Designing And Applying A Pricing Mechanism For Industrial Applications Residual Heat:

(1)

0

Residual Heat:

Designing And Applying A Pricing

Mechanism For Industrial Applications

MSc Thesis

MSc Technology & Operations Management

University of Groningen, Faculty of Economics & Business

Author: Sander Markus (s1536214) Supervisor: Prof. dr. I.F.A. Vis Second assessor: M. van der Werf M.Sc.

(2)

(3)

2 Abstract

The demand for energy in the world is rising. Due to this rise in demand a lot of alternative energy sources become profitable. One of these energy sources is residual heat.

In the current situation the price of residual heat is coupled to the natural gas price. Because the price of natural gas can greatly fluctuate the price of residual also fluctuates. Therefore the goal of this research was to design an independent pricing mechanism.

To do so the factors that influence the price had to be identified. Within the academic literature little is known about these factors, therefore expert interviews were conducted. These interviews yielded fourteen factors of which most turned out to be cost related. These factors were implemented in a new pricing mechanism.

The designed pricing mechanism was verified by applying data from an existing case. The verifcation yielded a price that is lower and more stable than in the current situation.

After the verification the mechanism was validated in two ways, numerical and practical. Numerical validation consisted of a sensitivity analysis. This analysis yielded that the designed mechanism is most suited when large amounts of energy are required (from 30000 MWh/year and up). Another important aspect of the mechanism is that the supply and demand are part of it. This means the mechanism can be applied dynamically.

Practical validation consisted of expert interviews with an expert who has a lot of practical experience and and an expert from the academic field. Both validated the identified factors and the designed mechanism. In extend to this it was stated that the designed mechanism can be applied in cases which consists of one supplier and one or multiple consumers.

(4)

3

List of tables

Table 1: Overview of identified factors ... 20

Table 2: Notations and definitions of the variables ... 22

Table 3: Used input... 26

Table 4: Price with profit margins ... 28

Table 5: Price with discount when coupled to natural gas ... 28

Table 6: Identified & implemented factors ... 41

List of figures

Figure 1: Research design ... 11

Figure 2: Conceptual model fixed costs ... 18

Figure 3: Conceptual model variable costs ... 18

Figure 4: Conceptual model continuity ... 19

Figure 5: Conceptual model overview ... 19

Figure 6: Comparison mechanism to natural gas price ... 30

Figure 7: Comparison mechanism to oil price ... 31

Figure 8: Continuity modeled as a factor ... 32

Figure 9: Continuity modeled as a fine/discount ... 32

Figure 10: Combination of fine/discount and a factor ... 33

Figure 11: Price differences ... 33

Figure 12: Influence of energy consumption ... 34

Figure 13: Consumed energy with cost factors ... 35

Figure 14: Influence of the grade of the returned energy ... 36

Figure 15: Influence of different grades of returned energy on different consumption levels ... 37

(7)

6

1. Introduction

1.1. Introduction

In present time (the year 2013) there is almost nothing as important in the world as energy. Without a sufficient energy supply the world would crumble (Dincer, 1999). In despite of the current economic crisis the demand for energy is still rising. This causes a rise of the prices of crude oil, natural gas and electricity. Nowadays there are a lot of alternative energy sources (e.g., Solar or wind energy). These energy sources do not cause problems like acid rain or the greenhouse effect (Dincer, 1999). However there is another energy source that is already available since the first steam engine was build. This energy source is the residual heat produced by industrial processes. This research will focus on this source of energy.

A lot of industrial processes (e.g., power plants or the production of steel) produce a lot of heat which is still mostly discharged. A small portion of the produced residual heat in the Netherlands is used by the (local) government to warm houses in the wintertime. However, there are a lot of processes (e.g., greenhouses, drying processes) who require heat that still use traditional energy sources to produce their required heat. These processes could also make use of residual heat instead. In order to make use of the residual heat a company should be situated near a heat producing source (some kind of industrial process) within a 10 to 20 km radius (Korhonen, 2001). The heat will then be transported in the form of steam or hot water from the source to the company via pipelines.

The heat producing source and the company using the heat are two important stakeholders in the process. However there is a third party that cannot be neglected. The third stakeholder is the

government. The government already has set up strict regulations for selling heat to the private market (Warmtewet 2013), but has not yet come up with a same set of rules for the industrial market. They could influence this market by taxing the sold heat but also by giving subsidies (SDE+ regeling, 2013) to stimulate the use of renewable energy sources. All these stakeholders will have to collaborate to come up with a pricing mechanism for residual heat that will work for all involved parties.

(8)

7

1.2. Problem Definition

The problem that rises is that at this moment there is not a good way to price residual heat. In the present situation (the current pricing mechanism) the price of heat is a combination of a fixed duty and a variable duty. The variable duty consists of a discount percentage per megawatt hour (MWh) compared to the price of natural gas per MWh (Markus, 2013). The problem with this mechanism is that as the gas price alters so will the price of heat. This is a problem because residual heat is often formed in processes that do not use natural gas (e.g., the steel industry or waste incinerators).The pricing mechanisms that are currently in place for electricity and natural gas cannot directly be applied so a new mechanism has to be designed. Next to that it has to be investigated which factors of the current mechanisms can be applied to this mechanism and what the limitations of the mechanism are. In the Netherlands the Flexiheat project is started, part of their goal is to find a solution for these problems. This research fits into the Flexiheat project however is not on behalf of the project.

The aim of this research is to design an independent pricing mechanism for residual heat, this leads to the following research question.

- How can the price of residual heat be established?

To answer the main research question the factors that influence the price of residual heat have to be identified. When the factors that influence the price are determined the aim is to design a pricing mechanism for one setting. The setting will consist of one supplier and one consumer of heat. From this the following general sub-questions can be formed.

o What do the current pricing mechanisms in the energy sector look like?

o Which aspects of current pricing mechanisms in the energy sector can be applied? o Which factors influence the price of residual heat?

o How can a pricing mechanism be designed for a specific setting?

The mechanism will be verified by applying the data supplied by a pre-selected case (which consists of one supplier and one consumer of heat). When the pricing mechanism is verified it will be tested by applying it on one or two other cases. The testing will show which aspects of the designed mechanism can be applied on other applications and which cannot be applied. Eventually, for each case, suggestions for alterations of the designed pricing mechanism will be made. This leads to the case specific sub-questions:

o How can the designed pricing mechanism be verified for a specific case? o How to validate the designed pricing mechanism?

After verification and validation the factors of the mechanism that can be generally applied have to be identified. This leads to the general sub-question.

(9)

8

2. Methodology

This paragraph discusses the research design and gives a stepwise approach on how all the questions will be answered.

Step 1

Because there already has been a lot of research conducted on pricing mechanisms in the energy sector, the first part of the research will consist of a literature study. This study will describe how the pricing mechanisms currently in place work and will answer the first sub-question. Firstly the current pricing mechanism for residual heat will be described. Secondly the literature study will describe the current pricing mechanism for electricity and natural gas. These are the two main sources of energy currently sold to industrial applications. Therefore it is important to find out how these pricing mechanisms work and to identify all the factors that influence these mechanisms as some of the factors could also be implemented in a pricing mechanism for heat.

Step 2

In this step all the factors that influence the, to be designed, pricing mechanism will be identified. An initial set of factors will come forth of the previous step. In addition interviews will be conducted, among both suppliers and consumers of heat. Interviews are used because the factors, which influence the pricing mechanism, the Flexiheat partners find important have to be identified. Interviews are used because it is a good primary data collection technique for gathering qualitative data (factors can be seen as qualitative data as they do not consist of numbers) (Cooper & Schindler, 2006). The interviews will consist of open questions with room for follow up questions. The interviews will be conducted among employees of each company who have a lot of knowledge about the process of delivering and usage of residual heat and therefore have a good insight in which factors might influence the price of residual heat. This way of interviewing can also be seen as semi structured (Cooper & Schindler, 2006). This step will answer the second and third sub-question. The interviews will be conducted in three different cases, each consisting of at least one supplier and one consumer of heat. These cases are:

- The waste incinerator and a consumer located in Wijster, the Netherlands

- The power plant (supplier) located in the Eemshaven and a malt factory (consumer), located near Delfzijl, the Netherlands

- The heat network of Groningen Seaports, located near Delfzijl, , the Netherlands

(10)

9 Step 3

Only determining the factors that influence the pricing mechanism is not enough to actually build the mechanism because it is not known how much influence one factor actually has. To determine the amount of influence of a factor on the mechanism a survey will be held. The choice for a survey is based on the fact that the factors are already known and need to be weighted. A good method to obtain this data is to use a survey (Karlsson, 2009). The survey is of a descriptive nature as it is used to determine which factors will have the most influence on the mechanism (Karlsson, 2009). The survey will use a forced ranking scale. When this scale is used the persons who take the survey are forced to rank the different factors which give a good insight in the importance of each factor (Karlsson, 2009). The survey will be conducted among the same people who were interviewed in the previous step. These people are chosen because they provided the factors and have a good insight in their importance. After the survey has been conducted the analytic hierarchy process (AHP) will be applied to the results of the survey. AHP is designed to calculate weights to elements (BMPSG, 2013). The method uses the forced rankings as an input and some mathematical formulas to calculate the weight (importance) of each identified factor (BPMSG, 2013). This method is chosen because it is relatively easy and better than the standard ranking methods (BMPSG, 2013). This step answers the fourth sub-question as the result of this step are

weighted factors. These can be seen as the building blocks of which the pricing mechanism is constructed.

Step 4

Step 4 is the actual building of the pricing mechanism. The mechanism will be built in Excel and will be based on the factors identified in the previous step. The mechanism will be built for one setting

consisting of one supplier and one consumer and will then be tested on multiple other applications. This setting is chosen because it is the simplest situation and can therefore serve as a good basis for more complex situations. Step 5 & 6 will further elaborate on the testing.

Step 5

The first step in the testing process is to verify the model. Verification will be done by applying the developed model to the first case in step 2. Verifying the model is the first step in testing the developed theory (the model). The right method to test a theory is to use a case study (Karlsson, 2009). The output of the case study consists of prices (numbers) for residual heat and can therefore be seen as quantitative (Eisenhardt, 1989). The verification will be done by using supplied data by the companies in the case. The data consist of information of ingoing and outgoing energy (heat) levels. The outcomes of the case study will be communicated to the participants, who on their turn will deliver feedback. This feedback might lead to alterations of the initial mechanism. This step answers the fifth sub-question.

Step 6

(11)

10 will be done for the case the initial pricing mechanism is applied to and can also be seen as a sensitivity analysis. Practical validation will be done by conducting interviews with experts from two different fields (practice and academic). Interviews are used because it is a good way to acquire qualitative data (Cooper & Schindler, 2006). Different fields are chosen to make sure the factors and the designed pricing

mechanism are correct from both views. In the interviews each of the identified factors, the relationship between the factors and de designed pricing mechanism will be discussed. This way of conducting interviews can be seen as semi-structured as only the discussion topics are defined (Cooper & Schindler, 2006). Based on the outcomes of the interviews it will become clear which alterations to the mechanism could be made for certain applications (different users of residual heat).

Step 7

In the final step the outcomes of steps 5 and 6 will be analyzed. The aim is to identify the factors that influence the pricing mechanism that are not case specific. These factors are the general factors and can be applied directly in other cases. The identification of the general factors will be done by comparing the results of each case. Within this comparison all the factors that influence the price in each case will be put next to each other. The factors that are used in each case will be considered as general factors. This step answers the seventh and last sub-question.

(12)

11

(13)

12

3. Theoretical Framework

This section gives an overview of literature on residual heat and the pricing mechanisms currently in place. It will also answer the first two sub-questions. Section 3.1 explains the concept of residual heat. Section 3.2 discusses the different pricing mechanisms currently in place. Section 3.2.1 will explain the pricing mechanism of residual heat that is currently in place. Sections 3.2.2 and 3.2.3 will explain the pricing mechanisms that are currently in place for electricity and natural gas. Section 3.3 will be a conclusion discussing the factors found in the earlier sections.

3.1. Residual Heat

The concept of heat can be interpreted in a lot of ways. Therefore it is important to clearly describe how the concept of residual heat is defined in this research. Within industrial processes two kinds of heat exist, low and high grade heat (Ammar, Joyce, Norman, Wang, Roskilly, 2011). High grade heat is the heat which is viable for capture by the industrial processes itself (Ammar et al. 2011). Residual heat is often referred to as low grade heat and can be defined as heat that is not viable to recover within the processes and is rejected to the environment (Ammar et al., 2011; Etmoglu 2012). However, this definition is not complete enough as it does not make clear that residual heat still has economic value. The EPSRC (2011) report gives a better definition, it is stated that residual heat is energy which is rejected from a process at a temperature high enough above the ambient temperature to permit the economic recovery of some fraction of that energy for useful purposes. But heat alone does not have any value. It needs to be captured in a medium. This medium is called a working fluid and the most common working fluid is water or steam (Ammar et al. 2011; Etmoglu, 2012). However certain applications might require a different working fluid like ammonia or pentane (Ammar et al, 2011). This research will only focus on the use of water, in both the liquid phase and gas phase (steam), as a transfer fluid because all the companies involved use this medium (Waste incinerator, 2012).

There is one other important issue about heat that has to be discussed. This is the fact that, because of high energy losses, heat cannot be transported over great distances (Korhonen, 2001). The range in which heat can be transported is limited to 10 to 20 kilometers from the source, which makes it a very local product (Korhonen, 2001).

3.2. Current Pricing Mechanisms

This section will answer the first sub-question “What do the current pricing mechanisms in the energy

sector look like?” and the second sub-question “Which aspects of current pricing mechanisms in the energy sector can be applied?” In this section the pricing mechanisms that are currently in place for the

different kinds of energy delivered to industrial applications are described. It will give a clear overview of how they work and which factors influence the price of the different forms of energy.

3.2.1. Current Mechanism for Residual Heat

(14)

13 the fuel prices change. Within the current mechanism for the Flexiheat partners, situated in the

Netherlands, the price of heat is coupled to the natural gas price (Waste incinerator, 2012). The variable duty is based on the amount of heat used. The amount of heat used is calculated to MWh after which the costs of the amount of natural gas that would be necessary to supply the same amount of MWh are calculated. The final heat price is the natural gas price minus a discount percentage (Waste incinerator, 2012). In some cases in Finland the price is calculated in the same way, however the heat price is not coupled to the natural gas price but to the oil price (Suhonen, 2005).

From the current mechanism two factors can be identified, the fixed and the variable duty. This research does not focus on the way heat is delivered from the supplier to the customer therefore it can be assumed that this does not alter. This leads to the conclusion that the fixed duty will not change and is a factor that should be included in the pricing of residual heat. One of the problems of this research (as stated in paragraph 1.1) is that in the current mechanism the variable duty of heat is coupled to a competing energy source (Suhonen, 2005; Waste incinerator, 2012). Therefore the current variable duty will not be taken into account as price setting factor.

3.2.2. Electricity

The pricing mechanism behind electricity used to be simple and straightforward. The prices were set by regulators and reflected the cost of generation, transmission and distribution (Geman & Roncoroni, 2006). This all changed in the nineties when the privatization of the energy markets started. Most western countries, including the Netherlands, opened their electricity market for external parties (Keppo & Räsänen, 1999; Geman & Roncoroni, 2006). What followed was a serious restructuring of the market which greatly influenced the way in which electricity is priced. The initial price of electricity is still set by the cost made (Muyachi & El-Hawary, 1998; Anderson & Hu, 2005) but the eventual selling price is determined by the (international) supply and demand (Geman & Roncoroni, 2006). This implies that electricity has become a product that can be bought and sold by anyone who has an interest in it. This is a big contrast to the old situation in which electricity was produced and directly sold by the generator to a consumer. Because anybody can buy or sell electricity, a new market, that looks a lot like the stock exchange erected. This market is called the spot market (Green, 1999; Anderson & Hu, 2005). On the spot market, generators offer different quantities of power which can be bought by third parties. These third parties can be consumers, but can also be a middle man who tries to maximize ones profit

(Anderson & Hu, 2005). One consequence of the spot market is that the price of electricity for which it is sold can differ greatly from the production cost and there are even great differences between similar quantities sold (Green, 1999; Keppo & Räsänen, 1999; Anderson & Hu, 2005). In the Netherlands the spot market is exploited by the APX-group (Rijksoverheid, 2013a).

(15)

14 transmission, or in the case of heat, transportation costs are a factor that could be included in the pricing mechanism. The final factor is the generation costs. Residual heat is a byproduct of an industrial product and therefore does not have any production or generation costs (Ammar et al., 2011; Etmoglu, 2012) so this factor will not be taken into account.

3.2.3. Natural Gas

Worldwide there are various pricing mechanisms for natural gas in place. Almost every country has its own mechanism. These mechanisms greatly differ from each other (Mercados, 2010). There are a lot of countries in which the price of natural gas is coupled to the oil price. The price in these countries is calculated by a formula that is based on the price of light and heavy fuel oil (ECS, 2007; Mercados 2010). This method was also used in the Netherlands until 2004. In this year the Dutch government decided to decouple the natural gas price form the oil price to make the market more transparent (Bannisseht, 2008). This lead to a situation that is very similar to the electricity market. Supply and demand became an important factor to set the price (Bannisseht, 2008). The Dutch government set up a market where natural gas could be traded and accessed by third parties. This market looks a lot like the spot market and is called the Title Transfer Facility (TTF) (Rijksoverheid, 2013b). However the big difference with the spot market is that prices are not only dependent on supply and demand, but they are also influenced by factors like the physical capacity of the gas network, the exchange rate between the Euro and the Dollar and the price development of long-term oil contracts (Bannisseht, 2008).

The factors that can be identified are supply and demand, coupling to the oil price, price development of long-term oil contracts, exchange rate between the Euro and the Dollar and the physical transport capacity of the network. The first three factors can be disregarded as explained in paragraphs 3.2.1. and 3.2.2. Coupling the heat price to the Euro/Dollar exchange course can be seen as coupling the price to an index. An index is only a substitute for a competing energy source (Okkonen & Suhonen, 2010) and therefore this factor also will not be included. However, the final factor (the physical transport capacity) identified can directly be applied to a heat network and will therefore be included as one of the price setting factors.

3.3. Identified Factors from Literature

Most factors identified in the different pricing mechanisms cannot be applied in a pricing mechanism for residual heat. Eventually the theoretical framework yielded three useful factors, these factors are:

- Fixed duty: an annual fee to be paid by the consumer for its connection to the heat delivery network.

- Transportation costs, the costs made while transporting heat (mostly energy losses) - The physical transport capacity of the network

(16)

15

4. Interview outcomes

This section discusses the results that come forth of the conducted interviews. The interviews were conducted as discussed in step 2 of section 2. The interview outcomes can be found in Markus, 2013. The following people were interviewed:

I. Manager project development at waste incinerator, Wijster II. Regulatory affairs at waste incinerator, Wijster

III. Project manager water and energy at water company, Groningen IV. Manager QSHE at power plant, Delfzijl

V. Plant manager at malt company, Delfzijl VI. Project manager at Groningen Seaports

It has to be noted that the plant manager of the malt company (V) indicated at the start of the interview that he did not have enough knowledge to answer the questions, therefore this interview did not yielded any results and is not used. The interview protocol used can be found in appendix A. The questions posed in the interviews come forth of the identified factors from literature. The literature did not specify any specific factors and only gives a superficial overview of how the price of residual heat is set.

Therefore the first part of the interview is focused to find out how the price is set in the current situation and where improvements could be made. The second part of the interview focusses on all the separate factors that influence the price as this is very important for the actual design of the new pricing

mechanism. The last questions are used to verify the found factors from literature. The goal of the interviews was to identify the factors that influence the price of residual heat. This section answers the third sub-question: “Which factors influence the price of residual heat?”

4.1. Identified Factors

The factors that were identified from the interviews can be placed into three different categories. These categories are: fixed costs, variable costs and continuity. These categories come forth of the interview outcomes. In all the interviews a difference was made between fixed and variable costs. However, some factors identified could not be placed in one of these categories. The project manager of the water company (III) stated that these factors are continuity factors as all these factors influence the continuity in which residual heat is delivered. Therefore it was chosen to create the third category called continuity. Each of the factors will be discussed below. The numbers behind each factor indicate which person mentioned the factor.

Fixed costs

(17)

16 which the investments are depreciated. If for example the contract length is 10 years, the

investments will be depreciated in 10 years.

- Maintenance costs (II), Regular maintenance to prevent breakdowns is necessary. The costs made for maintenance are (partly) charged to the consumer.

- Staff costs (II), to make sure heat is actually delivered to the consumer and the network is operated in the right way it is necessary to hire the right staff. These costs are also charged to the consumer.

- A yearly fixed duty (II, IV), every year a fixed duty has to be paid by the consumer for its connection to the heat network. This duty has to be paid in extend to all the costs made. Variable costs

- Decoupling costs (III, IV, VI), to make the residual heat available the heat has to be transferred to a medium (water in the case of this research) that is suited for transportation. These costs include the costs to be made to get the medium to the temperature and pressure as demanded by the consumer or the costs of electricity that otherwise could have been generated.

- The amount of energy used by the consumer (I, II, III, VI), the amount of energy is influenced by factors like temperature and pressure. How these factors influence the amount of energy require some calculations that lie in the field of thermodynamics. This field is beyond the scope of this research therefore only the amount of energy will be used.

- The amount of energy returned (II, III), in most cases heat is delivered in a closed system and will be delivered back to the consumer after is has been used. This could lead to a discount for consumer. On the other hand, when no heat is returned to the supplier, the supplier will be confronted with extra (energy and water) costs. This could lead to a penalty fee to be paid by the consumer.

- Transportation costs (I, III, VI), these costs consist of:

o Heat losses (I, III, VI), during transportation heat losses occur, a supplier could decide to charge the costs of these losses to the consumer. Heat losses are influenced by the volume, temperature and pressure of the medium that is transported. However, the calculations behind this are in the field of thermodynamics and are beyond the scope of this research.

o Pumping costs (III, VI), in some cases (dependent on the phase and pressure of the medium) it is necessary to pump the medium from the supplier to the consumer. These pumps require energy and thus create extra costs.

- Water costs (II, IV, VI), this factor is made up of the following factors:

o Water used (II, IV, VI), the amount of water used to get the heat from the supplier to the consumer.

o Costs of creating demineralized water (II, IV, VI), normal water cannot be used as it would cause serious erosion in the pipelines. Therefore the water has to be

demineralized before it can be used.

(18)

17 supplier because when the residual heat is used; less energy is not used, which also leads to a reduction in CO2 emissions. However, the CO2 emission rights still have to be paid although they might be lower than when traditional energy sources are used.

Continuity

- Demand pattern of the consumer (I, II, III, IV, VI), does the consumer needs heat 24/7 or only on working days. The pattern in which heat is required could greatly influence its price. When a supplier has a non-stop process and has a constant production of heat but the consumer only needs that same heat on specific moments this might cause problems for the supplier as it cannot get rid of its residual heat. This could lead to the situation in which the residual heat has to be dumped, which on its turn causes higher environmental costs.

- Availability of residual heat (I, II, III, IV, VI), this depends on the process of the supplier, does the supplier has a constant process or is the process shut down in the weekends. A consumer might have a constant need for heat while the supplier can only deliver heat on working days. This factor is similar to the demand pattern but is only viewed from the consumers’ perspective. - Reliability of the supplier (I, II, III, IV, VI), this factor implies the unscheduled downtime of the

supplier. These are mostly caused by breakdowns. The reliability can be seen as a percentage of the amount of time agreed between the supplier and consumer, in which the supplier actually delivers heat.

- Contract length (I, III), the length of a contract influences the depreciation costs of the

investments. The yearly depreciation costs are calculated by dividing the total investment costs by the length of the contract. The contract length may also influence the energy price set by the supplier. If, for example, a supplier wants to close a 10 year contract but the consumer only wants to sign for 5 years it might cause a higher energy price. This is because the income of the supplier becomes less certain so the supplier wants to earn more money in the shorter period to compensate for the uncertainty.

The interviews also yielded some factors that cannot be applied in the design of the pricing mechanism. - The physical transport capacity of the network is found not to be a factor as heat networks are

built in such a way that they always have more capacity than the maximum demand (II, VI). - Connection fee (II), it is common that consumers pay a one-time fee to get connected to a heat

network. Although this fee will probably have to be paid by the consumer it does not influence the price of 1 MWh or GJ of heat. It is a fee that has to be paid to gain access to the heat network.

(19)

18

4.2. Conceptual Model

To give a clear overview of how each factor influences the price of residual heat a conceptual model can be constructed. However, due to the high number of factors a conceptual model for each category is created. The final model will give an overview of how the categories are related to each other and how they influence the price of residual heat. In the models the + means that a factor increases the price of heat, the – implies that the factor decreases the price of heat. The conceptual models are shown in figure 2-5. The first model shows the factors influencing the fixed costs.

Figure 2: Conceptual model fixed costs

The second model shows the factors influencing the variable costs.

(20)

19 The third model gives an overview of the factors influencing the continuity. What is different about this model is that each factor has a + or a -. This is explained by the fact that each continuity factor could increase or decrease the price of one MWh. The reason behind this is, as explained before, that certain agreements between the supplier and consumer are made. If for example the consumer does not meet these agreements the supplier could charge a higher energy price or give a fine to the consumer.

Figure 4: Conceptual model continuity

The final model gives a clear overview of the relations between the different categories and how each category influences the price of residual heat.

(21)

20

4.3. Results from Interview Outcomes

The interviews yielded a lot of factors that influence the price of residual heat. All these factors, with their explanations, can be found in section 4.1. The interviews also yielded some factors that cannot be applied in het pricing mechanism. The reasons of why they are not applicable can be found at the end of section 4.1. The interviews rejected (two times) the factor “physical transport capacity” which was found in the literature therefore this factor will not be included in the pricing mechanism.

All the identified factors are summarized in table 1. The table also shows if a factor comes forth from literature or the interviews.

Table 1: Overview of identified factors

Factor Literature Interviews

Fixed costs

Depreciation costs X

Maintenance costs X

Staff costs X

Yearly fixed duty X X

Physical transport capacity X

Variable costs

Decoupling costs X

Amount of energy used X

Transport costs X X

Amount of water used X

Environmental costs X

Amount of energy returned X

Continuity

Demand pattern X

Availability X

Reliability X

Contract length X

From the table it can clearly concluded that, within the scientific literature, there is not much known about the factors that influence the price of heat. Because these factors are not known there also is not a good way to price residual heat to be found in the scientific literature.

After all the factors were identified several conceptual models were built to show the relations between the different factors and how each factor eventually influences the price of residual heat. These models can be found in section 4.2.

(22)

(23)

22

5. Design of the Pricing Mechanism

This section contains the actual design of the pricing mechanism. The design is based on the factors identified in sections 3 and 4. Before the pricing mechanism can be build the variables have to be

defined. This is done in section 5.1. Section 5.2 will give the actual pricing mechanism. Finally this section will answer the fourth sub-question: “How can a pricing mechanism be designed for a specific setting?”

5.1. Notations

The table below gives an overview and description of the used symbols.

(24)

23

5.2. The Pricing Mechanism

This section describes the actual pricing mechanism. It will show how each category is built from the different factors before calculating the total costs of residual heat per MWh.

5.2.1. Fixed Costs

The fixed costs are calculated by dividing the sum of the fixed cost factors (depreciation-, maintenance- and staff costs) by the total amount of energy used as can be seen in (1). The depreciation costs have to be calculated at forehand and this is done by dividing the investment costs by the contract length (2). The investment costs are calculated by subtracting the investment costs paid by the consumer in advance from the total investment costs as can be seen in (3).

To calculate the total fixed costs per MWh:

The depreciation costs per year are calculated by:

In which the investment costs are calculated by:

With the following condition: α ≤ 1

5.2.2. Variable Costs

The variable costs are a summation of all the identified variable factors (decoupling-, transport-, water-, environment costs) minus the total costs of energy returned. This number is than divided by the total amount of energy used (4). The transport costs are the sum of the pumping costs and the energy losses (5), in which the energy losses are calculated by the total amount of energy lost multiplied by the price of one MWh as can be seen in (6). In the same way the price of the energy returned is calculated (7). The water costs are calculated by the summation of the price of one m3 of water and the costs to

demineralize one m3_{of water. This number is multiplied by the total amount of cubic meters of water} that are necessary (8).

To calculate the total variable costs per MWh

The transport costs are calculated by:

(25)

24 In which costs of the energy losses are calculated by:

The total price of the energy returned is calculated by:

The costs of the water used are calculated by:

5.2.3. Continuity

The factors in the continuity category can be modeled in two different ways. The first way is in the form of a multiplying factor. For example, if reliability is only 80% a discount of 5% over the (variable + fixed) costs is given to the consumer (9). The second way is in the form of giving an actual financial discount. So in the same example, not a 5% discount is given but a discount of €10.000,- (10). This leads to two formulas for the continuity. It has to be noted that the contract length is not modeled as an amount of years. It will be modeled in the same way as the other factors. The value given to the contract length in this situation depends on the contract demand of the supplier. This is because a supplier might charge a higher price to a consumer who only signs a contract for five instead of ten years. The reason behind this is the loss of future income by the supplier.

The continuity can be calculated with the following formulas:

The following conditions apply:

If Av1 has a value Av2 = 0 & if Av2 has a value Av1 = 1 If CL1 has a value CL2 = 0 & if CL2 has a value CL1 = 1 If DP1 has a value DP2 = 0 & if DP2 has a value DP1 = 1 If Re1 has a value Re2 = 0 & if Re2 has a value Re1 = 1 5.2.4. Total Costs

Now all the calculations for the different categories are known, the final costs can be calculated. The total costs are calculated by the sum of the fixed and variable costs which is multiplied by Cont1. Hereafter Cont2 is added and the profit margin can be added. Finally the fixed duty price per MWh is added (11).

The formula for the total costs is:

(( ) )

(26)

25

5.3. Conclusions on the pricing mechanism

The designed pricing mechanism is fairly simple and straightforward. The basis of the price is set by the fixed and variable costs (cost based). These costs are made up of all the factors that influence the price of residual heat. The simplicity of the mechanism is therewith the strength of the mechanism as it can easily be applied in different cases by just filling in all the required variables. The mechanism might be interpreted as a very static model; however the continuity factors do not have to change only on a yearly basis. They might be set every week, day or even every hour. This makes the mechanism dynamic as it can easily be adjusted for when supply and demand (availability and demand pattern) differentiate over time. When the mechanism is dynamically applied it shows similarities with the found mechanisms for natural gas and electricity as they are both based on supply and demand. The same can be said for the dynamically applied mechanism.

Another strength of the mechanism is that it is designed in such a way that a factor can easily be left out of the equation if it does not apply to a certain setting. In extend, a factor that is not yet included in the mechanism can easily be added, although the factor has to be categorized first (fixed or variable costs, or a continuity factor).

(27)

26

6. Test results

This section will discuss the results of the experiments conducted with the designed pricing mechanism. Testing was done in two steps. The first step is verifying the model as described in step 5 of section 2. The second step is validating the mechanism, which is in accordance with step 6 of section 2.

6.1. Verification

This section will answer the fifth sub-question: “How can the designed pricing mechanism be verified for

a specific case?” Verification is done by applying the data, obtained via the Flexiheat project, to the

pricing mechanism. The case used is the setting with a waste incinerator and its consumer located in Wijster. Unfortunately the data was not complete enough to fill in all the variables. Therefore the following assumptions are made:

- Staff costs are set on €100.000,- per year. This is based on the assumption that the total overhead costs for personal are equal to two full time employees. The staff costs have a small impact on the eventual price as they are low compared to other costs made (such as investment costs and decoupling costs).

- Environmental costs are set on € 0,- It is not known if the waste incinerator exceeds its emission rights therefore it is assumed they do not. The influence of the environmental costs is low as one kiloton of CO2 costs around €5000 (EEX, 2014) and the emissions of the waste incinerator are 495 kiloton per year (NFM Drenthe, 2012).

- The continuity factors are set to their default values (1 and 0) because it is not known if the agreements made are met. The influence of the continuity factors will be tested in section 6.2, validation.

With these assumptions made the model can be filled in. Table 3 gives an overview of the used input variables. These variables come forth of the case situated in Wijster. The input was given by the participants.

Table 3: Used input

Total costs Fixed costs Variable costs Continuity (1/2)

Variable Value Variable Value Variable Value Variable Value

β 1 α 0% Cdec €688.459,24 Av 1/0

C(tot)inv €0 Cdem €1/m3 CL 1/0

(28)

27 Within this case the depreciation costs, the heat losses and the maintenance costs are all included in the fixed duty and therefore have a value of 0. The decoupling costs are made up of the costs made for unsold electricity. The price for the heat returned is 0 as the consumer is obliged to do so (Waste incinerator, 2012). Finally an overview of the energy streams can be found in appendix B.

The given input gives the following calculations. Fixed costs:

The first step is to calculate the fixed costs per MWh. This gives the following calculations: The depreciations costs per year are:

The fixed costs per MWh are:

Variable costs:

The second step is to calculate the variable costs per MWh. This gives the following calculations The costs of the heat losses are:

The costs of the water used are:

The costs of the energy returned to the supplier are:

The variable costs per MWh are:

(29)

28 Continuity:

The third step is to calculate the continuity factors. This gives the following calculations:

Total costs:

Finally the total costs can be calculated using formula (11).

( ) ⁄

Thus when the pricing mechanism was applied a price of €25,02/MWh is obtained. This is only the cost price and it can be assumed that the supplier wants to make some profit on the delivered energy. Therefore the price for a profit margin of 10%, 15% and 20% are calculated and shown in table 4.

Table 4: Price with profit margins

Profit margin Price per MWh

10% € 26,83

15% € 27,74

20% € 28,65

The average natural gas price in 2013 was €26,99/MWh (APX, 2014). With the addition of the fixed duty the price in the current situation is €33,87/MWh. When comparing this price to the prices with a profit margin it can be concluded that the prices with a profit margin are still lower the price in the current situation. With the designed pricing mechanism the use of residual heat would still be a viable option for the consumer.

In the current situation the price for one MWh is calculated by giving a discount percentage on the natural gas price not including the fixed duty (Waste incinerator, 2012). Table 5 gives an overview of different discount percentages given in the situation in which the price is coupled to the natural gas price.

Table 5: Price with discount when coupled to natural gas

Discount percentage Price per MWh

10% € 31,18

15% _{€ 29,83}

20% € 28,48

(30)

29 The only situation in which the price is higher is when a 20% discount on the natural gas price given and a 20% profit margin is applied. However, the natural gas price fluctuates and so will the price of one MWh of heat when the current mechanism is applied. The designed mechanism gives a price that is more stable because it is not influence by the natural gas price.

In the current situation it is hard to implement the pricing mechanism due to the fact that a contract between the supplier and consumer is already in place and it is very unlikely that it is broken up. Still, when trying to attract new consumers the designed mechanism could be directly applied to the new case. This implies a lower relative profit for the supplier but when it leads to more consumers the absolute profit will be higher.

6.2. Validation

The next step in the research is validating the designed pricing mechanism. Step 6 of section 2 describes two kinds of validation, numerical and practical. The numerical validation is done by altering various input variables in the current setting. Practical validation was done by interviewing experts in both the practical and academic field. This section will answer the sixth sub-question: “How to validate the

designed pricing mechanism?”

6.2.1. Numerical validation

For the numerical validation the following experiments were conducted:

- A comparison between the prices obtained from the mechanism and the situation in which the price is coupled to an alternative fuel.

- The influence of the continuity factors on the price per MWh and the difference in the price when they are calculated as a multiplying factor or an actual financial discount/fine.

- The influence on the price of the energy consumed by the consumer. - The influence of the grade of the returned energy.

All the experiments conducted used the same setting as was used in section 6.1 with the exception of the tested variables.

Experiment 1

In the first experiment a comparison is made between the price of one MWh of heat when it is coupled to an alternative fuel and when the price is obtained from the designed mechanism. This experiment is done to create a link with real life settings. The chosen fuels are natural gas and oil because it is common to couple the heat price to these fuels (Suhonen, 2005). In both comparisons an inflation rate of 2% per year was applied to the prices obtained from the designed mechanism.

(31)

30

Figure 6: Comparison mechanism to natural gas price

It can be concluded that the natural gas price highly fluctuates. This implies that when the price of heat is coupled to the natural gas price this price would also greatly fluctuate. However, the average gas price within these years is €20,47. This means that when a discount is applied the average price is always lower than the price obtained from the mechanism. This is a great advantage for the consumer. The problem with this low average is that it is caused by low prices in 2007 and 2009. The trend since 2009 is that the price of natural gas is rising and therefore one might say that it will result in a higher price of heat in the future. Next to that, as mentioned before, the pricing mechanism provides a price that is much more stable.

(32)

31

Figure 7: Comparison mechanism to oil price

It can be concluded that a one on one coupling to oil was very beneficial for the consumer in the past. Up till 2005 this would give a very low price of heat per MWh. However after 2003 the price of oil sky rocketed and became instable over the years. This implies a great rise and instability in the price of heat. Experiment 2

In the second experiment the influence of the continuity factors is tested. In the first part the continuity factors were modeled as a multiplication factor. In the second part they were modeled as a fine or discount to be paid or received by the consumer. In both settings the cost price of one MWh was used (so no profit margin was applied).

(33)

32

Figure 8: Continuity modeled as a factor

(34)

33 Because the last two settings did not yield any satisfying results a third setting was tested. In this setting a fine of €100.000 was modeled and a discount of €100.000 was modeled. This led to new cost prices. In extend to this after the fine and discount were modeled there was also a continuity factor modeled. A scenario where this could happen is when a consumer only signs a contract for 5 years while the supplier expected 10 years. A supplier could than charge extra costs for lost income in the future. Of course there are still agreements on how much energy there should be delivered thus certain multiplication factors do still apply. The results of this scenario are shown in figure 10.

Figure 10: Combination of fine/discount and a factor

Because of the different cost prices it was expected that the difference between the prices would increase when a factor was applied. As one can clearly see in figure 10 the lines do not have the same gradient. This implies that the price difference grows when the multiplication factor grows. To prove this the difference between the prices per MWh was calculated. The results are shown in figure 11.

(35)

34 Figure 11 clearly shows a rise in price differences. This is caused by the different standard prices set. One might argue that when a contract of only 5 years is signed instead of 10 the depreciation costs double. This is also true and will have the same effect as the scenario sketched because the standard price would increase.

It can be concluded from this experiment that there is no difference when applying the continuity factors as a multiplication factor or as a discount or fine. When these factors are combined a price difference does occur.

Experiment 3

In this experiment the influence of the amount of energy consumed is tested. In this setting the amount of consumed energy will be varied and again the standard price is calculated. The results are given in figures 12.

Figure 12: Influence of energy consumption

(36)

35

Figure 13: Consumed energy with cost factors

The results in figure 13 are more up to the expectations. There is still a steep decline in costs when the amount of energy consumed is low. This was also expected and can be explained by the fact that investment costs of a heat network are high.

(37)

36 Experiment 4

In the final experiment the influence of the grade of the returned energy was investigated. The grade of the returned energy can be explained as the temperature of the return stream. If the temperature goes up, so will the grade of the heat (making it more high grade). To model this, the price of one MWh of returned energy was adjusted. The thought behind this is that if the temperature is higher the supplier has fewer expenses, thus the returned energy has a higher value. The results are shown in figure 14.

Figure 14: Influence of the grade of the returned energy

The figure clearly shows that when the grade of the returned heat is higher the price goes down. This result was expected and it makes perfect sense because of the way the amount of returned energy is modeled. The model is built as such that an increase of the price of returned energy of €1 will decrease the price per MWh with a fixed factor. This explains the linear decrease of the price.

(38)

37

Figure 15: Influence of different grades of returned energy on different consumption levels

Figure 15 shows the expected results. The price difference between the different grades of returned heat is constant. This is because the amount of energy returned is a fixed factor compared to the energy consumed. The actual price that comes forth of the experiment becomes lower when the grade of the returned heat increases. This is again in line with the expectations as the grade of the residual heat has a negative influence on the final price per MWh.

6.2.2. Practical validation

The designed mechanism and identified factors were practically validated by interviewing experts in two fields, the academic field and an expert from practice. The experts consulted are:

- H. Oost, Production manager at the waste incinerator located in Delfzijl - Dr. Ir. W.H.M. Alsem at the Rijksuniversiteit Groningen

The case of Delfzijl gives a different view on the mechanism because the waste incinerator is a supplier for multiple consumers. So it is different from the case in Wijster which consists of only one supplier and one consumer. The interview with Dr. Alsem gives an academic view on the designed mechanism. In extend to this Dr. Alsem has extensive experience in the energy business. Both experts were not interviewed in previous research steps.

(39)

38 In extend to this H. Oost stated that the designed mechanism would be a good alternative for the

mechanism currently in place.

Concluding, both interviews confirmed both the correctness of the identified factors and the designed pricing mechanism. Based on these interviews no adjustments to the mechanism have to be made for a case that has only one supplier. Unfortunately no case with multiple suppliers was used, therefore it is not known if the factors and mechanism are also correct for such a setting.

Finally it has to be noted that it was also attempted to acquire data from different business cases and apply it to the designed mechanism. However it turned out that the companies approached were very reluctant to supply the data needed.

6.3. Conclusions after testing

Verification of the designed pricing mechanism yielded a price that lies close to the price that comes forth of the mechanism that is currently in place. From this it can be concluded that the designed pricing mechanism yields realistic results and could therefore be implemented in the used case. Besides this the obtained prices from the design are lower than the prices in the current situation. This might be a disadvantage for the supplier, but it could lead to more consumers which would lead to an increased profit.

A second conclusion that comes forth of the verification is that some factors have more influence on the price than others. This is due to the fact that some factors represent a bigger amount of money than other factors do. So an alteration within these big factors can result in a substantial alteration in the eventual price.

The validation yielded that the designed mechanism gives a price that is more stable and lower than when the price is coupled to an alternative energy source. The price is more stable because of the cost based approached. Various costs are fixed en therefore do not fluctuate over time.

Secondly the way in which the continuity factors are modeled (a factor or a discount/fee) is not very interesting as both ways yield the exact same results. However, when these ways are combined the price of residual heat compared in different situations does differ. Still the designed mechanism is sensitive to the continuity factors which make it dynamically applicable if the supply and demand (availability and demand pattern) fluctuate over time.

Variations in the price are bigger when the amount of used energy is low. When the amount of used energy goes up the fluctuations in the price go down. So it can be concluded that the pricing mechanism cannot be implemented for consumers that require low amounts of energy, however the mechanism is very suited for implementation when the consumed energy levels are high.

(40)

39

7. Generalization

In this section discusses the general applicability of the pricing mechanism. Generalization is done by comparing the results from verification and validation conducted in section 6. By conducting the generalization the last sub-question “How to test the general applicability of the pricing mechanism?” will be answered.

The factors that were identified and applied in the mechanism were confirmed in the expert interviews. No factors were left out of the pricing mechanism and no factors have to be added. The numerical validation yielded that the mechanism gives a more stable price when the amount of consumed energy is high. Finally, the verification and validation was limited to a situation in which there is only one supplier in the network.

(41)

40

8. Discussion

The goal of this research is to design a pricing mechanism for residual heat that is independent of alternative energy sources.

The literature review and the conducted interviews yield all the factors that have to be implemented in the mechanism. A total of 14 factors are found, but some of these factors are made up of different variables. This results in a total of 27 variables that are eventually implemented in the pricing mechanism.

All the identified factors are straightforward and can easily be given an economic value; therefore it was not necessary to weigh the factors. Because of the straightforwardness of the factors the designed mechanism is fairly simple and easy to apply. Due to the simplicity the results that come forth of the mechanism do not leave much room for a different interpretation than what they are, a price per MWh. Unfortunately, due to the decoupling costs, the mechanism is not totally independent from alternative energy sources because these costs consist of upgrading the heat and/or the costs of unsold electricity. However these costs are only a fraction of the total price and it can therefore be stated that only a light coupling remains.

The cost price obtained from the designed mechanism is €25,05/MWh. A coupling to the natural gas price, which is the current situation, gives a price of €33,87/MWh. On first sight the price obtained from the mechanism is significantly lower, however a discount is given on the natural gas price and a profit margin has to be added to the price of the designed mechanism. Only in the situation of a 20% and a 20% profit margin the current system yields a lower price.

Although the continuity factors were modelled both as a factor and a fine/discount the results were exactly the same. This is because a fine/discount can also be seen as a factor also represents a certain amount of money and vice versa.

The mechanism turns out to be sensitive to the continuity factors which makes the pricing mechanism dynamically applicable as the supply and demand are two of the continuity factors. In extend to this the mechanism is sensitive to the amount of heat consumed. When the designed mechanism is applied to the used case it turns out that the use of residual heat is already viable when only 30.000 MWh/year are demanded.

Finally, the designed pricing mechanism can be applied to cases that consist of one supplier and one or multiple consumers.

The designed mechanism is a cost based approach to establish the price of residual heat. The price that comes forth of the mechanism is lower and more stable than the current situation. Therefore

(42)

41

9. Conclusion, recommendations & further research

Section 9.1 reflects on the main objective and research questions. Section 9.2 contains the recommendations and section 9.3 discusses the limitations and suggestions for further research.

9.1. Conclusion

The main objective of this research was to design an independent pricing mechanism which establishes the price of residual heat.

The mechanisms that are currently in place for alternative energy sources cannot be applied because they depend on the property that these sources can be transported over great distances. Unfortunately, heat is a very local product and when transported over great distances the energy losses are very high. Because the current mechanisms cannot be applied a new mechanism had to be designed. The bulk of the factors, which are the building blocks of the mechanism, were identified from real life situations. Within the scientific literature little is known about how residual heat can be priced and which factors influence its price. The identified factors are represented in the table below.

Table 6: Identified & implemented factors

Fixed costs Variable costs Continuity factors

Depreciation costs Decoupling costs Demand pattern

Maintenance costs Amount of energy used Availability

Staff costs Transport costs Reliability

Yearly fixed duty Amount of water used Contract length Environmental costs

Amount of energy returned

These factors result in a fairly simple, cost based mechanism. Applying the mechanism to a real-life case gives a realistic price that is lower and more stable when compared to the current situation. The

continuity factors allow that the mechanism is not just a static model but that it can be dynamically applied.

The mechanism gives a different and new perspective on how to price residual heat and can be applied to situations which consist of one supplier and one or more consumers. For both parties it is important to realize that investment costs for building a heat network are high and is only a viable option when the consumed energy levels are high to ensure a price that is lower than the natural gas price.

So it can be concluded that the price of residual heat can be established by using a cost based approach. Due to this approach the price of residual heat becomes more stable and is lower than the price

(43)

42

9.2. Recommendations

 One general recommendation is to decouple the price of residual heat from alternative energy source. Because a lot of suppliers do not use any of these energy sources in their process to produce the heat it yields a price that cannot be traced back to their original process. Because of this suppliers become depend of factors beyond their scope of control.

 Prices of alternative energy sources show great fluctuations. Therefore it is recommended to implement the designed pricing mechanism within already existing and future projects. It will result in a more stable price.

 Before the pricing mechanism can be implemented it is important to carefully map all the involved costs. When this is not done the price could greatly differ from the predicted price and might even exceed the natural gas price.

 The continuity factors can have great influence on the price. Therefore it should be clear for both parties (supplier and consumer) what is expected from each other. If for example the demand is higher than assumed by the supplier it can result in a price that is much higher than expected by the consumer.

9.3. Limitations & further research

 The research was limited to a situation in which only one supplier is present in the heat network. Therefore a suggestion for further research would be to investigate if the designed mechanism is applicable in a situation with multiple suppliers and if the designed mechanism has to be altered for this situation.

 Although within the validation it is confirmed that the designed mechanism can be applied to a situation in which the supplier delivers heat to multiple consumers, the mechanism was not applied to such a case. So it is not known what the effect on the price will be. Further research, in which the mechanism is applied to a case with multiple consumers, could show the effects on the price of heat in such a setting and if price differences occur between the different

consumers.

 Prices in this research are calculated on a yearly basis, however, the demand pattern and the availability might fluctuate throughout the year. A suggestion is to investigate how the designed mechanism can dynamically be applied in which prices are set on a more short-term basis (e.g., weekly, daily or hourly).

 Another limitation is that the research only focused on the connection between supplier and consumer. However a consumer does not always use all the heat supplied (heat is often

(44)

43

References

Ammar, Y. Joyce, S. Norman, R. Wang, Y. Roskilly, A.P., (2011), Low grade thermal energy sources and

uses from the process industry in the UK, Applied Energy: 89 (2012) pp. 3-20

Anderson, E.J. Hu, X., (2005), Supply Function Equilibrium in Electricity Spot Markets with Contracts and

Price Caps. Journal of optimization theory and applications: Vol. 124, (2005), No. 2, pp. 257–283

Bannisseht, Q. von (2008), Niet olie maar TTF bepaalt gasprijs, Vakblad voor de Bloemisterij, Nr. 41, pp. 14-16.

Cooper, D. R., & Schindler, P. S. (2006). Business Research Methods (ninth edition). New York: McGraw-Hill/Irwin.

Dincer, I. (1999), Renewable energy and sustainable development: a crucial review, Renewable and Sustainable Energy Reviews: 4 (2000) pp. 157-175

ECS (Energy Charter Secretariat), (2007), Putting a Price on Energy: International Pricing Mechanisms for Oil and Gas, Brussels: Dépot Légal

Eisenhardt, K.M., (1989), Building Theories from Case Study Research, The Academy of Management

Review: Vol. 14, No. 4 (Oct., 1989), pp. 532-550

EPSRC: Thermal Management of Industrial Processes, (2011), National sources of low grade heat

available from the process industry. Newcastle: Newcastle University

Etemoglu, A.B., (2012), Thermodynamic investigation of low-temperature industrial waste-heat recovery

in combined heat and power generation systems, International Communications in Heat and Mass

Transfer: Volume 42, March 2013, pp. 82-88

Geman, H. Roncoroni, A. (2006), Understanding the Fine Structure of Electricity Prices, The Journal of Business, Vol. 79, No. 3 (May 2006), pp. 1225-1261

Green, R, (1999), The Electricity Contract Market In England And Wales, The Journal of Industrial Economics, Volume XLVII, No 1, pp. 107-124

Karlsson, C. (2009), Researching Operations Methods, New York & London: Routledge

Keppo, J. Räsänen, M. (1999), Pricing of electricity tariffs in competitive markets, Energy Economics: 21 (1999) pp. 213-223

Korhonen, J. (2001), Co-production of heat and power: an anchor tenant of a regional industrial

ecosystem, Journal of Cleaner Production 9 (2001) 509–517

Markus, S. (2013), Residual Heat, Designing and Applying a Pricing Mechanism for Industrial Applications:

Designing And Applying A Pricing Mechanism For Industrial Applications Residual Heat:

Residual Heat: