Approaches towards maximizing network utilization

(1)

Approaches towards maximizing network utilization

How to deal with unused capacities at Gas Transport Services?

Gregor Schneider

Groningen, July 3, 2009

Master thesis Technology Management Faculty of Economics and Business, University of Groningen

Supervisor: Drs. C. de Snoo

(2)

Abstract

(3)

Acknowledgements

At this point, I want to thank everybody who has directly or indirectly contributed to the achievement of this thesis. This research project has been the last part of my studies and at the same time the first step into my professional career. I really enjoyed the topic and hope being able to work in the same field later on.

First of all, I have to acknowledge Bert Kiewiet, who offered me the opportunity to do my final year project at the Gasunie. He did not only give me the chance to write my thesis on a challenging topic, but also to learn about the Dutch working culture and the fascinating world of gas transport, I hardly knew anything about in advance. I appreciated his help and patience during the practical part of this project and especially during the writing of my thesis.

In the same way I want to thank Cees de Snoo who has been my supervisor at the Rijksuniversiteit Groningen. His enthusiasm and critcial suggestions supported me a lot in developing this research. He helped me in particular to align the practical and academic requirements of this thesis

(4)

1 Introduction

1.1 Outline

This thesis completes my Master of Science program in Technology Management at the Rijksuniversiteit Groningen. It has been written as a final year project within a company. Topic of this diagnosis research is the capacity planning of the Dutch gas infrastructure network at N.V. Gasunie. Since the liberalization of the gas market, the challenges that capacity planning is facing have changed considerably. Focus of this diagnosis will be, to analyze how the utilization rate of the Dutch gas infrastructure has changed and can further be improved for the future. The main question is how capacity planning can further be enhanced, to assess the availability of underutilized capacities. Following sections will give an overview of the company, describe the problem statement and elaborate on the research design.

1.2 Gasunie

N.V. Nederlandse Gasunie is a Dutch company which is specialized in the transportation of natural gas. Gasunie was founded in 1963 and has its headquarters in Groningen.

It is a gas infrastructure company having a high pressure grid consisting of over 15,000 kilometers of pipelines. The Gasunie network is one of the biggest in Europe and extends over the Netherlands and northern Germany. The throughput reaches 125 billion cubic meters per year.

The company’s mission is to deliver gas transport and related services safely and reliable to an integrating European gas market. Furthermore, it pursues the strategy of becoming the “gas roundabout” of northwest Europe, by adapting to the long-term transport demand of Dutch and foreign gas flows, offering storage capacities, participating in LNG terminal projects and international pipeline projects, such as the Nordstream.

(6)

Gas Transport Services B.V. (GTS) is the network operator in The Netherlands. It is responsible for providing gas transport services, planning and expanding the Dutch network. GTS performs its tasks fully autonomously from N.V. Nederlandse Gasunie, as it is required by Dutch law.

The Construction & Maintenance division is responsible for building and ensuring a safe, reliable and sustainable gas transport system.

Gasunie Participations & Business Development is responsible for the development of non-regulated services and products. It is also responsible for the participation in national and international gas infrastructure projects, such as the BBL that links the Dutch network to the United Kingdom (Balgzand-Bacton-Line).

The Gasunie Engineering & Technology division (GET) is specialized in research and engineering services for gas transport and energy efficiency. Among other things it is doing research and development on energy, infrastructure and sustainability issues.

Gasunie Deutschland GmbH & Co. KG is the operator of the North German pipeline network. Its responsibilities include operating, managing and developing gas transport and offering related services in Germany.

1.3 Gas Transport Services

The work for this thesis was executed within Gas Transport Services B.V. (GTS). As national transmission system operator (TSO) for natural gas, it is responsible for the management, operation and development of the gas transmission grid in the Netherlands. GTS’ mission is the independent provision and implementation of gas transmission services, for the benefit of a properly functioning liberalized gas market in Europe. Safety, reliability and security of supply are its main priority.

(7)

GTS also has public tasks related to the security of supply in extreme cold weather conditions and related to the small-fields policy (the extraction of natural gas from small fields). To this end, the company liaises with the Minister of Economic Affairs regarding the estimates for the next twenty years. GTS is bound by the Gas Act. In this Act it is stated that the national transmission operator has to treat its customers all equally in equal circumstances.

GTS anticipates market developments, provides new services and invests in the provision of pipelines, compression capacity and the storage of natural gas through capacity planning. The primary objective of the capacity planning process is to offer sufficient transport capacities to GTS its customers and to ensure a balanced transport network that matches the supply and demand of natural gas. It is a crucial activity within GTS to manage, operate and develop the gas transmission grid. In order to plan GTS studies problems such as fluctuations in supply and demand or temperature, affecting the main transport network and assesses appropriate solutions. The planning has to achieve optimal transmission plans, to assure the allocation of transport capacities, and drafts the design of large network expansions to match future needs in transport capacities. Therefore capacity planning needs not only detailed information about the network configuration, but also about the supply and demand of natural gas that has to be transported through the Netherlands. Planning is a vital activity of network management and will be the focus of this research.

The system for selling GTS´ transport services is known as the entry-exit system. The pipeline network has about 50 so called ´entry´ and 1,100 so called ´exit´ points. At an entry point the gas can be physically injected in the system. At an exit point the gas can be physically removed from the system (for example, a gas delivery station or an export station).

The customers of GTS are called shippers. Shippers are parties who enter into a contract with GTS for the transmission of natural gas. Shippers can be traders, energy companies, gas producers or end users from the chemical industry. The transmission of gas from entry to exit points is organized by means of contracts between GTS and shippers.

(8)

maintained between the volume of gas injected into the system and the volume off-taken within a gas day.

GTS provides both firm (assured) and interruptible (non-firm) capacity. Firm capacity means that a customer knows for sure he can use the capacity he has contracted with a 100% guaranty that the capacity is available on every day in the year on every hour. Interruptible capacity runs a certain chance of being interrupted and is therefore offered to a lower price. This means that its availability is below 100%. It will not be offered until firm capacity is sold out. Interruptible capacity is offered based on the past utilization rates of firm capacities that have been hardly used.

1.4 Problem Statement

The issue is that planning has become much more complex since the liberalization of the European gas market in 2005. Since then, the former Gasunie was split into two independent companies. Gasunie is still the Dutch gas infrastructure company, but trading and selling gas became the solely responsibility of Gasterra.

Liberalizing the European gas market means that all parties on the market must have non-discriminatory access to the national gas transmission grid and local distribution grids (Services included, 2008). For example, Gasunie is not allowed to favor any specific shipper. It has to treat all its shippers equally and assure them full access to its network, when capacity is physically available.

Before the liberalization of the gas market, the system for selling gas transport and services was known as the point-to-point system. This means that the combination of entry and exit points, that customers were going to use, were already fixed and known beforehand. The new regime, called decoupled entry-exit system, is creating more flexibility for the customers on the one hand, but decreases the level of predictability of gas transport routes on the other hand (Services included, 2008).

(9)

of a GTS customer that is buying gas on the same. Due to the possibility of gas exchange, shippers and traders have the opportunity to buy or sell gas anonymously. Although facilitating the gas market by creating more flexibility, the process of capacity planning in the gas network becomes more unpredictable.

Within the Netherlands there is a tendency of shippers booking more entry or exit capacity than they actually are about to use. Shippers do this to create flexibility in their portfolio. In this way the shipper can buy and sell gas at different trading hubs. Shippers do not want to minimize any transportation restrictions on their trading activities. This means, that the contracted capacity of transporting natural gas is higher than the final allocated capacity in the transmission grid. This is leading to unused contracted capacities in the network.

Actually, this outcome is counterproductive to the proper functioning of a liberalized gas market and against the first European Gas Directive from 1998, which was supposed to prevent distortive market behavior. According to Gas Transmission Europe (GTE, 2005), the liberalized European gas market should ensure the maximum use of capacity, prevent anti-competitive behavior; ensure non-discriminatory and transparent access to the transmission network. Instead, artificial capacity shortage and price implications became the side effects of liberalization.

The Commission de Régulation de l’Electricité et du Gaz (CREG, 2005) defined this behavior as capacity hoarding. By fully booking the available entry or exit capacity a shipper can prevent another competitive shipper from booking capacity and offering gas to the customer at a given point. By withholding capacity from primary and secondary markets, shippers can block or hamper their access to other competitors. The CREG (2005) mentions several motives for that. First of all, shippers may want to avoid new entrants on the market to keep a dominant position and control downstream gas prices. Secondly, they may want to create capacity congestion on the primary market to create speculative scarcity of resources on the secondary market, like the TTF for instance.

(10)

The activities of Gas Transport Services are regulated. The activities are monitored by the so-called ‘Energiekamer’. The Energiekamer wants to strive for a maximum competitiveness in the region and wants to ensure a maximum utilization of the given infrastructure. The behavior of capacity hoarding could result into inefficient infrastructure investments and inefficient route planning.

Here emerges the challenge for GTS to adapt to a new market situation. In order to make liberalization a success, it is necessary to assess and further optimize planning activities where possible. In parallel to the professional perspective of optimizing capacity planning this research has also relevance from a scientific point of view. So far, there are no ready-made approaches or methods provided by science for capacity planning, tackling the inefficient use of contracted network capacities in a liberalized gas market. Especially methods and approaches that can cope with the inefficient use of network capacities and make more capacity available to the market would be interesting.

1.5 Research Objective

The problem described above can be seen from a marketing perspective and an operations management perspective. The marketing perspective would involve an overall understanding and analysis of the market. This would enable Gasunie to make more entry and exit capacity available to the market (and thus increasing revenues). The focus would be on understanding the behavior of shippers and evaluate the way GTS is selling entry and exit capacities to its customers.

In this case, we’re looking from an operations management perspective on the problem statement. The focus is on understanding the way planning the gas infrastructure is done within GTS and if new concepts could be triggered for making even more capacity available to the market. This would enable GTS to make its route planning more efficient and make better use of its physical capacities. Here the process of capacity planning has to be investigated.

(11)

science, most literature is focused on logistics, but seldom directly connected or relevant to the transportation of natural gas.

Capacities at entry and exit points of gas transmission systems will be the main concern of this research, because the utilization of entry and exit capacities in the liberalized gas market are the main driver for capacity planning. Moreover, capacity is the main keyword and refers to the domain of capacity management in operations management. Hence, capacity management is chosen as a suitable scientific framework, for this kind of research topic. According to David Barnes (2008), capacity management is concerned with the ability to meet customer demand and respond to changes over time.

Capacity management aims at solving several basic issues of any production system. David Barnes (2008) mentions the problem of measuring the productive capacity of an organization, forecasting the right demand and opting for the right strategic approach in capacity planning.

This research will only consider gas flows between cross border points in the Netherlands, which will reduce the amount of considered entry and exit points. Cross border points are essential to analyze the effect of the liberalized European gas market on infrastructure planning. They connect the Dutch national gas transmission network with gas transmission networks in neighboring countries, such as Germany, Belgium and the United Kingdom. To summarize, the aim of this research is to analyze the current way of capacity planning within GTS and suggest possible approaches for improvement. The assumption is that possible improvements will deliver a higher business performance with regard to the efficiency of route planning and capacity utilization for instance.

1.6 Research Design

Relating to the problem statement and research objective, keywords as infrastructure management and capacity management give the direction for the research question. Furthermore, looking for improvement of capacity planning is an essential part of the research. The research question is formulated as follows:

(12)

In order to structure the thesis and elaborate the research question, the main question will be broken down into sub questions. These sub questions will enable a stepwise approach to answer the research question.

The first sub question is concerned with analyzing the current capacity planning at GTS and will be treated in chapter 2. Understanding the planning process is the basis for analysis and making improvements. Therefore the business process of capacity planning has to be analyzed and depicted carefully. The question is formulated as follows:

1.) How is capacity planning executed within GTS?

Besides, it is relevant to see how the liberalized gas market affects GTS’s capacity planning in order to improve the same. Here it is interesting to know the gas flows at cross border points in the Netherlands, which will be analyzed in chapter 3. The second sub question is:

2.) Which magnitude do the unused capacities have at cross border points?

To improve capacity planning at GTS, it is necessary to analyse current literature available on capacity management in the liberalised energy market, including gas and electricity. Furthermore, more general literature available on capacity management in infrastructure and transportation science will deliver new insights. This will be treated in chapter 4. The third sub question is formulated as follows:

3.) Which alternative or additional approaches to capacity planning could be suitable?

After all three sub questions have been answered successfully, the main research question can be answered. Due to previous findings, it will be possible to assess alternative planning approaches, complementary to the existing capacity planning on the GTS case. Outcome of this research will be recommendations that have the intention to improve the current way of capacity planning. The recommendations are formulated in chapter 5.

1.7 Research Methodology

(13)

we became and introduction to the activites of Gasunie and the European energy gas market by experts from various fields. Afterwards the collection of quantitative data has been started. From September till November historical data of gas flows through the Netherlands has been collected and analysed. In the second half of the student project, simulation studies based on the Dutch gas transmission grid were done. These had the objective to obtain quantitative data on the network utilization rates based on historical gas flows. In the academic part, a desk research was conducted to find literature on capacity management approaches in infrastructure and transportation science. Finally expert meetings with capacity planners from GTS helped to find suitable solution approaches and assess their feasibility.

In order to answer the research questions an appropriate methodology is needed. As the aim of the research is to identify possible ways of new approaches to the capacity planning process within Gasunie, the methodology of Thomas H. Davenport seems to be appropriate. In his book called “Process Innovation: Reengineering work through information technology” (1993), Davenport describes a framework for process innovation. He mentions that a business should not be seen in terms of certain functions, products or divisions, but rather in terms of key processes. His methodology has been chosen, because it has an organizational focus with the clear aim to achieve business transformation. Furthermore, Davenport’s framework has been helping and guiding in identifying and developing opportunities of maximizing network utilization. In this case the key process is capacity planning, which has constantly to be evaluated and innovated to offer new services where possible and help further facilitating the liberalized Dutch gas market.

Though, the question that has first to be clarified here is what is meant by the words process and innovation. “A process is thus a specific ordering of work activities across time and place, with a beginning, an end, and clearly identified inputs and outputs: a structure for action. This structural element of processes is the key to achieving the benefits of process innovation.” As stated by Davenport (1993:5). Concerning the word innovation, he makes a clear distinction between improvement and innovation. In line with Davenport, process improvements happen on an incremental level of change, are continuous and rather moderate in their risk. On the other hand, process innovations are on a radical level of change, are happening one-time and having a high-level of risk.

(14)

Davenport presents a high-level approach to process innovation. It is based on five steps, but its sequence is not binding for the user. In some cases, it is better to start upside-down or just in the middle of the framework. The five steps are as follows:

1. Identifying processes for innovation 2. Identifying change enablers

3. Developing process visions 4. Understanding existing processes

5. Designing and prototyping the new process

The first step applies to the exploration of processes that can be potential innovation carriers. Identifying processes for innovation starts with enumerating major processes and determining their boundaries. Afterwards it is important to assess their strategic relevance to the company. It is also relevant to explore the condition and situation of each process, so that opportunities for improvement can be identified.

The second step is the identification of change enablers, which involves searching for technological and human opportunities. These could be for example information technology, organizational or human resource issues. Identifying the constraints and opportunities of potential enablers is crucial for a successful process innovation. Applying an enabler should not occur without considering its constraints and determining which can be accepted.

Developing a process vision or several process visions is the next step of the framework. Through a process vision for a selected process, an initial statement of process innovation is given. It gives an idea how things can be done differently and better. Furthermore it gives guidelines about the objectives and attributes of the same. A process objective implies a process goal, a certain type of improvement and a quantitative target for the innovation. Whereas a process attribute is more descriptive, it gives the process operation in a future state and includes the function of an enabler. Last but not least, the process vision should be aligned with the business strategy of the company and also consider process users and customers.

(15)

assess the processes in terms of current technological and human enablers. In addition, the processes have to be measured in terms of process objectives and process attributes.

Finally, designing and prototyping the new processes is the last step of the framework. In this stage new process designs have to be brainstormed and developed for a selected process. Then the resulting design alternatives have to be assessed on their feasibility, risk and benefit. In the end the preferred process design will be opted. Afterwards prototypes of the new process design, a transition strategy and new organizational structures and systems will have to be developed.

(16)

2 Planning Process

2.1 The Dutch gas transmission grid

GTS is responsible for the management, operation and development of the national gas transmission grid. In this section, the underlying infrastructure and functioning will be clarified to give a better understanding of GTS’ its daily task.

Since the unbundling of the gas market, the gas transport network in the Netherlands is operated as fully uncoupled entry-exit system. Its main task is to determine the available transport capacity for its customers based on the given infrastructure. The uncoupled entry-exit system means that capacities at entries and entry-exits are sold separately. A customer can choose the combination of entries and exits he wants to buy. Hence GTS doesn’t know beforehand which route shippers are using to deliver gas through the network. For instance, a customer can inject gas into the network, at a specific entry, and withdraw it at any other of, for example, the two exits he booked in advance.

In the past GTS had to plan gas transport on a point-to-point basis and its main task was security of supply. At that time, GTS was an integrated company, supplying and transporting the gas. There was only one shipper and gas streams were more predictable, because the demand and technical capacity were well known (Dick Vermeulen, 2008). Furthermore GTS was able to adapt the gas supply according to its own strategy. Hence it could decide where the gas has to come from and how it should be routed through the network, enabling an overall optimization of the network. The point-to-point system was easier for the planning, as the routes of gas flows were known beforehand.

(17)

obtain the required quality of gas, there are blending stations that mix the different types of gas. To give an example of the different gas qualities, the high calorific gas, called H-gas,

has a calorific value of about 50 MJ/m3 according to the Wobbe index and the low calorific

gas, called G-gas, has a Wobbe index of about 44 MJ/m3.

Figure 1 gives an overview of the transportation grid. This technical scheme represents the main components and routes of the network.

(18)

2.2 Capacity Planning

The process of capacity planning has to achieve optimal transmission plans and drafts the design of large network expansions to assure the reliability of the gas transmission grid within technical constraints. Understanding the planning process is the basis to analyze and make improvements. Here the first step of Davenport’s methodology will be applied: Identifying processes for innovation. Following sections will describe the procedure of capacity planning and identify its major sub processes.

Capacity planning starts with the collection of demand and supply data for the transport of natural gas (Kiewiet, 2007). Inputs are contracts, prognoses and market information. The elementary data used are the entry and exit contracts of shippers with varying contract lengths (typically 1 to 20 years). This information is usually not enough for the prediction of long term trends and the planning of network expansions since new developments in the gas market, new commercial consumer projects (power plants, industry) or new international flows are not know at forehand. Gas Transport Services can only expand the network if the investments are substantiated by firm capacity contracts. In this way the investments will be earned back, based on the additional income generated by new capacity contracts. Market parties however can not commit themselves at an early stage in business development. At the same time, the lead time of typical gas infrastructure projects is 5 to 10 years. This indicates the time wise complexity in assessing future trends in gas transport. In order to have more certainty about future needs and developments on the market, GTS is undertaking meetings with its shippers. In these meetings, shippers indicate their long term needs of transport capacities (more than 5 years ahead). If the estimated supply and demand leads to bottlenecks and evokes the need of new investments into the transmission grid, it is very important to have reliable supply and demand data. With the information resulting from markets, prognoses, contracts and Open Season, the uncertainty whether or not capacities will be contracted on a longer term becomes more predictable.

After the supply and demand data has been collected, the contracted capacities and other information about supply and demand at entry and exit points will be used to run computer simulations of the gas transmission network.

(19)

compressors, blending stations and valves for example. In addition, the database also contains the information about contracted entry and exit capacities. The Plato database is the backup for all other software applications and provides them with the necessary network and capacity information.

The main tool used for capacity planning is MCA, a comprehensive simulation model for the Dutch gas network. MCA stands for Multi Case Analysis and allows running simulations based on a piecewise linear optimization algorithm. It enables to calculate transport capacities, and to calculate the reliability of the network and make failure analysis. The algorithm tries to minimize the overall cost function of the transportation grid within constraints. The constraints are given by the technical network configuration and the contracted entry and exit capacities. Costs could be the fuel needed for compressor or blending stations, for instance. The graphical user interface of MCA includes a detailed map, visualizing the distribution of gas flows, pressures, qualities and temperatures all over the Dutch network. Through the graphical user interface, also new network parts can be implemented and modeled, by which means future developments and network expansions can be assessed.

