Method for route selection of transcontinental natural

gas pipelines

Fotios G. Thomaidis1
National and Kapodistrian University of Athens
Department of Informatics and Telecommunications
[email protected]
Abstract. The route of transcontinental natural gas pipelines is characterized by
complexity, compared to national and cross-border pipelines, since their large
magnitude results in the examination of parameters that do not exist or are
considered negligible for pipelines of smaller scale and that require
management of more information. The aim of the present dissertation is the
development of a route selection method, able to deal with the aforementioned
complexity of transcontinental pipelines. The developed algorithm examines
the validity of the conditions for economic viability of the pipeline, defines the
alternative routes, selects the weights of criteria that affect the pipeline design
and compares the routes, taking into consideration the available data, the
experience and knowledge of the decision maker. The consistency and
sensitivity of the results is examined. The method is applied in the case of a
transcontinental pipeline transporting gas from the broader Caspian Region to
Western Europe. Different scenarios of criteria weights are used and discussed
at the results of the application. The software tool Gas-PRS, allows quick
application of the method and facilitates the decision maker in examining the
consequences of different choices.
Keywords: Optimum route method, natural gas, transcontinental pipelines,
Caspian Region E.U. energy corridor

1. Introduction
Route selection of transcontinental natural gas pipelines is characterized by
complexity compared to the cases of national and cross-border pipelines. This
complexity arises from their large magnitude, which results in the examination of
parameters that do not exist or are considered negligible for pipelines of smaller scale
and that require processing of more information. Large volumes of natural gas need to
be transported for long time periods, of 15 20 years, so that the transcontinental
pipeline is economically viable; thus construction of such a project can only proceed
if long-term gas supplies from large fields are secured and final markets with
adequate demand to buy all the transported volumes are selected. The large pipeline
1

Dissertation Advisor: Dimitrios Mavrakis, Assoc. Professor

length also affects the complexity of route selection, as it results in passing through
territories with different political, economic, institutional and environmental
characteristics. Additional issues concerning transcontinental pipelines are the
prospects of supplying gas to the intermediate markets and the high costs, that depend
on the geological characteristics of the crossed terrain and the transit fees paid to third
countries. The need to examine large amounts of information regarding the pipeline
route increases the complexity of the problem.
Several methods have been developed to select the optimum route of national and
cross-border hydrocarbon pipelines, using the Analytic Hierarchy Process (AHP),
Multi-Attribute Utility Theory (MAUT) and Verbal Decision Analysis (VDA) and
Geographic Information Systems (GIS) [1 11]. These methods can be applied
effectively to small scale projects, but fail to meet the complexity of transcontinental
pipelines. None of the existing methods examines the natural gas supply and demand
and evaluates issues related to the intermediate regions, such as the political,
economic and institutional conditions of the transit countries, the payment of transit
fees and the prospects of supplying gas to third markets. The present dissertation aims
to develop an optimum route selection method, able to address the aforementioned
complexity of transcontinental pipelines.
2 Optimum route selection method
The route selection method for transcontinental natural gas pipelines results in the
selection of the optimum route, after examining a group of alternative solutions which
are defined by the decision maker. The method combines the selections of the
decision maker, regarding the importance of the route parameters, with the available
information and his ability to evaluate them. From this point of view, the selected
optimum route reflects the choices of the decision maker.
The algorithm of the method is divided into three phases (Figure 1), which include
validation of the conditions for economic viability of the pipeline, definition of the
alternative routes, selection of criteria weights, comparison of the alternative routes,
taking into consideration the available data, the experience and expertise of the
decision maker, test of the comparisons consistency and sensitivity analysis of the
results.
Phase 1: Examination of natural gas supply and demand
In the first phase of the method, the two main conditions for economic viability of
the pipeline are validated; the necessity to secure sufficient gas supplies at the
entrance of the pipeline and demand at the targeted markets. In this respect, the
supply potential of the pipeline is assessed and the demand of the final markets is
examined.
The supply potential of countries that constitute possible suppliers for the pipeline
is assessed by examining the countries availability of gas, large gas fields that have
adequate production to supply the pipeline, infrastructure that can transport gas from

the fields to the pipeline entrance, political and financial stability. Gas demand of the
final markets is investigated by examining the countries projections of future
consumption that cannot be covered by indigenous production or existing gas supply
contracts, import infrastructure that is competitive to the studied transcontinental
pipeline, political and financial stability.

Figure 1: Flowchart of optimum route selection method.

If the supply potential is limited, or the demand at the final markets cannot justify
transportation of large gas volumes, then construction of the pipeline is not
economically viable, thus the method is terminated.
Phase 2: Examination of alternative routes
In the second phase the decision maker selects the group of alternative routes. The
alternative routes are then compared using the hierarchic structure of Figure 2 [12].
Criteria level

Subcriteria level

Alternative
routes level

Political and financial

stability
Project risks

Institutional framework
Route 1
Environmental impact

Construction and
operation cost
Optimum route of
transcontinental
gas pipeline

Project costs

Route 2

Transit fees

Use of existing
infrastructure
Route n
Transit markets
Intermediate
markets

Neighboring markets

Regional gas market

Figure 2: Hierarchic structure for comparison of alternative pipeline routes.

The project risks criterion reflects the characteristics of the transit regions that
influence the construction and operation of the pipeline. In includes examination of
the political and financial stability of the transit markets, the homogeneity of
institutional frameworks along the pipeline and the potential environmental impact of
the pipeline. The project costs criterion includes the pipeline construction and
operation cost, the transit fees paid to third countries and the potential use of existing
infrastructure to transport gas to other markets. The intermediate markets criterion

concerns the prospects of supplying gas to the transit markets, neighboring markets
and to regional gas markets.
Calculation of the criteria weights and the performance of the alternative routes on
the criteria is carried out using the AHP method. This method is selected because it
allows examination of both quantitative and qualitative criteria and takes advantage of
the available data on the alternative routes and the experience and expertise of the
decision maker [12]. Comparison matrices are constructed for the criteria, subcriteria
and alternative routes. The elements of each matrix are compared in pairs and graded
using a numeric scale 1 9. The primary right eigenvector of each matrix is then
calculated, to define the local priorities [13]. The uncertainty of the comparison
matrices is examined by performing consistency tests. Two methods are used, the
consistency ratio of Saaty (eq. 1) and the method of Alonso Lamata (eq. 2) [12].
,

(1)

where CR is the consistency ratio, RI the random consistency index, CI the

consistency index, max the maximum eigenvalue of the comparison matrix and n its
size.
(2)
where the consistency limit, defined by the decision maker.
Comparison of the alternative routes requires the examination of large amounts of
information regarding the routes. The method applies methodological tools that
support the decision maker, by facilitating the management of all this information.
These tools are:

Natural gas database, developed on the GIS, which includes worldwide

information regarding the natural gas markets, fields and infrastructure.

Risk analysis, using country risk and regulatory risk indicators of each
routes transit countries. The routes are classified into risk groups
according to their minimum risk indicator and for each route the mean,
coefficient of variance and coefficient of semivariance are calculated.

Estimation of external costs, which reflect the environmental and social

impact of the pipeline. To assess these costs for each route, the
environmental sensitivity and population density of the transit regions are
taken into consideration.

Calculation of supply costs, which reflect the total cost for transportation
of gas from the production field to the final market. The supply costs
include the gas production cost at the field, the transport cost, which
corresponds to the minimum gas price at the final market resulting in
depreciation of the project investment after a given period of time and the
transit fee, which depends on negotiations between the transit country and
the company that transports the gas. If not all the required economic
information of the pipeline are available, the transport cost can be

calculated by applying a learning curve linear function, that uses data

from projects that have already been completed [14].

Method that assesses the prospects of the transit countries for integration
in a regional natural gas market. The Technique for Order Preference by
Similarity to Ideal Solution (TOPSIS) method is used with criteria
regarding the countrys network characteristics, natural gas regulatory
framework and gas market functioning [15].

Phase 3: Selection of optimum route

In the third phase, the local priorities of the second phase are used to calculate the
global priorities of the alternative routes (eq. 3), which define the ranking of the
alternatives. The route with the highest global priority is considered the most
preferred.
(3)
where Ptot,i the global priority of route i, wj the weight of subcriterion j (calculated
by multiplying the weight at criteria level with the weight at subcriteria level) and pij
the performance of route i at the subcriterion j.
The ranking of the routes depends on the selections of the decision maker,
concerning the importance of the criteria. Sensitivity analysis is performed to examine
the connection between the final results and the initial conditions of the problem. The
method is concluded with the selection of the optimum route.
3 Method application
The method is applied in the case of a transcontinental pipeline that belongs to the
energy corridor connecting the broader Caspian Region to the European Union [12,
14].
Phase 1: Examination of natural gas supply and demand
Assessment of the supply potential of the region (Azerbaijan, Egypt, Iran, Iraq,
Kazakhstan, Turkmenistan, Uzbekistan) shows that the examined countries have
abundant available natural gas volumes however transportation of the produced gas
from the large fields to the entrance of the transcontinental pipeline is hindered due to
the lack of sufficient export infrastructures. The final markets of the application
(Austria, France, Germany, Italy) have increasing gas demand and high dependency
on imports, thus flow of Caspian gas to these markets would allow them to cover part
of the future consumption and to diversify their supply sources.
Phase 2: Examination of alternative routes
Four alternative routes are examined, with starting point the town of Erzurum at
eastern Turkey and ending point the Austrian gas system (Figure 3); the older route of

the Nabucco pipeline (route 1), the current route of Nabucco (route 2), the transBalkan pipeline (route 3) and a route suggested by KEPA, passes through northern
Greece, embodying part of the Greece Italy interconnection (route 4).

Figure 3: Alternative routes of transcontinental pipeline.

Since the application was carried out without the participation of a decision maker
who would proceed to the final decision for construction of the pipeline, three
scenarios of criteria importance are examined:

Scenario 1 (S1): Project risks are the most important criterion.

Scenario 2 (S2): Project risks and costs have equal importance.

Scenario 3 (S3): Prospects of supplying gas to the intermediate markets is

more important than the other criteria.

The local priorities of the hierarchic structure are calculated using the AHP
method. The weights of the criteria and subcriteria levels for the three scenarios and
the performance of the alternative routes are presented in Figure 4.
Phase 3: Selection of optimum route
Table 1: Global priorities of alternative routes.
Route 1
Route 2
Route 3
Route 4

S1
0.217
0.316
0.096
0.371

S2
0.220
0.285
0.130
0.365

S3
0.246
0.251
0.089
0.414

Equation 3 is used to calculate the global priorities of the alternative routes. Route
4 has the highest global priority for all three scenarios (Table 1), while sensitivity
analysis shows that this route remains better than the other alternatives, regardless of
the selected weights. Thus route 4 is selected as optimum.

Figure 4: Local priorities of the hierarchic structure.

4 Gas-PRS software tool

The software tool Gas-PRS (Gas Pipeline Route Selector) has been developed with
a view to facilitate the decision maker in applying the route selection method [16].
The software has the following characteristics:

It allows quick implementation of the method.

It is easy to use and includes step-by-step instructions.

It facilitates the decision maker in studying the results of different initial

conditions.

It is connected to a natural gas database and to GIS maps, which can be

easily updated.

It automatically tests the consistency of the comparison matrices.

5 Conclusions
The method presented in this dissertation has been designed to select the optimum
route of a transcontinental natural gas pipeline from a group of proposed alternatives.
It addresses the complexity of the problem by examining the two main conditions for
economic viability of the pipeline, assessing all the parameters that determine the
pipeline route though a suitable hierarchic structure and facilitating the management
of the required large amounts of information with a set of appropriate methodological
tools. The final results of the method depend on the selections of the decision maker
regarding the criteria weights and the availability and reliability of information for the
alternative routes.
The method can be used by consortia constructing transcontinental pipelines,
Governments of countries that aim to export, import or transit natural gas,
International Organizations, such as the European Commission and international
funding organizations such as World Bank and European Bank for Reconstruction
and Development.
The method is applied in the case of a transcontinental pipeline of the south energy
corridor, connecting the broader Caspian Region and Egypt with the E.U. The
application has shown that despite the large natural gas resources of the Caspian
Region, the pipeline cannot be filled, due to the lack of export infrastructure. Four
alternative pipelines starting in Erzurum and ending in the Austrian borders are
compared. The three examined scenarios and the sensitivity analysis lead to the
conclusion that the optimum is the route which travels through northern Greece,
embodying part of the Greece Italy interconnection, and then branched north to
Bulgaria. This route is a win win option as it transits only through European
member states, thus ensuring a level of political, financial and regulatory stability, it
has decreased construction and operation costs due to the use of the Greece Italy
interconnection, it crosses large (Hungary, Romania) and emerging (Greece) natural
gas markets, its location in the center of S.E. Europe allows connection of the pipeline
to the Western Balkan gas systems.

References
1. P.K. Dey, An integrated assessment model for cross-country pipelines, Environmental
Impact Assessment Review J., Vol. 22, 2002, pp. 703 721.
2. P.K. Dey, S. Gupta, Feasibility analysis of cross country petroleum pipeline projects: a
quantitative approach, Project Management J., Vol. 32, Issue 4, 2001, pp. 50 58.
3. M. Ozdemir, T.L. Saaty, The unknown in decision making: What to do about it,
European Journal of Operational Research, Vol. 174, 2006, pp. 349 359.
4. Y. Andreyeva, O. I. Larichev, N. E. Flanders, R. V. Brown, Complexity and uncertainty in
Arctic resource decisions: The example of the Yamal pipeline, Polar Geography and
Geology J., Vol. 19, Issue 1, 1995, pp. 22 35.
5. O. Larichev, R. Brown, Numerical and verbal decision analysis: comparison on practical
cases, Journal of Multi-Criteria Decision Analysis, Vol. 9, 2000, pp. 263 273.
6. S. Feldman, R. Pelletier, E. Walser, J. Smoot, D. Ahl, A prototype for pipeline routing
using remotely sensed data and geographic information system analysis, Remote Sensing
of Environment J., Vol. 53, 1995, pp. 123 131.
7. M.R. Delavar, F. Naghibi, Pipeline routing using geospatial information system analysis,
ScanGIS'2003 - The 9th Scandinavian Research Conference on Geographical
Information Science, 4-6 June 2003, Espoo, Finland Proceedings, 2003, pp. 203 213.
8. M. Iqbal, F. Sattar, M. Nawaz, Planning a least cost gas pipeline route: a GIS & SDSS
integration approach, Proceedings of International Conference on Advances in Space
Technologies, IEEE, 2006, pp. 126 130.
9. J.K. Berry, M.D. King and C. Lopez, A web-based application for identifying and
evaluating alternative pipeline routes and corridors, GITA Oil and Gas Conference, 20-23

2004,
Houston,
Texas,
:
http://www.
gisdevelopment.net/proceedings/gita/oil_gas2004/papers/02gos2004.pdf.
10. R. S. Wasi, B. J. Darren, Spatially Enabled Pipeline Route Optimization Model,
International Pipeline Conference, 2004, Calgary, .
11. M. Mohitpour, H. Golshan, A. Murray, Pipeline design & construction: A practical
approach, ASME Press, 3 , 2007.
12. F. Thomaidis, D. Mavrakis, Optimum route of the south transcontinental gas pipeline in
S.E. Europe using AHP, Journal of Multi-Criteria Decision Analysis, Vol. 14, Issue 1 3,
2006, pp. 77 88.
13. T.L. Saaty, The Analytic Hierarchy Process, McGraw-Hill International Publications,
New York, 1980.
14. D. Mavrakis, F. Thomaidis, I. Ntroukas, An assessment of the natural gas supply potential
of the south energy corridor from the Caspian Region to the EU, Energy Policy J., Vol. 34,
2006, pp. 1671 1680.
15. F. Thomaidis, P. Konidari, D. Mavrakis, The wholesale natural gas market prospects in
the Energy Community Treaty countries, Operational Research An International Journal,
Vol. 8, 2008, pp. 63 75.

10

16. F. Thomaidis, N. Kontinakis, D. Mavrakis, Gas-PRS a decision support tool for

selection of optimum transcontinental natural gas pipeline routes, Euro-Asian Journal of
Sustainable Energy Development, Vol. 1, Issue 1, 2008, pp. 33 44.

11