Javier Gutierrez Puebla, Rafael E. Gonzalez Aguayo


TRANSPORT IN EUROPE:
A Study of Train Accesibility Using GIS

Abstract

This paper reports the design, implementation and results of a trans-european scale transportation accesibility model. This study report only the results of the railways portion of the model, which includes multiple categories of lines within the transportation hierarchy.

The rail network was digitized over the road network and the design considers not only the railways but also the possibility of taking another means of transport in order to access the train. Implementation of the model used indicators of 'absolute accesibility' similar to those reported in Gutierrez et al. (EGIS'93), except for important differences resulting from enhanced network analysis capabilities in the most recent version of the GIS software. Analysis includes construction of isoaccesibility contours zones showing net increase in accesibility and differences maps.

Introduction

The antiquity of the conventional railways in relation to the continuous advance of the rest of the means of transport and the necessity of complementary means that imply annoying breaking of load zones, have forced to develope an increasing investigation effort to transform the train in a more competitive mean. At dawn of the Second World War the concept of the High Speed Train arises as a possible solution to this problem. The High Speed in Europe doesn't appear until September 1981, with the partial opening of the new Paris-Lyon railway -- though this wasn't the first one who began to be built. This railway was a big success attracting even some of the trips that until then had been made by plane or road, and generating a larger demand besides. In answer to this kind of iniciatives, some countries of the European Community continued the developement of High Speed railways paying attention to their different national needs, but following the same common objective:

"give the train a different offert from the actual one, more competitive, capable of winning a bigger share of the travel market, taking place as an alternative to the airplain and the road".

Seeing the interest of each of the countries of the European Community and facing the possibility of each country designing the railway as he pleases,and considering only his own necesities at the margin of the european view, the European Community decided to stop this tendency proposing the creation of an European High Speed Network, and working out a guiding plan of it (Guiding Plan of the European High Speed Network). This plan extends till 2010 (see map 1) and includes the following railways:

- 9.000 Km. of new railways, capable of carrying
- 250 Km/h or higher velocities.
- 15.000 Km. of improved lines, prepared for velocities of arround 200 Km/h.
- 1.200 Km. of connection (or framework) railways.

If the foresights of the previous table are fulfilled, Spain will have a total of 46 kilometers of new created railways for each million of inhabitants, more than twice the corresponding ones of Germany and Italy, being the second after France. If we consider the number of network kilometers in relation with the number of square quilometers of territory, then Spain will be in the fourth place in spite of having a lesser density of population. The High Speed train is thought of as an integrator element:

"The High Speed train, as it reduces the duration of the travel it also reduces the distances. This is the reason of it being a powerful instrument of arrangement of the territory. The European network will have an organizing effect on the communitary space and it will promote the regional developement and the relation between regions"
Besides the high speed train is a better option for carrying out the environmental criteria and it has a lesser energy consumption. Nevertheless not all are advantages:

"The building of new infrastructures has an important environmental cost and implies a very high investment. On the other hand, the territorial effects of these infrastructures is very punctual because only the main cities have train stations which has the aditional danger of increasing the territorial umbalance already existent."

Other of the more bothersome problems is the possible incompatibility of the different High Speed networks. It's not completely clear if a direct connection among the sections of the different countries can be achieved, being this the final purpose of the Transeuropean network.

The role of G.I.S.

The Geographic Information Systems provide a new analysis tool which is specially appropriate to support the accesibility studies because they allow:
The G.I.S. is a useful tool for planning and managing graphic and alphanumeric information, their utility has being probed in lots of works and studies. The current presentation is a part of a more ambitious project dedicated to the study of the transeuropean networks of transport, and this communication is specifically centered in the railway transport. With this objective a georeference data base has been created which includes:
This study has two horizonts, corresponding to the years 1992-93, and to the planned situation for the year 2010. The previously mentioned spatial data base is formed by covers in ArcInfo format (revision 6.1.1), because this was the elected G.I.S. for the project. The design considers not only the railways but also the possibility of taking another means of transport and as a result there are some cases in which the accesibility by train depends, apart from the train network of the European Community, on the use of roads or ferry boats too. The system has been able to calculate times of access from durations of travels assigned to the arcs as impedances and, in some cases, to the nodes too (mainly delays for changing of means of transport to other different from the train which results in a penalization over the whole time of the travel). The attributes asociated to the train arcs of the data base are:

COUNTRY:

Represented by the initials of the country of the European Community that is crossed by the arc.

FROMNODE:

Origin location of the arc, in the language of the corresponding country.

TONODE:

Destiny location of the arc, in the language of the corresponding country.

TYPE:

Kind of infrastructure, distinguishing among conventional train, improbed train and high speed train.

DISTANCE:

Real distance represented by the arc (in Km. of network).

TIME:

Time it takes to travel along the arc.

The network of the year 2010 shows the foresighted final situation after the actions fixed in the Guiding Plan of the European High Speed Network had taken place. This network has been obtained starting from the 1992-93 digitized network and adding the new railways being built and registering in the data base the improvements of the already existing ones.

INDICATOR USED

The accessibility indicators try to "give a measurement of the separation of activities or human settlements that are connected among them by a transport system". It is important to take into account that all accessibility index must have some necessary elements: "the proximity measurement or separation between two or more points; the relation to a transport system that allows to cover the distance between them; the "effort" (time, cost...) needed to cover it; and the relation with some type of activity in which the user wants to take part". Taking this definition as a reference point, it was decided to establish an indicator to undertake this study. The indicator formulation is based on calculating the average impedance of each node in relation to the diferent economic activities centers through the network (using the minimum cost path), taking as a weight factor the Gross Domestic Product (GDP):



Where:
Ai is the accessibility of the node i.
Iij is the real impedance between nodes i and j.
GDPj is GDP of the destination economic activity center.

So the easiness to access an economic activity center is a function of the resistence to move between the origin and destination nodes. This resistence is expressed in impedance units, and the accessibility value to reach each node is the average distance to the center, taking as weight factor the importance of the destination economic activity center. Therefore, the accessibility levels (derivated from the indicator used) show the influence of the geographic siting of the nodes and the infrastrutures influence. The measurement of the impedance between origin and destination nodes is just the addition of the arcs impedances covered to get from the first to the second node using the minimum path. But besides, in the case of train it can happen that the origin point of a travel is not a primary node (a node with a train station connected to the network), which implies that to calculate the impedance it is necessary to add an extra cost for the complementary conexion to the road network. In conclusion, the easiness to access depends on the following factors:

  1. Easiness of access by road until the nearest node with train station.
  2. Delay associated to the change of mode from road to train.
  3. Easiness of access by train from the node of connection to the network to the destination economic activity center.

Therefore, the real impedance calculation results from:


I= Ir + Imc + In


Where:
I is the real impedance.
Ir is the impedance by road from the origin node to the nearest primary node.
Imc is the impedance for changing of mode.
In is the network impedance between the origin and destination node.


The impedances of the railway arcs have been calculated considering for each of them a cost equal to the time spent by the faster existing train to cover this arc, according to the 1993 Thomas Cook´s European TimeTable data. At the final situation, the average speed for High Speed is 275 Km/h, and for enhanced sections 200 Km/h have been considered. The node impedances assigned by gauge change between Spain and Portugal are estimated in 30 minutes. The impedances for crossing over big cities are not considered bearing in mind the scale used in this job. The impedances associated to functional transfers aren't considered too because this study is only dedicated to observe the intrastructures influence.

RESULTS

One of the goals of this study is to obtain an accessibility zones map for the two planned horizons. The Geographic Information System ArcInfo was used to achieve this end following the next proccess: For the real impedance calculation (I) it is needed to know the distance between every origin node and destination economic activity center accross the network. To obtain it the INTERACTION ArcInfo command was used. This command calculates the spatial interaction rate between two nodes of a network, using the origin and destination nodes as interaction centers. With this intention the network was provided with arc and node topology (AAT and NAT) coding these according to what has been previously said. For the calculation of interactions it was necesary to design a centers table with the nodes for which it is needed to measure the distances. As the process was applied to more than 4000 nodes and 94 economic activities centers, it was necessary to develop an AML program with two loops changing the origin and destination nodes of this table automatically until finishing the total number of trips; resulting in a table with more than 368.000 registers, with all possible connections and the interaction rate an time cost associated. The centers file contained the following items:


ARC#

(4,5,binary)

COV#

(4,5,binary)

COV-ID

(4,5,binary)

CITIES

(1,1,integer)

The indicators for the 94 economic activities centers used as reference were calculated using the results obtained from getting the interactions and adding the impedances of mode changing and road use. The NODEPOINT command was used to convert these centers into points which allows to manage them separately. and enabled their use to generate a Digital Elevation Model (DEM), using the points coordinates as coordinates and the indicator obtained for each of them as the Z value. With the TIN resulting from this process properly checked, the model was converted into a Lattice to proceed later to define the ranges that will allow to obtain the isoaccessibility zones. Therefore, ranges in the database were defined, transforming again the final result into an isozones cover with polygon topology. As this cover had triangulation in undesired zones, it was cut away (CLIP ArcInfo command) using the CE countries' border. This proccess was realized for both the periods considered in this study (1992-93 and 2010).

REMARKS TO THE RESULTS

As the object of this work is to show the process and not to comment the final results, these remarks are going to be very concise: