614 937.4280452 2060.280659 

0
1000
2000
3000
4000
5000
6000

2
0

2
5

4
0

4
5

6
7

.5 8
8

1
1

2

1
3

7
.6

2
1

4
7

6
5

2
5

0

3
0

2
.5

4
1

6

4
2

5

4
3

5
0

7
0

0
0

EU
I 

conditioned area 

new eui

Predicted new eui



48 

 

12 1121.409517 701.5628323 1822.972349 

13 1119.96642 -799.5301781 320.4362416 

14 1119.004355 -853.5546906 265.4496644 

15 1116.88523 -716.8852299 400 

16 1111.78887 2328.447028 3440.235897 

17 1103.370804 -1005.953354 97.41744966 

18 1102.985978 -1024.80283 78.18314832 

19 1097.057254 -972.5218944 124.5353597 

20 1094.952738 3930.892455 5025.845192 

21 1083.407961 -999.8664022 83.54155911 

22 1082.325638 -978.0634987 104.2621398 

23 1082.325638 -808.6352855 273.6903529 

24 1067.293378 -845.9603697 221.3330079 

25 610.3126474 -446.1922194 164.120428 

26 547.1771518 763.5841492 1310.761301 

27 291.6287171 -109.3851888 182.2435283 

As mentioned in EUI for companies the t-test shows confidence interval 

between (78.18, 5025.84). In the regression model here noticed the 

weakness of the linear assumption between conditioned area and EUI. 

Also, performed one-way ANOVA to prove that there is a direct effect 

between Conditioned Area and EUI, we found no significant result (p=.4), 

so the size has no effect on EUI which should be taken in consideration as 

shown above   

After establishing the relationship between conditioned area and EUI, 

figure 6 present a graph that represents this relationship, the graph presents 

a polynomial shape, the equation is: Y = 2.8093x2 - 123.77x + 2054.7 The 

percentage of R² was 8%. Thus, we cannot predict the future EUI based on 

the conditioned area 



49 

 

 

Figure 6: New EUI and conditioned area in companies 

ECI (Energy Cost Index) 

The factor affecting this indicator were studies, which is related to 

the energy price for every square meter, big variability was found, which 

indicates that some factors had a main effect and some do not, which will 

be seen later in this chapter.  

One-Sample T: ECI new  

Confidence interval for companies ECI mean value was calculated 

using t-test, the test assumes unknown variance and data normal 

distribution. Results are presented in table (14). It is shown that the average 

ECI value equals to 179.7 Nis/m
2
.year, with a 95% confidence interval 

varies between 96.6 to 262.8 Nis/m
2
.year 

Table 04: Descriptive statistics for Companies ECI 

Variable  N Mean  SD SE Mean  95% CI 

ECI new  26 179.7 205.8 40.4 (96.6, 262.8)  

 

y = 2.8093x2 - 123.77x + 2054.7 
R² = 0.0811 

0
1000
2000
3000
4000
5000
6000

new eui 



50 

 

Tolerance Interval: NEW ECI  

 

Figure 7: Normal Probability plot for Companies ECI 

all values are confined between the highest and lowest value by a 95% 

ratio. (Table 15) 

Table 05:  Tolerance  probability for Companies ECI 

N Mean SD Lower Upper 

26 179.70 205.772 15.324 647.6 

ECI Regression Analysis (Conditioned Area V.S ECI) 

Assuming a linear relationship, the regression equation presented in results 

(Table 16, 17, and 18), r²= .02 it is shown that the values is low and the 

modelled equation does not explain the data variability. 

Table 06:  Regression ECI and Conditioned area 

Multiple R 0.152321 

R Square 0.023202 

Adjusted R Square -0.01587 

Standard Error 204.2845 

Observations 27 

 



51 

 

Table 07: ANOVA for Companies ECI 

ANOVA     

  SS MS F Significance F 

Regression 24781.28 24781.28 0.593817 0.448168 

Residual 1043304 41732.17   

Total 1068085       

Table 08: Standard error 

  Coefficients Standard 

Error 

t Stat P-value Lower 

95% 

Upper 

95% 

Intercept 189.4704 43.01568 4.404684 0.000174 100.878 278.0629 

conditioned 

area 

-0.01786 0.023182 -0.7706 0.448168 -0.06561 0.029881 

Figure 8 represent the relationship between the predicted and actual ECI in 

order to show the difference and the percentage of error (Table 19). The 

assumption of a linear relationship between conditioned area and 

occupancy rate also failed.  

Table 09: Actual ECI VS predicted ECI 

     

Observation Predicted new eci 

nis/m
2
 

Residuals Standard 

Residuals 

new eci 

nis/m
2
 

1 184.9473 -169.623 -0.846 15.32 

2 185.0044 -164.844 -0.822 20.16 

3 181.8782 -160.302 -0.800 21.56 

4 182.039 -165.597 -0.826 16.44 

5 188.9345 -120.968 -0.603 67.966 

6 184.0666 -158.876 -0.793 25.190 

7 188.2646 -130.783 -0.652 57.481 

8 111.7617 -85.0877 -0.424 26.674 

9 187.3268 -134.827 -0.673 52.5 

10 188.7202 -127.292 -0.635 61.42 

11 187.4697 -131.175 -0.654 56.29 

12 188.1842 -151.24 -0.755 36.944 

13 179.6452 -134.482 -0.671 45.16 

14 64.42191 -26.7076 -0.133 37.714 

15 181.8782 -127.344 -0.635 54.534 

16 188.3986 -116.565 -0.5819 71.83 



52 

 

17 188.7559 46.31912 0.2312 235.07 

18 188.6666 -63.5999 -0.3175 125.06 

19 189.1132 117.4868 0.5865 306.6 

20 187.684 188.866 0.9428 376.55 

21 189.0238 458.5762 2.2892 647.6 

22 187.8984 238.4652 1.1904 426.36 

23 102.3831 167.9841 0.8385 270.36 

24 189.0238 414.5762 2.0695 603.6 

25 186.2549 373.7729 1.8659 560.02 

26 183.7539 270.0586 1.3481 453.81 

27 187.012 -106.792 -0.533 80.220 

 

 

Figure 8: Actual ECI VS predicted ECI 

After establishing the relationship between conditioned area and ECI, 

figure 9 shows a graph that represents this relationship, the graph presents a 

polynomial shape, and the equation is: 

Y = 0.6955X
2
 - 29.295x + 407.62 

 

 

 

0
100
200
300
400
500
600
700

2
0

2
5

4
0

4
5

6
7

.5 8
8

1
1

2

1
3

7
.6

2
1

4
7

6
5

2
5

0

3
0

2
.5

4
1

6

4
2

5

4
3

5
0

7
0

0
0

EC
I N

EW
  

conditioned area  

new eci

Predicted new eci



53 

 

 

Figure 9: new ECI VS conditioned area 

The percentage of R² was 18.3%. Thus, cannot predict the future ECI 

based on the conditioned area. After failed to prove that conditioned area 

has a direct effect on both energy indicators, connected conditioned area 

with energy, where energy is the main factor in both indicators and any 

change to energy can affect energy indicators. So, study will define the 

relationship between energy and conditioned area and find the regression 

equation.  

Regression for Energy V.S Conditioned Area for company 

Regression analysis was done between energy level and conditioned 

area; after find that conditioned area affect energy using a regression 

equation and by calculating the standard error (Table 20-23). R² equals 

0.38 and adjusted R square equals 0.36. a significant relationship was 

found (p=0.0005). the conditioned area lies between the values (218.5855 

and 685.7852)m
2
.  

 

0
100
200
300
400
500
600
700

2
0

2
5

4
0

4
5

6
7

.5 8
8

1
1

2

1
3

7
.6

2
1

4
7

6
5

2
5

0

3
0

2
.5

4
1

6

4
2

5

4
3

5
0

7
0

0
0

EC
I 

 

conditioned area 

new eci



54 

 

Table 21: Energy VS condition area 

Regression Statistics 

Multiple R 0.623426 

R Square 0.38866 

Adjusted R Square 0.364207 

Standard Error 999502.5 

Observations 27 

Table 20:  ANOVA Energy for company 

  SS MS F Significance 

Regression 1.59E+13 1.59E+13 15.8938 0.000513 

2.5E+13 9.99E+11   

Total 4.09E+13       

Table 22:  Standard error 

  Coefficients Standard 

Error 

t Stat P-value Lower 95% Upper 95% 

Intercept 98920.32 210462.7 0.470013 0.64242 -334536 532376.4 

conditioned 

area 

452.1853 113.4234 3.986703 0.000513 218.5855 685.7852 

Equation Y=98920+452x 

Table 23:  Predicted and actual energy 

RESIDUAL OUTPUT    

Observation Predicted 

energy KJ 

Residuals Standard 

Residuals 

Calculated 

energy KJ 

1 213413.7 -193618 -0.19755 19795.9732 

2 211966.7 -187612 -0.19142 24354.3624 

3 291099.1 -246788 -0.2518 44311.4094 

4 287029.4 -252276 -0.2574 34753.2886 

5 112485.9 -102633 -0.10472 9852.88591 

6 235706.4 -198034 -0.20206 37671.9463 

7 129442.8 -110127 -0.11236 19315.9732 

8 2065927 -1352003 -1.37946 713923.862 

9 153182.6 -121329 -0.12379 31853.9597 

10 117912.1 -104595 -0.10672 13317.5839 

11 149565.1 -113676 -0.11599 35888.8591 

12 131477.7 -117774 -0.12017 13704.1611 

13 347622.3 -225889 -0.23048 121733.154 

14 3264218 -1988513 -2.0289 1275704.7 



55 

 

15 291099.1 -174781 -0.17833 116318.4 

16 126051.4 -104516 -0.10664 21535.5705 

17 117007.7 -70649.9 -0.07208 46357.878 

18 119268.7 -82206.8 -0.08388 37061.8792 

19 107964 -78332.9 -0.07992 29631.1409 

20 144138.9 38158.38 0.038933 182297.235 

21 110225 -30785 -0.03141 79440 

22 138712.6 42592.06 0.043457 181304.698 

23 2303324 4086637 4.169643 6389961.34 

24 110225 -8728.47 -0.00891 101496.483 

25 180313.7 438928.8 0.447844 619242.462 

26 243619.6 1364651 1.392369 1608270.46 

27 161150.7 -106102 -0.10826 55048.5906 

Figure9 shows that the actual and predicted energy almost match. 

However, after studying the differences can see that the reason was related 

to the nature of the company, number of employee and the system that the 

company follows.  

 

Figure 9:  The relationship between actual energy kj and the predicted energy kj. 

 

 

0

1000000

2000000

3000000

4000000

5000000

6000000

7000000

2
0

2
5

4
0

4
5

6
7

.5 8
8

1
1

2

1
3

7
.6

2
1

4
7

6
5

2
5

0

3
0

2
.5

4
1

6

4
2

5

4
3

5
0

7
0

0
0

En
er

gy
  

conditioned area  

energy

Predicted energy



56 

 

EUI for Hospitals  

One sample T test: 

Number of hospitals in this sample was 6, only 4 answered the 

questionnaire, as presented in table (24), it is shown that the average EUI 

value equals 559 kJ/m
2
, with a 95% confidence interval varies between 635 

to 2415 kJ/m
2
 average confidence interval.  

Table 24: Descriptive Statistics for Hospitals EUI 

Variable  N Mean  SD  95%CI 

New EUI 4 559 280 (635,2415) 

Tolerance Interval: new EUI   

As shown in Table 25, all values are confined between the highest and 

lowest value by a 95% ratio, also here we note due to lack of sample 

number of hospital there is no figure for tolerance interval  

Table 25: Probability for Hospital EUI 

N Mean  SD Lower  Upper 

4 1524.94 559.11 1073.17 2332.28 

The P value was much higher than .05, which indicates high 

variance, differences in conditioned area and differences in energy 

indicators. (Table 26-29) show an r²=.31 which is a very low value. 

 

 

  



57 

 

Regression for EUI V.S Conditioned Area for Hospital 

Table 26: Regression Statistics 

Regression Statistics 

Multiple R 0.559338 

R Square 0.312859 

Adjusted R Square -0.03071 

Standard Error 567.6339 

Observations 4 

Table 27: ANOVA EUI for Hospital 

ANOVA     

  SS MS F Significance F 

Regression 293406.1 293406.1 0.91061 0.44066208 

Residual 644416.4 322208.2   

Total 937822.5       

Table 28: Standard errors 

  Coefficients Standard 

Error 

t Stat P-value Lower 95% Upper 95% 

Intercept 1993.546 567.1802 3.514838 0.072279 -446.83282 4433.926 

condition

ed area 

-0.10416 0.109155 -0.95426 0.440662 -0.5738186 0.365494 

Table 29: Residual Output 

RESIDUAL OUTPUT    

Observation Predicted new 

eui 

Residuals Standard 

Residuals 

new eui 

1 1669.081 -225.17 -0.48583 2332.283 

2 1056.086 17.09023 0.036874 1250.415 

3 1681.06 -430.644 -0.92917 1443.911 

4 1693.559 638.7239 1.378131 1073.177 

Figure 10 shows a high variance and high percentage of error between 

calculated and actual EUI . 



58 

 

 

Figure 10: Predicted EUI kj/ m
2

VS Real measures. 

After establishing the relationship between conditioned area and hospital 

EUI, figure 11 present a graph that represent this relationship, the graph 

presents a polynomial shape, the equation is: 

y = 177.78X
2
 - 1247.3x + 3309.8 

The percentage of R² was 81.9%. Thus, we can predict the future EUI 

based on the conditioned area. 

 

Figure 11: new EUI VS conditioned area in Hospitals. 

 

 

0

500

1000

1500

2000

2500

2880 3000 3115 9000

 n
e

w
 E

U
I 

 

conditioned area 

new eui

Predicted new eui

y = 177.78x2 - 1247.3x + 3309.8 
R² = 0.8196 

0

500

1000

1500

2000

2500

2880 3000 3115 9000

new eui 



59 

 

ECI FOR HOSPITAL 

One-Sample T Test 

Confidence interval for hospitals ECI mean value was calculated using t-

test; the test assumes unknown variance and data normal distribution. 

Results are presented in table (30). It is shown that the average ECI value 

equals to 267.4 kJ/m
2
, with a 95% confidence interval varies between 146.8 

to 387.9kJ/m
2
.  

Table 31: Descriptive Statistics for Hospitals ECI 

Variable N Mean SD SE Mean 95% CI 

ECI new 4 267.4 75.7 37.9 (146.8, 387.9) 

Tolerance Interval: ECI NEW  

As shown in the normal probability plot (Figure 12), all values are confined 

between the highest and lowest value by a 95% ratio.  

 

Figure 12: Normal Probability plot for Hospitals. 

After assumed a linear relationship and applied the regression 

equation the result shown in Table (31-34) r²=.83 (p=.08) which is slightly 

higher than .05. The equation was: 



60 

 

Table 30: Regression for ECI V.S Conditioned Area for hospitals 

Multiple R 0.913229 

R Square 0.833988 

Adjusted R Square 0.750981 

Standard Error 76.66505 

Observations 4 

Table 32: ANOVA ECI for hospital  

ANOVA     

  SS MS F Significan

ce F 

Regressio

n 

59053.23 59053.23 10.04729 0.086771 

Residual 11755.06 5877.53   

Total 70808.29       

Table 33: Standard Error 

  Coefficien

ts 

Standar

d Error 

t Stat P-value Lower 

95% 

Upper 

95% 

Intercept 426.6649 76.6037

8 

5.56976

3 

0.03075

6 

97.0654

5 

756.264

3 

conditioned 

area 

-0.04673 0.01474

3 

-

3.16975 

0.08677

1 

-

0.11016 

0.01670

2 

Table 34: RESIDUAL OUTPUT 

 

Observation Predicted 

new eci 

Residuals Standard 

Residuals 

new eci 

1 292.082 86.87936 1.387921 378.9613 

2 286.4744 -55.04 -0.87928 231.4343 

3 281.1004 -34.233 -0.54688 246.8673 

4 6.093276 2.393721 0.03824 8.486997 

Figure 13 represent the relationship between the actual and the predicted 

ECI  



61 

 

 

Figure 13: The relationship Predicted ECI VS Real measures. 

After establishing the relationship between conditioned area and hospitals 

ECI, figure 14 present a graph that represent this relationship, the graph 

presents a linear shape, the equation is: 

y = -0.0467x + 426.66 

The percentage of R² was 83.3%. Thus, can predict the future ECI based on 

the conditioned area or based regression equation because R² Values are 

close together. 

 

 

 

 

Figure 14: New ECI and energy in Hospitals 

 

0

50

100

150

200

250

300

350

400

2880 3000 3115 9000

EC
I 

Conditioned area  

new eci

Predicted new eci

y = -0.0467x + 426.66 
R² = 0.834 

0

100

200

300

400

0 2000 4000 6000 8000 10000

new eci 



62 

 

Regression for Energy vs. conditioned area  

Regression analysis was done between energy level and conditioned 

area; did find that conditioned area affects energy using a regression 

equation and find the regression equation. R² equals 0.75 and adjusted R 

square equals 0.36.  (Table 35) 

Table 35: Regression Statistics 

Regression Statistics 

Multiple R 0.868627 

R Square 0.754513 

Adjusted R Square 0.63177 

Standard Error 1609898 

Observations 4 

ANOVA 

 MS F Significance F 

Regression 1.59318E+1

3 

6.14707808 0.131372798 

Residual 2.59177E+1

2 

  

Total    

The sample size here is 4, a small sample size will give us huge variability 

in data and large margin of error because of that regression equation is 

week but R
2
is high= 63%so the model fits the data. The higher the R

2
 

value, the better the model fits the data.  The p value (.13) was higher than 

.05 which indicates that conditioned area does not affect energy (Tables 35 

and 37).  

Table 37 shows the residual error. Energy = 2703121+767.55*conditioned 

area 

 



63 

 

Table 36: Standard error 

  Coefficients Standard Error P-value Lower 95% Upper 95% 

Intercept 2703121 1608610.816 0.23489455 -4218172.75 9624415 

condition

ed area 

767.5528 309.5806998 0.1313728 -564.465474 2099.571 

Table 37: RESIDUAL OUTPUT 

 

Observation Predicted energy  Residuals Standard 

Residuals 

1 5094047.837 -

596263.8375 

-0.453613373 

2 9611095.88 47494.44348 0.036131849 

3 5005779.269 -

1254533.269 

-0.954398091 

4 4913672.937 1803302.663 1.371879616 

Figure 15 represents the relationship between predicted energy and 

actual energy; it is noticeable that the two lines almost match with slight 

differences, which indicates that the nature of hospitals in the city is the 

same in terms of consumptions. The high error indicates that the sample 

does not take energy indicators in consideration.  

 

Figure 15: Predicted Energy kj VS Real measures kj. 

0

2000000

4000000

6000000

8000000

10000000

12000000

2880 3000 3115 9000

en
er

gy
  

conditioned area  

Predicted energy

energy



64 

 

EUI for restaurants &hotels 

One-Sample T test: new EUI  

Confidence interval for restaurants and hotels EUI mean value was 

calculated using t-test; the test assumes unknown variance and data normal 

distribution. As presented in table (38), the average EUI value equals to 

1053kJ/m
2
, with a 95% confidence interval varies between 366 to 1740 

kJ/m
2
.  

Table 38: Descriptive statistics for Restaurants and Hotels EUI 

Variable N Mean SD 95%CI 

condition EUI 12 1053 1082 (366,1740) 

Tolerance Interval: new EUI 

As shown in the normal probability plot (Figure 16), all values are confined 

between the highest and lowest value by a 95% ratio. (Table 39) 

 

Figure 16: Normal Probability plot for Restaurants and Hotels EUI. 

Here as shown in (figure 16) Upper tolerance interval and lower tolerance 

interval so far between each other, which is related to the differences in the 

nature of work for each establishment, more specific, detailed and precise 



65 

 

classification are needed for future study. That’s why the data shown in the 

graph was nonparametric 

Table 39: Probability Restaurants and Hotels EUI 

N Mean SD Lower Upper 

12 1052.95 1081.58 171.8 4000 

Regression Restaurants and Hotels EUI VS conditioned area  

Assuming a linear relationship, the results were (r²=.079, p=.19) which is 

higher than .05 (Table 40). The regression equation was Y= 1486.7-.322X 

Table 40: Regression 

Regression Statistics 

Multiple R 0.404369 

R Square 0.163514 

Adjusted R Square 0.079865 

Standard Error 1037.495 

Observations 12 

Table 41: ANOVA EUI for Restaurants and Hotels 

 

 

 

Table 42: Standard error 

ANOVA     

  SS MS F Significan

ce F 

Regression 2104110 2104110 1.954773 0.192309 

Residual 10763961 1076396   

Total 12868070       

 Coefficients Standard 

Error 

t Stat P-value Lower 

95% 

Upper 

95% 

Intercept 1486.796 431.2617 3.44755 0.006252 525.8853 2447.707 

conditioned 

area 

-0.32245 0.230628 -1.39813 0.192309 -0.83632 0.191423 



66 

 

We drew a graph (Figure 17) representing the relationship between 

the actual and the predicted EUI. The results were shown in table (43), 

noticed that high percentage of error, the results were much different than 

the actual values. 

Table 43: Predicted EUI VS Real measures 

Observation Predicted new 

eui kj/m
2
 

Residuals Standard 

Residuals 
new eui kj/m

2
 

1 1454.315 2545.685 2.573445 4000 

2 472.0167 111.3166 0.11253 583.3333 

3 1454.103 545.8965 0.551849 2000 

4 28.40082 353.0277 0.356877 381.4286 

5 897.056 352.944 0.356793 1250 

6 1386.356 -219.689 -0.22208 1166.667 

7 1148.225 188.1764 0.190228 1336.401 

8 922.5108 -668.838 -0.67613 253.6731 

9 1355.237 -1161.83 -1.17449 193.4118 

10 922.5108 -236.797 -0.23938 685.7143 

11 1349.755 -1177.96 -1.1908 171.8 

12 1244.96 -631.942 -0.63883 613.0176 

 

 

Figure 17: Predicted EUI kj/m
2

 VS Real measures kj/m
2

 

0

500

1000

1500

2000

2500

3000

3500

4000

4500

n
ew

 e
u

i 

conditioned area  

new eui

Predicted new eui



67 

 

After establishing the relationship between conditioned area and EUI, 

figure 18 present a graph that represent this relationship, the graph presents 

a linear shape, the equation is:  

Y=-0.3224x + 1486.8 The percentage of R² was 16.35%. Thus, we cannot 

predict the future EUI based on the conditioned area. 

 

Figure 18: New EUI and Energy 

Sample size here was 14, 12 answered the questioner. After 

analysing the data, noticed that confidence interval is good and regression 

equation is not good for to predict in the future trend of EUI and energy 

consuming in hotel and restaurants. Also, R
2
 here 16.35%. Also, type of the 

hotel and restaurant play a role in energy consuming and other building 

follow the main play a role. 

 

 

 

0

1000

2000

3000

4000

5000

0 1000 2000 3000 4000 5000

new EUI 



68 

 

ECI for RESTURANTCE &HOTEL 

One-Sample T: new ECI  

Confidence interval for restaurants and hotels ECI mean value was 

calculated using t-test; the test assumes unknown variance and data normal 

distribution. Results are presented in table (44). It is shown that the average 

ECI value equals to 232 kJ/m
2
, with a 95% confidence interval varies 

between 67.1 to 397.2 kJ/m
2
.  

Table 44: Descriptive Statistics for Restaurants and Hotels ECI 

Variable  N Mean  SD SE Mean  95% CI 

ECI new  12 232.2 259.7 75.0 (67.1,397.2)  

Tolerance Interval: New ECI  

As shown in the normal probability plot (Figure 19), all values are confined 

between the highest and lowest value by a 95% ratio. (Table 45) 

 

Figure 19: Normal probability plot for Restaurants and Hotels 

Also here the data shown in the graph was nonparametric in (Figure 19) 

Upper tolerance interval and lower tolerance interval so far between each 

other, which is related to the differences in the nature of work for each 



69 

 

establishment, more specific, detailed and precise classification are needed 

for future study. 

Table 45: Probability for Restaurants and Hotels 

N Mean SD Lower Upper 

12 232.169 295.744 34.28 774.44 

Regression ECI for RESTURANTCE &HOTEL vs conditioned Area  

Table 46 shows the regression statistics (r²=.09) 

Table 46: Regression Statistics 

Multiple R 0.301399 

R Square 0.090841 

Adjusted R Square -7.5E-05 

Standard Error 214.9157 

Observations 12 

One-way ANOVA: new ECI versus condition area  

By used ANOVA test to prove the effect of condition area which is the size 

multiplied by occupancy rate, p value was more than .05, which indicate no 

significant effect on energy consumption as shown in table (47-48) 

Table 47: ANOVAECI for Restaurants and Hotels 

  SS MS F Significance F 

Regression 46150.84 46150.84 0.999179 0.341082 

Residual 461887.6 46188.76   

Total 508038.4       

 

 



70 

 

Table 48: Standard error 

Table 49: RESIDUAL OUTPUT 

Observation Predicted new eci Residuals Standard 

Residuals 

new eci 

1 242.2694 -4.89986 -0.02391 237.3695 

2 96.79079 -20.6533 -0.10079 76.13746 

3 242.238 147.7275 0.720924 389.9655 

4 31.09118 24.05178 0.117375 55.14296 

5 159.7392 -85.9824 -0.4196 73.75675 

6 232.2045 0.172913 0.000844 232.3775 

7 196.9374 577.507 2.818288 774.4443 

8 163.509 -112.272 -0.5479 51.23714 

9 227.5959 -194.017 -0.94682 33.57843 

10 163.509 -45.6519 -0.22279 117.8571 

11 226.7841 -192.504 -0.93944 34.28 

12 211.2638 -93.4782 -0.45618 117.7856 

Figure 20 represent the relationship between actual and predicted ECI  

 

Figure 20: Predicted ECI nis/ m
2

 VS Real measures nis/ m
2

 

0

100

200

300

400

500

600

700

800

900

new eci Predicted new eci

  Coefficien

ts 

Standar

d Error 

t Stat P-value Lower 

95% 

Upper 

95% 

Intercept 247.0798 89.3352

8 

2.76575

9 

0.01993

2 

48.0284

2 

446.131

2 

conditioned 

area 

-0.04775 0.04777

4 

-

0.99959 

0.34108

2 

-0.1542 0.05869

3 



71 

 

After establishing the relationship between conditioned area and 

EUI, figure 21 present a graph that represent this relationship, the graph 

presents a polynomial shape 

The percentage of R² was 9%. Which indicate that the equation could not 

be used, and the results can’t be adopted. This might be related to the other 

underestimated factors, like wedding halls in Hotels, meeting rooms, 

swimming pools and number of rooms. Thus, can predict the future ECI 

based on the conditioned area. 

 

Figure 21: New ECI nis/ m
2

and conditioned area m
2

 

We considered energy a main factor affecting both energy indicators, 

that’s why considered a linear relationship and can apply regression 

statistics in order to get a relationship between energy and conditioned 

area.  

 

 

y = 9E-07x2 - 0.0517x + 249.13 
R² = 0.0909 

0

200

400

600

800

1000

0 1000 2000 3000 4000 5000

ECI NEW 



72 

 

Regression for Energy V.S Conditioned Area  

Regression analysis was done between energy level and conditioned 

area; find that conditioned area affect energy using a regression equation 

and by calculating the standard error. R² equals 0.55 and adjusted R
2
equals 

0.50 (Table 50-52) 

P value .005 is smaller than .05 and equation for regression is 

Y=301136.6825+424.9754109X 

Table 51: Regression statistics 

Regression Statistics 

Multiple R 0.744520852 

R Square 0.554311299 

Adjusted R Square 0.509742429 

Standard Error 542095.9833 

Observations 12 

Table 50: ANOVA Energy for Restaurants and Hotels 

  SS MS F Significance F 

Regression 3.65E+12 3.65489E+12 12.43719 0.005477 

Residual 2.94E+12 2.93868E+11   

Total 6.59E+12       

Table 52: Standard error 

  Coefficients Standard 

Error 

t Stat P-value Lower 

95% 

Upper 

95% 

Intercept 301136.6825 225336.2 1.336388246 0.211033 -200944 803217.1 

conditioned 

area 

424.9754109 120.5044 3.52663939 0.005477 156.475 693.4758 

Y=424.9x+301136 

 



73 

 

The data were represented in Table (53) and figure (22). It’s noticeable that 

the standard residual decreases and the data are proximal which proves our 

theory that conditioned area is a predictor for energy. 

Table 53: RESIDUAL OUTPUT 

Observation Predicted energy kj Residuals Standard Residuals kj 

1 343945.2747 58982.43 0.114115029 

2 1638577.954 197232 0.381590415 

3 344224.4295 -141447 -0.27366129 

4 2223247.115 -498094 -0.9636761 

5 1078392.139 1207786 2.336737876 

6 433513.635 -70104.9 -0.13563398 

7 747360.8639 655860.5 1.26891242 

8 1044843.652 -600916 -1.16260897 

9 474526.6502 -395615 -0.76540716 

10 1044843.652 155156.3 0.300185495 

11 481751.2321 -408736 -0.79079386 

12 619868.2407 -160105 -0.30975987 

 

 

Figure 22: The relationship between: Predicted Energy kj VS Real measures kj 

 

 

 

0

500000

1000000

1500000

2000000

2500000

0 1000 2000 3000 4000 5000

Predicted energy

energy



74 

 

For schools 

One-Sample T: new EUI  

Confidence interval for schools EUI mean value was calculated 

using t-test; the test assumes unknow