There are three distinctive transport capacity calculations in the capacity planning methodology of GTS. These form also the main output of the capacity planning activity. The first one is aimed at calculating the short term need of transport capacities, the second at studying the future developments and the third one at determining the maximum available entry and exit capacities for shippers.

Capacity calculations for the need of short term planning look at a horizon of one to five years. Main focus of attention is to evaluate if the network meets its design criteria, based on the short term contracts and prognoses. For GTS the secure transportation of gas, in a safe and reliable manner, has the highest priority. Hence bottlenecks in the system have to be identified timely to avoid unscheduled interruptions in gas transport.

(20)

while a specific combination of entry and exit points is upgraded, because the maximum capacity will depend on the overall load of the network. In this way the maximum entry or exit capacity that could be made available to the market is calculated.

In previous sections a general description of capacity planning and its sub processes have been presented. Figure 2 is an adequate representation of the capacity planning on a high level process flow chart.

Figure 2: Capacity planning process flow chart

(21)

Above mentioned elements will be described in the following sections of this chapter, including their definition, meaning and procedure. First of all, the relation between these four elements will be illustrated here.

The relationship between the process steps of the MCA simulations is unidirectional and ends up with estimating the free capacities in the system. Usually the first step is to generate scenarios of possible supply and demand combinations in the network. After the scenarios have been generated, they will be assessed on feasibility in MCA. If the set of scenarios can be accommodated by the network, the next step will be to analyze the free capacities in the transmission network. Otherwise the configuration of the network will have to be modified, in order to accommodate failing scenarios. Again, a decision has to be made at this point. If the necessary modifications are minimal, the free capacities of the system can be identified subsequently. If they are bigger network changes will be necessary to have the network up to standard again.

This means, that there are basically three possible sequences in the MCA simulations process. The first starts with “Generate Scenarios”, “Assess Scenarios” and “Free Capacities” and would be an ideal situation. But most of the time, the sequence includes all four steps illustrated above. Then the sequence follows the order “Generate Scenarios”, “Assess Scenarios”, “Configurations Network” and “Free Capacities”. The third sequence includes a loop before reaching the state of analyzing free capacities. Here the sequence could be “Generate Scenarios”, “Assess Scenarios”, “Configurations Network”, “Generate Scenarios”, “Assess Scenarios”, “Configurations Network” and “Free Capacities”. The number of loops can also be higher in the third sequence.

Generate Scenarios Assess Scenarios Configure Network Free Capacities

(22)

2.3 Scenarios

To understand the meaning of the term scenarios as it is used by GTS, an explanation has to be given first. In short, scenarios try to translate probable shipper behavior into routes of entry and exit combinations. That’s why the term “generate scenarios” is used in the previous graph. GTS assesses these scenarios to check if the network can accommodate the flows properly.

In a point to point system it would not be necessary to develop shipper scenarios, because the supply of gas, the delivering route in the network and the end destination are well known at forehand. Looking closer at the possible routings of gas transport, it is important to keep in mind the particularities of the decoupled entry-exit system. The customer is free to choose the combination of entry and exit points, according to his portfolio of contracted capacities. Later on a concrete example will elaborate on these options. Shippers do not want to have transport limitations and usually book more entry than exit capacities. So that they can flexibily choose the amount of gas they want to deliver to a certain point.

Besides the transport capacity sold on the primary market, gas that has been physically injected into the network can change ownership on the secondary market. This can happen before the gas physically leaves the network at any exit. This means that the gas, already in the system, is sold to another shipper on the short term gas market. Due to the fact that natural gas injected or extracted cannot be distinguished physically, it can easily change ownership, which finally makes gas trade possible.

Given the many different choices the shippers can make, Gas Transport Services developed a set of fixed scenarios that cover these worse case entry/exit combinations, according to which capacity calculations are performed. These scenarios form the design basis for the network.

(23)

has to balance the grid. This means that the supply and demand have to be matched despite different gas qualities. If 70% of the supply is H-gas and 50% of the demand is G-gas, the 20% surplus of H-gas have to be blended with nitrogen to get enough G-gas. These different demand and supply situations with respect to gas quality and exchange between H- and G-gas, are covered in the earlier mentioned set of scenarios.

2.4 Generating Scenarios

As mentioned before, a scenario describes the way in which shippers can use their contracted entry and exit capacity. When a scenario is generated, one flow situation in the transmission grid is fixed and balanced (supply equals demand). Figure 4 is a simplified model of a gas transmission network. It will be used to describe the complexity and possibilities in developing scenarios.

Technical Capacities Entry1 200m3/h Exit1 200m3/h Entry2 200m3/h Exit2 200m3/h Connector 200m3/h

Figure 4: Simplified gas transmission network

The model has two entries, two exits and three pipes. One pipe is going from Entry 1 to Exit 1, another from Entry 2 to Exit 2 and the third one is connecting both of them. Each

entry and exit has a technical capacity of 200m3/h, which means that no more than 200m3/h

of gas can flow through each pipe. Regarding supply and demand, 400m3/h would be the

maximum amount of natural gas that can be injected and extracted from the system. Furthermore, the technical capacity equals the maximum firm capacity that can be contracted by shippers

(24)

scenarios determine which capacities can flow through the system. A multitude of scenarios can be proposed, but only those scenarios that will determine the available capacity are taken into account. This selection is based on two main aspects (Bert Kiewiet, 2007). The first is to figure out how realistic and important a scenario could be. Secondly, a final selection of worst-case scenarios is made. This final selection has to make sure that all critical situations of the network are covered.

At GTS scenarios are defined by several features. The combination of entries, exits and supply and demand capacities are the basic data. But in general a scenario is not only defined by a special gas routing in the network. Another feature is the quality of gas injected into the system. A scenario is also determined by the amount of H-gas injected into the network. The amount refers to the H-gas strategy (Bert Kiewiet, 2008) of GTS.

If the strategy is at its maximum, much more H-gas then needed would be injected. On the other hand, if the H-gas strategy is minimal, just sufficient H-gas to supply the H-gas demand would be injected by shippers. Therefore, the scenarios have to be balanced, which means that the different supplied and demanded gas qualities have to be matched. As mentioned in previous section, if 70% of the supply is H-gas and 50% of the demand is G-gas, the 20% surplus of H-gas have to be blended with nitrogen to get the expected amount of G-gas.

Besides routings and gas qualities, GTS is considering different temperatures in its scenarios. In general GTS is using three different temperatures for that, these are -17, -7 and +10 degree Celsius. It is necessary to investigate the effect of different temperatures, because especially domestic consumption changes considerably with temperature and therefore the load of the gas transport infrastructure changes. For instance, if the temperature is lower, the domestic consumption for heating will increase and compressor stations will need more energy, thus making more costs, to compress the gas and keep it above the required minimum pressure. On the other side, if the temperature is about ten degrees, transportation costs and capacities will be rather low. But the need of gas by industries and plants remains stable and rather independent from the variation of temperature.

(25)

Figure 5 gives an example of three scenarios. For each of them, the amount of used transport capacities is defined and represents in which way shippers could use their contracts. Generally speaking, transport capacities describe the total utilization of firm contracted capacities by shippers. Applying other features such as gas qualities and different temperatures would go beyond the scope of this simplified model and will not be considered here.

In this case the contracted capacities are equal to the technical capacities of 200m3/h per

entry and exit. This means that the firm capacity is sold out. Nevertheless, it is assumed that

the specified transport capacities in each scenario are lower than 400m3/h for supply and

demand, because shippers usually never utilize all firm contracted capacities. Scenario 1

has a total supply of 300m3/h and a total demand of 300m3/h. The required transport

capacity at each entry and exit point is about 150m3/h. In Scenario 2, the total amount of

supply and demand in natural gas is also 300m3/h, but the distribution of transport

capacities is totally different. At Entry1 the capacity is 200m3/h, 100m3/h at Entry2,

100m3/h at Exit1 and 200m3/h at Exit2. In Scenario 3 the supply is lower than the demand

of natural gas. The supplied transport capacity is 250m3/h and the demanded transport

capacity 300m3/h, this is not a feasible situation since the shippers would be unbalance.

Scenario 1 Transport Capacities Scenario 1: Supply 300m3/h, Demand 300m3/h 150m3/h 150m3/h 150m3/h 150m3/h Scenario 2 200m3/h 100m3/h 200m3/h 100m3/h Scenario 3 200m3/h 200m3/h 100m3/h 50m3/h Scenario 2: Supply 300m3/h, Demand 300m3/h Scenario 3: Supply 250m3/h, Demand 300m3/h

Figure 5: Simplified gas transmission network scenarios

(26)

2.5 Assessing Scenarios

After the generation and final selection of scenarios, GTS is using a set of 50 scenarios to run its network simulations in MCA. By this means, GTS is assessing if the contracts and prognoses of transport capacities are feasible and if GTS can assure a safe and reliable transport of natural gas, given these demand and supply scenarios. This assessment occurs in the comprehensive environment of MCA. This simulation program allows analyzing the flow of natural gas through the Dutch transmission network

After calculating the given set of scenarios, the simulation and optimization tool MCA shows the results on its graphical user interface. This enables the user to go through all fifty scenarios and check if they can be accommodated, by the Dutch transmission network. If a scenario can be accommodated, it will be categorized as a success, if the scenario can not be accommodated it will be categorized as a failure.

A failed scenario will be further analyzed. MCA shows where the bottlenecks are and suggests the cause of problem. But most of the time, the capacity planner has to retrace the cause of the bottleneck in the network manually, to fully understand the problem. Here expert knowledge is used to assess the situation.

For the simplified network that was introduced in previous section, the process of assessing scenarios is illustrated by different colours for the simulated gas flows, as can be seen in

Figure 6. In Scenario 1 the gas flow is 150m3/h at each network point and coloured in

orange. This means that 150m3/h is flowing from Entry 1 to Exit 1 and as well from Entry 2

to Exit 2. The third pipe interconnecting both other pipes will not be employed in that case. Comparing this simulation result with the predefined transport capacities shows that Scenario 1 is a success and can be accommodated by the network.

Scenario 2 is also a success, but the simulated gas flow makes use of the interconnector

between vertical pipelines. As the simulated gas flow is 200m3/h at Entry 1 and only

100m3/h at Exit 1, the residual gas flow of 100m3/h has to go somewhere. In this case it is

going to Exit 2 through the interconnector. Therefore the simulated gas flow is red between Entry 1 and the interconnector, as well as between the interconnector and Exit 2. The rest of

the system is transporting 100m3/h and coloured in green.

Scenario 3 is more complicated, because gas is flowing in the simulation, but the full

(27)

Exit 1 and simulated in red, which is conform with the specified transport capacities. But

the pipe connecting Entry 2 and Exit 2 is marked in blue, which means that only 50m3/h is

flowing to Exit 2. According to the specified transport capacities, the flow to Exit 2 should

be 100m3/h. In this case the scenario needs further analysis and the network might have to

be reconfigured to get the missing 50m3/h to Exit 2, which will be further discussed in next

section.

Figure 6: Simplified gas transmission network assessment

2.6 Configuration Network

Depending on whether the assessment of scenarios was a success, this sub process has to be undertaken to accommodate failing scenarios. Usually GTS is retracing the problem of a bottleneck and looking at different options to accommodate the failing scenario.

One option implies minimal modifications in the network by varying the configuration of technical components. For instance the compression ratio of a compressor station could be too low to deliver sufficient gas to subsequent points in the network. Hence the compression ratio has to be increased by installing additional compressor units (high investment). Another example could be that a closed valve in the network has to be opened, and by doing this using another or more connecting pipelines to deliver sufficient gas to an exit.

For the failed Scenario 3 this means that the cause of the problem has to be retrieved carefully, as can be seen in Figure 7. The bottleneck is at Exit 2, because the actual needed

output of 100 m3/h gas can not be reached. The simulation is lacking 50m3/h of supply to

Exit number 2. Getting the missing amount of gas from Entry 2 is impossible, because the

(28)

missing 50m3/h from Entry 1 to Exit 2 through the interconnector. But as the Entry 1 is already at its maximum and delivering all the gas to Exit 1, this situation is infeasible. Hence, the only option that remains is to alter Entry 2. Because on one hand there are no technical components that could be switched and on the other hand investing in a higher technical entry capacity would not be necessary.

In order to accommodate Scenario 3, the predefined transport capacity at Entry 2 has to be

increased about 50m3/h. This implies that Scenario 3 has to be regenerated and reassessed.

The renewed scenario has a total supply of 300m3/h and a total demand of 300m3/h. As can

be seen in following figure, Scenario 3 can be successfully accommodated by the network.

Now there is a simulated gas flow of 100m3/h between Entry 2 and Exit 2.

Problem 200m3/h 200m3/h 100m3/h 50m3/h Infeasible 200m3/h 200m3/h 100m3/h 50m3/h Scenario 3 200m3/h 200m3/h 100m3/h 100m3/h 50m3/h 150m3/h 200m3/h 100m3/h Correct Scenario 3 50m3/h

Figure 7: Simplified gas transmission network configuration

To give an overview of the new and successful set of scenarios, the three actual and assessed scenarios are listed in Figure 8.

(29)

2.7 Free Capacities

After the successful assessment of all 50 scenarios, GTS is analysing available free capacities in its network. This is done in order to know which transport capacities can be awarded to shippers. Furthermore, the analysis of free capacities allows to determine the maximum available capacity of entry and exit points, which is composed out of the actually contracted and free capacities. The free capacities are basically dependent on the amount of firm capacity that has already been contracted at various entry and exit points. If the firm contract available capacity at every point in the network has been sold out, there are no free capacities left.

GTS is using an approach called NX-methodology (Bert Kiewiet, 2008) to identify free capacities and assess the maximum available capacities in the network. Here an important criteria is that gas entering at an entry point must be able to be transported to any exit point in the transmission grid, due to the uncoupled entry and exit system.

First of all the NX-methodology considers the whole network and uses the predefined set of scenarios as base case. Then the next step is to draw some combinations of worst case entry-exit routes for one specific entry and repeat that procedure for all other entries in the network. Afterwards the maximum capacities can be assessed in MCA by applying the entry/exit routes on the given set of scenarios. This occurs by selecting an entry-exit route that has to be increased, while leaving all other entries and exits fixed to their base case values. This procedure usually demands to increment the value of the selected route successively until one or more scenarios fail in the simulation. The highest increment that can be accommodated by the set of scenarios is equivalent to the free capacity of the selected route.

In general this is the standard approach used to identify available capacities in the transmission grid. But as entry and exit capacities are sold separately in a decoupled entry and exit system, it can be of additional value to have a look at isolated points or to try different approaches. This enables to analyze the free capacities for one point or special situations in the network.

The NX-methodology has also been applied to the simplified model. Those results can be seen in Table 1. Four routes have been available to increase the given transport capacities. The staring point was route Entry 1 to Exit 1, where only Scenario 1 could be increased

(30)

they already reached the technical possible maximum at their entries. Consequently the free capacity was zero for that route. This can be read in the last column of Table 1, which is called “MaxCap” and stands for the maximum available free capacity on a given route. Using the NX-methodology for the other routes delivered the same result, the free capacities left in the simplified model are equal to zero.

Table 1: Free capacities for routes of the simplified network

Instead of looking only at routes, an other option is to check the free capacities at individual

entry and exit points. The results are listed in Table 2. In this case, there are 50m3/h

available at Entry 2, because the transport capacities at this point vary between 100 and 150

m3/h. On the other hand, the remaining points reach at least once the maximum capacity of

200m3/h. Therefore the final free capacities are 0m3/h at Entry 1, Exit 1 and Exit 2 but

50m3/h at Entry 2.

(31)

3 Diagnosis of capacity flows

3.1 Aim of Diagnosis

After having identified capacity planning as a process for innovation and having described its sub processes in Chapter 2, this chapter will follow the next step of the research methodology. Understanding existing processes is important to know how to deal with underutilized capacities. It will help to find unrecognized problems and assess its performance. The aim of this diagnosis is to gain further insight into the effect of the liberalized gas market on capacity planning. Because of the tendency of shippers to buy more entry and exit capacity than they will actually use, we can assume that the overall gas load of the grid is lower than expected. As the elementary data used for capacity planning are the entry and exit contracts of shippers.

GTS is responsible for the management, operation and development of the national gas transmission grid. One of the main tasks is to determine the available transport capacities for shippers based on the given infrastructure. However, the behavior of capacity hoarding could result into inefficient infrastructure investments and inefficient route planning, due to underutilized contracted capacities.

The design of the network is based on a fixed set of supply and demand scenarios that the network always should be able to accommodate. The set of demand and supply scenarios contains worst case, but realistic, combinations of demand and supply. The set of scenarios contains scenarios for the temperature range between -17°C to +20°C. Each scenario is unique, in the sense that it has pre defined the way that demand and supply is matched. Three variables define the character of the scenario: different geographic orientation (for example most supply from entries in the North East or South West), gas quality (assuming a minimum or maximum surplus of H-gas, calling for quality conversion into G-gas), temperatures (-17°C being the design case for the system to +10°C focussing on minimum flow situations, quality conversion, and filling of storages).

(32)

In order to get insight in the utilization degree at entry and exit points, the difference between contracted and finally allocated capacity is studied. For this diagnosis, all 13 cross border points of the Dutch grid with Germany, England and Belgium will be considered. Cross border points are interesting, because they represent big gas flows through the network and represent mainly commercial gas flows. The other entry and exit points are much more dispersed, smaller, dedicated to the local household market and more predictive due to their temperature correlated demand.

After evaluating gas flows at cross border points and making conclusions about their utilization rate, it is necessary to update the design scenarios mentioned before and to replace the contract figures in these scenarios with the estimates based on realizations. The next step is to reassess the design scenarios of the Dutch gas infrastructure and investigate if there are any additional free capacities left in the network. Here it will be interesting to assess to what extend capacity planning could be partly based on realizations. The possible gain in overall available capacity is evaluated and the feasibility of this approach is discussed at the end of this chapter.

(33)

3.2 Approach, methods and data

The Figure 9 shows the approach taken for the diagnosis. First of all, the historical gas flow data of cross border points has been be collected and analysed. Next, the current set of 50 scenarios were reassessed, by replacing the contract figures at cross border points with estimates of realizations instead. This set of scenarios based on estimates was used to run simulations in MCA. After the successful balancing of the set of scenarios in MCA, the free capacities in the network could be estimated. This has been done by comparing the free capacities in the network, based on regular contract values at cross border points with the free capacities in the network based on estimated values. This gives an indication of the overall gain in free capacity, if one would base the capacity planning on realizations.

Processing the historical data includes extracting data, plotting data and doing some statistical analysis. To get the historical gas flows of all 13 cross border points, it was

(34)

relevant to extract the relevant data from Gasunie’s businesswarehouse database. Therefore, the relevant data had to be defined. The first restriction was already set, because the selection of points had been restricted to cross border points, which help to better understand transit flows through the Netherlands.

The second restriction was the time horizon. As the liberalization started in 2005, the relevant period of time was given. But due to the fact, that the first year of liberalization could not be seen as representative enough by the staff of GTS, the time horizon has been restricted to July 2006 until July 2008. The idea was that after a start-up period, the behavior of shippers would be more normal and experienced. This on the other hand would allow visualizing the patterns of their behavior at cross border points.

The third condition was that the data includes the relevant transport capacities. Here the contracted capacities and the allocated capacities on an hourly basis, were input to the analysis of gas flows and utilization rates. As the database registers the network gas flows in cubic meter per hour, the datasets resulted into 17542 hours for contracted and allocated capacity per cross border point. Finally, the last condition was to specify which kinds of capacities were used for contracts and allocations. The reason is that there are firm capacities and interruptible capacities. In this diagnosis only firm capacities haven been considered, as all capacity planning is based on them. Even interruptible and backhaul capacities are calculated upon firm capacities.

After having carefully processed datasets of all 13 borderpoints in spreadsheets, the data has been plotted by means of three different diagrams. The first type of diagram was made to show the difference between contracted and allocated capacities for each cross border point (See Figure 10 and 11). Here the standard line-digram has been applied to plot the allocations and contracts over the last two years. These plots are useful, because they given an overview of characteristics underlying a cross border point, for example the volatility of allocations and seasonal demands.

(35)

The third type of diagram was applied to indicate unused capacities that have been contracted (see Figure 12). Starting in July 2006 and going until July 2008, the diagram visualizes to what extend contracted capacities were actually utilized. These type of diagram is based on the ratio of allocated to contracted capacity.

In order to assess free capacities at cross border points, some analysis is needed. These analysis formed the basis for later assessments in the next steps of the diagnosis. The main aim of the statistical analysis is to understand the realized physical gas flows ar cross border points in distinction to capacities shippers had contracted before. This delivers the average used capacity of each point and helps to categorize them by degree of utilzation. The method used to define realistic used capacities is applying the LDC and Utilization diagram instead of an arithmetic average. The arithmetic average by definition takes all numbers in a list to make their sum and divide them by the number of items in the list. This is correct in general but not sufficient for covering the average amount of used capacities. The arithmetic value delivers a number that is too low and could lead to overestimated free capacities in the system, because for a significant amount of time the actual used capacities will exceed this arithmetic mean.

Using the maximum allocated capacities of a point by excluding peaks delivers the right approach. Here the LDC has been used to distinguish the peak hours from the normal loads. Tracing the curve from the right to the left and cutting off the part with the steepest inclination results into the value of average utilized capacity. This actual value gave information about the flow, but not the percentage of utilization.

3.3 Trend Analysis

(36)

Exit K 0 200 400 600 800 1000 1200 01- 07-2006 01- 08-2006 01- 09-2006 01- 10-2006 01- 11-2006 01- 12-2006 01-2007 01- 02-2007 01- 03-2007 01- 04-2007 01- 05-2007 01- 06-2007 01- 07-2007 01- 08-2007 01- 09-2007 01- 10-2007 01- 11-2007 01- 12-2007 01-2008 01- 02-2008 01- 03-2008 01- 04-2008 01- 05-2008 01- 06-2008 01- 07-2008 Al lo c a te d qua nti ty [ N m3 /h our] Exit Contracted

Figure 10: Contracted and allocated capacities for exitpoint K

Figure 10 is a diagram of the cross border point K, which shows the relation between contracted and allocated capacity for an exit. The allocations are represented by the blue line and the contracts by the pink line. It is rather typical for the observation made during the diagnosis at other points, because in most of the cases the allocations did not reach the contracted value. For K the allocations are always below the contract line, except for some peaks. At these peaks the shippers utilized more exit capacitiy than contracted, this will have resulted in a commercial penalty for the shipper. If this excessive exit capacity would have jeopardized the stability of the gas transport network, the shippers/consumers would have been shut down to such extend that the nominated capacity is within contractual limits, this however happens very rarely.

The allocations were much higher during winter than during summer months. This is the reason for the line to have two humps, for each winter period in the example. Here, the demand of gas seems to be temperature correlated, as the demand for gas is much higher for heating in households during winter days.

(37)

two humps is difficult to read. Another general explanation for the volatility of the allocation line is that the demand for gas is much higher during day than night and also higher during weekdays. To conclude, K can be seen as a typical example of the household demand for natural gas.

Other cross border points show a totally different allocation curve and also their relation to the contract line is different, for example exit A shown in Figure 11. As can be seen in the following picture, the allocation curve and the relation between the pink and blue line differs from previous example. It is typical of the industrial demand for natural gas, which can be explained by two main aspects. First of all, the allocated demand was similar during summer and winter time, therefore no strong temperature correlation. Secondly, the

allocated demand was flat out and matched the contract line most of the time. Interruptions could be due to (un)scheduled maintenance, weekday/week and shifts.

Exit A 0 200 400 600 800 1000 1200 01-0 7-2006 01-0 8-2006 01-0 9-2006 01-1 0-2006 01-1 1-2006 01-1 2-2006 01-0 1-2007 01-0 2-2007 01-0 3-2007 01-0 4-2007 01-0 5-2007 01-0 6-2007 01-0 7-2007 01-0 8-2007 01-0 9-2007 01-1 0-2007 01-1 1-2007 01-1 2-2007 01-0 1-2008 01-0 2-2008 01-0 3-2008 01-0 4-2008 01-0 5-2008 01-0 6-2008 01-0 7-2008 Al lo c a te d qua nti ty [ N m3 /h our] Exit Contracted

Figure 11: Contracted and allocated capacities for point A

(38)

The utilization graph of point K has been restricted to the winter months as can be seen in Figure 12, because the utilization in winter months is usually higher than in summer months. Hence winter months form the bottleneck for capacity planning, because much more gas is flowing through the network in that period. If winter months can be

accommodated, summer months can be accommodated as well. Winter months have been restricted to October to and including April and summer months from May till September. Therefore the estimate values identified from the winter months are more realistic for the calculation of estimates in the statistical analysis, when considering the design of the network. Since the network should be able to accommodate this maximum flow winter situation.

In Figure 12 the utilization of K is never above 100%, apart from one exception. This exception means that a shipper or shippers went above their contracted capacity for an hour. As already explained before, in this case they have to pay a penalty to GTS. Normally the utilization of K is between 30% and 90%. Only a few times the utilization is above 95%. These peaks are seen as exceptions, these datapoints are incorporated in the load duration curve of Figure 13.

Exit K Winter months 0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% 80.00% 90.00% 100.00% 110.00% 10/1/2006 10/11/2006 10/22/2006 11/2/2006 11/13/2006 11/23/2006 12/4/2006 12/15/2006 12/26/2006 1/5/2007 1/16/2007 1/27/2007 2/7/2007 2/17/2007 2/28/2007 3/11/2007 3/22/2007 4/1/2007 4/12/2007 4/23/2007 10/3/2007 10/14/2007 10/25/2007 11/5/2007 11/15/2007 11/26/2007 12/7/2007 12/18/2007 12/28/2007 1/8/2008 1/19/2008 1/30/2008 2/9/2008 2/20/2008 3/2/2008 3/13/2008 3/23/2008 4/3/2008 4/14/2008 4/25/2008 Used capacity

(39)

In the LDC it is easier to see peak times, idle times and average capacities. On the other hand both diagrams, the LDC and the utilization diagram, have been used to not underestimate the average used capacity of cross border points.

LDC Exit K Winter months 0.00 200.00 400.00 600.00 800.00 1000.00 1200.00 1 552 1103 1654 2205 2756 3307 3858 4409 4960 5511 6062 6613 7164 7715 8266 8817 9368 9919 Cumulative hours A ll o c a te d qu a n ti ty [N m3 /hour s ] LDC Winter

Figure 13: LDC of K for the winter months

The LDC of K in Figure 13 shows that only for a short period of time the curve is above

900 m3/h. Looking at the left side of the diagram, it becomes apparent that the gradient is

almost vertical between 800 and 1000 m3/h. The final cut-off is made at 900 m3/h, which is

the exact value before the gradient becomes almost zero. The related percentage of utilization for this point is 90%. This procedure demonstrated for K was also applied to the other cross border points. The results will be given in following section.

3.4 Categorization of cross-border points

(40)

“Estimated”. For the contract values, the highest value of the specific winter or summer period between July 2006 and July 2008 were taken. The estimated values were identified by cutting off peak hours in the LDC of allocations and represent the cut off for each point in winter and summer.

As can be seen in Table 3, the estimated capacities resulting from allocations are higher in winter than in summer.

Table 3: Estimated capacities for the cross border points

Estimated Capacities

07/2006 - 07/2008

Actual Contracted Estimated

Forward Exit Winter Summer Winter Summer

A 967.30 967.30 967.30 967.30 B 999.65 1000.00 959.67 900.00 C 1000.00 1000.00 800.00 700.00 D 975.60 1000.00 585.36 650.00 E 1000.00 545.11 1000.00 408.84 F 1000.00 1000.00 800.00 700.00 G 1000.00 995.89 700.00 597.53 H 1000.00 989.97 850.00 197.58 I 1000.00 1000.00 700.00 101.89 J 974.92 1000.00 877.42 550.00 K 1000.00 995.02 900.00 597.01

Forward Entry Winter Summer Winter Summer

L 1000.00 985.19 940.00 935.93 M 1000.00 1000.00 800.00 900.00 N 1000.00 1000.00 650.00 620.50

Note: original data normalized between value 0 to 1000.

(41)

Table 4: Estimated utilization rates for cross border points

Categorization of entry/exit points

07/2006 - 07/2008 (total: 17542 hours)

Avg. Used capacity (%) Above average (hours)

Forward Exit Winter Summer Winter Summer

A 100% 100% 13 hours 0 hours B 96% 90% 59 hours 29 hours C 80% 70% 615 hours 98 hours D 60% 65% 336 hours 85 hours E 100% 75% 53 hours 212 hours F 80% 70% 264 hours 76 hours G 70% 60% 33 hours 19 hours H 85% 20% 27 hours 839 hours I 70% 60% 86 hours 67 hours J 90% 55% 123 hours 31 hours K 90% 60% 80 hours 80 hours

Forward Entry Winter Summer Winter Summer

L 94% 95% 12 hours 48 hours

M 80% 90% 71 hours 120 hours

N 65% 60% 20 hours 0 hours

Total hours per period 1086 1599

3.5 Reassessing existing set of scenarios in MCA

(42)

are L and M. These are the most important entry points into the Dutch network. Afterwards follows the cluster H/K in the east, here represented with K. In the southeast is the cluster B/C, here represented with B. In the southwest is the cluster N/O, which is going to become a bidirectional point in 2010, with an entry and an exit. Here it is represented by the new exit O. The last cluster is E in the west. These five clusters are important and will be the main points of investigation for the analysis of free capacities. Therefore it is sufficient to take one point of each cluster as an example of free capacities in the network. The reason is, that comparing few points spread all other the network gives a good insight into the behavior of transit flows. In this case the points are entry L and M, exit K, B, O and E.

Figure 14: Network map with cross border points used for free capacity analysis

(43)

3.6 Estimating Free Capacities

As mentioned before, in order to analyse free capacities it is worthwhile to concentrate on the main clusters and the points selected before. The available free capacity for the different cluster was determined for both the original situation based on contracted capacities and the situation where estimated at these cluster points were taken into account.

Table 5 and Table 6 give the main results from estimating free capacities in MCA. First of all the capacities have been increased in the contract-file for each point separately to reconsider the already existing free capacity in the network. Afterwards the contracted capacity and free capacity have been added up to obtain the total capacity for each point. Then, the free capacities have been increased for the estimate-file. Again, the estimated capacities and the free capacities have been added-up to obtain the total capacity for each point. Comparing these capacities will give an indication of the overall gain in capacity, for the situation when one would base planning activities on realizations from the past.

One of the conclusions is that free capacities based on the estimate file are considerably

higher then the capacities based on the contract file. If they are about 100 m3/h in the

contract file, they are about 300m3/h in the estimate file for instance. An other positive

aspect is, that the Total Capacities in the contract file are lower than in the estimate file, which makes planning based on allocated capacity much more interesting.

Table 5: Free Capacities for contracted values, capacity in m3/hour, data normalised between 0 and 1000

Contract File

Contracted Existing FreeCap Total Capacity

(44)

Table 6: Free Capacities for estimated values, capacity in m3/hour, data normalised between 0 and 1000

Estimate File

Estimated FreeCap Total Capacity

Exit K 882.55 258.27 1140.81 Exit B 959.67 136.68 1096.35 Exit E 742.20 306.83 1049.03 Exit O 0.00 80.00 80.00 Entry L 940.00 179.89 1119.89 Entry M 800.00 205.79 1005.79

The last Table 7 shows the difference between the total capacities of both files, where the difference is called additional free capacity. This column indicates how the capacity for these points could be increased if one would assume the estimated realized capacities at all cross border points. It should be noted that these capacities are calculated individually, that means that the additional capacity for one point is for the condition that all the other points are left at the lower realized capacity. The additional capacities for these points are not available at the same time (therefore a ceteris paribus approach). The conclusion is that all six cross border points could individually be significantly increased, with the only exception for M. This means that the free capacities based on the estimate file are higher in total. At M the additional capacity could not be increased, because the network was already saturated in that area. This had to do with the fact, that the estimated capacity for M are already very low and that it is a bidirectional point (D/M). At a bidirectional point, the net flow will be determined by the entry flow minus the exit flow. When based on estimates, the netinflow at the bidirectional point D/M is much higher than for the contracted one, therefore the free capacities could not be increased that much.

Table 7: Additional Capacities as difference of Total Estimates and Total Contracts values, capacity in m3/hour, data normalised between 0 and 1000

Additional Capacities

Total Capacity Estimates Total Capacity Contracts Delta AddCap

(45)

Figure 15 is a bar-chart of the results found in previous tables. All six points have been attributed three bars. The first bar of a point shows its total capacity based on estimates, the second bar shows its total capacity based on contracts and the third bar shows the difference or delta between previous total capacities. The blue part of a bar represents the estimated value; on the contrary a pink bar represents the contracted value and the yellow part of a bar represents the free capacities identified through the simulations in MCA.

Free Capacities -200.00 0.00 200.00 400.00 600.00 800.00 1000.00 1200.00 E x it K E x it B E x it E Exit O Ent ry L En tr y M Delta FreeCap Contract Estimate

Figure 15: Bar chart of Free Capacities for cross border points

Approaches towards maximizing network utilization