Using gravity model in regional rail transport planning

Transcription

1 Using gravity model in regional rail transport planning VÍT JANOŠ Department of Logistics and Management of Transport Czech Technical University in Prague, Faculty of Transportation Sciences Konviktská 20, Praha 1 CZECH REPUBLIC janos@fd.cvut.cz Abstract: The importance of transport planning in public services gradually becomes fully appreciated in the context of the Czech Republic. Public transport - as well as investments and operationally intensive industries - requires a guaranteed long-term strategy, not only in the field of transport infrastructure but also in the field of transport services. The first prerequisite for successful transport planning is the most accurate description of passenger flows, including the parameters affecting the choice of transport mode. These inputs constitute an important basis for the following steps: setting up the line network, timetable design, creating circulation of vehicles and courses of staff, controlling, and any eventual requests for infrastructure improvements. On the example of one of the Czech regions responsible for public transport order, Ústecký kraj, usage of a gravity model is specified to determine the significance of proportional inter - town transport relations within the region. Practical application of the results of this gravity model is stated further transformation of the timetable concept on the selected rail relation, including effects in the change of the number of passengers. Key-Words: transport planning, public service, passanger transport, railway timetable, gravity model, timetable planning 1 Defining the objectives of transport planning Currently, the transport planning in the Czech Republic stands in its factual beginnings. Institutionally, transport planning is established by the "Law on public passenger transport services" (194/2010 Coll.) and individual orderers of public transport are thus required to build transport plans for its scope. Public transport, in the concept of quality of transport service, is more than just a social service for those who have no other option but its aim is to achieve an overall transport accessibility of the region, so that public transport represents an attractive alternative to individual transport. It is clear that public transport can not cover all the passenger flows, not even all the transport needs of individuals. Public transport with regard to the need for compensation (income from fares does not cover the cost of providing) must include elements of the mass. If there are common transportation needs of individuals in a particular place and time, the transport flow is formed, which can be then addressed by public transport organized on a collective basis. With regard to the rational planning of public transport and the final volume of funds for compensation payments, these resources cannot be spent inefficiently on transportation of single individuals, not having then the resources to ensure satisfactory transport on relations with much higher demand. Theoretical public transport planning consists of 7 classic steps: 1) Estimation of passenger flows, part of transport modelling 2) Setting up the line network plans 3) Implementation line network to the infrastructure, modal choice revision 4) Timetable design 5) Circulation plans of vehicles and stuff courses ISBN:

2 6) Evaluation of the operation performance, controlling 7) Defining the requirements for the infrastructure improvements In this article we will pay attention only to the first point. Any information regarding the transport relations and passanger is important for line network planning and timetable improvements. Outputs of gravity model introducing the core and collecting lines with high (not yet used) potential will be gradually applied to the ongoing optimization of single areas of the region. On the example of Ústecký kraj (one of the 14 regions in the Czech Republic responsible for regional public transport order) will be presented, how the simple principles of well known gravity model were used in the public transport planning and what practical result was achieved. 2 Conditions for survey of transport relationships in the Czech Republic The issue of identifying the transport relations is in a very difficult situation in the Czech Republic. In most cases when it is necessary to define transport relations using traffic modeling, you can not use the classic four-stage transport model or EVA model based on principles of transport demand and supply. The classic intelligent transport modelling methods in most cases lack the necessary data bases, the available data in many cases lack the temporal relationship to the "same time". Many data important for the transport demand (level of population, number of jobs, educational facilities etc.) usually collected for each zone are simply not available in the required quality and details. Although orderers of public transport (public authorities) are responsible for transportation planning in their territory, creating high-quality transport model would not pose a major priority and usually are not even willing to finance the necessary surveys to obtain data bases, bases of populationmobility and travel behavior. In the event that models based on classical four-stage transport model are applied in the Czech Republic, very rough data or generally available external data (e.g. from Germany) are often used. In view of the foregoing, it is necessary to set priorities in public transport very easily - this tool presents gravity models using the basic data that are generally available in the Czech Republic. The information of passenger flows is currently available mainly from the CSO data (Census 2001, 2011) - data on the number of permanent residents in the municipalities and their local areas, information about regular commuting. Further information is available from carriers - information about the load-factor, number of boarding and alighting passengers at each stops, shortly will be provided the origin-destination matrices. More information on the origins and destinations relate mainly to number of employment opportunities (data from major employers) and regular commuting school children - these data are used in abundance in the case of larger optimization of timetables in the region. Available CSO data are not broken down by mode of transport attributable to a specific origindestination, because this is the information about the total passenger flows - across all modes of transport. On the contrary, data from carriers are related only to a particular type of service and thus have explanatory power only for passengers who already use the public transport. For a description of all passenger flows in the region across all transport modes can only be used the theoretical transport models. Thus established passenger flows can be followed on the basis of the availability of origins and destinations of share assigned to each mode of transport - theoretically calculated passenger flows thus assign theoretical modal split. It shows how important is the role of public transport in a whole transport system where there is generally a high demand for transport and the passenger flows which makes it suitable to be targeted and at the same time, in which relations there is a most unused potential for public transport. ISBN:

3 Today, no region has yet processed quality transport model, because the benefits of highquality and comprehensive transport modelling in the Czech area are still not fully appreciated. What is missing, is the quality travel surveys, surveys on the transport behaviour and sensitivity of users to parameters of public transport offer. Finally, it is difficult, often impossible, to group the time incommensurable data needed for the transport model setting. For the purpose of transport planning, incremental theoretical gravity models of passenger flows in the region are therefore often used. The output of these models is the proportional comparison of significance of passenger flows, often associated with the theoretical calculation of the modal split. From the gravity model taking into account the modal split can be seen, how the public transport is successful in individual relations. Clearly then follows the disproportion in the region and the relations that represent the greatest potential for growth in public transport. Juxtaposition of relations can also be inferred where is potential for core railway lines, collecting railway lines and core bus services. 3 Using the gravity model and application of outputs In 2011 author of this article created an extensive gravity model of the region Ústecký kraj, based on the known "Lill s travel law": v v ij k Q Z i j ij = k (1) wij Q i Z j w ij where: passanger flow between i-th a j-th place gravity constant source / origin potential of i-th place destination potential of j-th place travel resistence The model is based on the assumption, that the transport relationship between settlements (Q i a Z j ) increases with their size and decreases with transport resistance (usually time or geographic distance). In the case of proportional comparison may be omitted constant k, however, in this solution was territorial, regional and district administrative hierarchy reflected by constants from 0.7 to 0.9. Travel resistance w ij in this case was determined as the square of travel time. Source and destination (Q i, Z j ) in this case was determined as the square root of the number of inhabitants of the cities (by the square root can be taken into account a higher level of attractiveness / emitivity of individual cities). Travel matrix was assumed to be symmetrical. The calculation results are then only numeric values, which after conversion reflect the "relative size" of traffic flow. The aim of developing the gravity model was creation of the proportional comparison significance of individual transport relations in the region, according to the relative significance. The aim was to find out, in which relations is generally highest theoretical travel potential, and what public transport offer is today provided in relations with "similar" significance. Furthermore, the relative comparison was made of the availability with public / private transport, and ultimately defined relationship growth potential for public transport, and endangered relations (where is urgently needed to carry out a fundamental change in the operational concept of public transport). Within the configuration of the model all the towns in the region Ústecký kraj with more than 5,000 inhabitants (27 towns) were taken into account, including also the usual destinations in regular commuting to neighboring regions and the Capital city of Prague (13 additional targets). The basis for the model processing were the CSO-data (Census 2001), travel times comparison was based on then current timetables (2010/11) and publicly available Route planner (in relation center - center). Taken into account was that the travel time by public transport with connections showing regularity, or "usual" travel time. For reaching a public transport station was considered generally a ISBN:

4 surcharge of 20 min. Middle waiting time for a service / connection has been neglected. During assembly of the model were categorized relations showing daily commute from sessions without daily commute. Interval connection was initially ignored, as it applies premise, that in daily commute it is hardly acceptable that the interval of the connections in public transport will significantly exceed the regular travel time. For determining modal-split were used outputs from research and development project "Configuration the IFIT-nodes in the Czech rail network" carried out during by the Czech Technical University, commissioned by the Czech Ministry of Transport. Fig. 3 Cut from the matrix of the theoretically achievable Modal-Split The last step was to calculate the proportional transport relations, that were weighted by theoretically achievable modal-split, what defined relations with the greatest potential for public transport development and at the same time it was clear, where the offer of the public service is already close to the theoretical saturation. Fig. 3 Cut from the matrix of the theoretically achievable Modal-Split weighted by travel time ratio Fig. 1 Modal-Split of public transport on the transport market By processing the outputs of a gravity model were first compared the travel times by public transport and by individual transport. Fig. 2 Cut from the time-availability matrix On the basis of the travel-time comparison was then calculated theoretically achievable modal-split for public transport in individual relations. Based on the outputs of the gravity model (among other outputs) relations with a similar "travel-importance" were compared and their offer of transport connections was evaluated. For functional testing of the hypothese of "theoretical similarities" of transport relations, regional rail relation Litoměřice - Ústí nad Labem was selected in cooperation with the region Ústecký kraj. At this relation it was calculated by the gravity model very similar theoretical significance of transport relation to another regional relation Roudnice nad Labem - Lovosice - Ústí nad Labem. In both cases, those are the relations with intensive daily commute, where the main destination is the city of Ústí nad Labem. In areas with a daily commute the morning peak hours transport connections offer (which is shorter and sharper than the afternoon rush hours) is decisive. While in relation Roudnice nad Labem - Lovosice - Ústí nad Labem in the morning peak hours (about 5:30 to 8:00) 30-minute interval of connections was already offered at the time of the ISBN:

5 modelling (2011), in the relation Litoměřice - Ústí nad Labem only 4 connections kept in irregular intervals were offered at the same time. For both compared relations is given, that in the transport peak in the early morning only regional rail connection is offered (in the morning rush hours neither long-distance trains, nor connection by bus operate in these relations). Passengers in these relations can travel only by regional rail or individually. There was performed a change in the timetable and operation concept in 2012 in the relation Litoměřice - Ústí nad Labem, so that 30-minute interval connection was introduced in the morning peak hours (creating an identical offer as on relation Roudnice nad Labem - Lovosice - Ústí nad Labem with identical teoretical significance). In 2013 the change in the number of passengers on the relation Litoměřice - Ústí nad Labem was evaluated, namely on the basis of autumn campaign census on rail, and comparisons for Table 1 Development of the number of passengers between Litoměřice - Ústí n.l. Between 2010 and 2011 original operational concept functioned, while in 2012 and 2013 the timetable was reorganized according to the new extended concept. In view of the fact, that the results of the census are not public, the number of passangers was relativized, with a default considered year 2010 (100% of passengers). Due to the introduction of the new offer of connection the increase of passengers by 8% occurred after 2 years in relation Litoměřice - Ústí nad Labem. For completeness, it is necessary to add, that throughout the condition "ceteris paribus" was followed, ie that there were no major changes which could affect the demand for transport - except changes in the operational concept in the railway timetable. There is generally verified in the Czech Republic, that the total response time for passengers (settling their number) occurs about three years after the concept change of the timetable. At this point, however, only data for two years of the new operation (2012 and 2013) can be evaluated, and is thus premature to draw conclusions about the elasticity of demand of passengers related to interval / frequency connection in regional traffic. 4 Conclusion Although the use of gravity models in operational concept planning of public transport represents only a very limited support tool, on the case of regional rail services in the Region of Ústecký kraj it has been demonstrated, that such tool can be used to detect the potential of development of public transport and can contribute to the improvement of the offer in the relation, where this improvement has tangible benefits. For the orderers of public transport the consciousness is very important, in which areas there is a potential of growth for public transport. Proportional comparison of the significance of single transport relations can contribute even to the elimination of historically originated incommensurability. In addition, the outputs of the gravity model are utilized in optimization of public transport projects, in the construction of backbone lines, even by modal optimization and in planning the necessary capacities. Practical application in the region Ústecký kraj with clearly positive result on regional rail relation Litoměřice - Ústí nad Labem, contributed to the fact, that the Region Ústecký kraj will further continue in modifying and developing operational concepts in regional transport in such relations, where the treated gravity model shown significant potential for increasing the modal-split of public transport. ISBN:

6 References: [1] Janoš, V. - Baudyš, K.: Transport Planning of Public Services. In Proceedings of the 11th European Transport Congress. Praha: České vysoké učení technické v Praze, Fakulta dopravní, 2013,. ISBN [2] Hesse, W. - Baudyš, K. - Janoš, V. - Pospíšil, J. - Guckert, M.: Project of nationwide Integrated Periodic Timetable in Czech Republic. Eisenbahn-Revue International. 2010, Jahr. 2010, nr. 11, S ISSN [3] Baudyš, K. - Janoš, V. - Polák, O. - Walla, L.: Hierarchic Planing in Regional Traffic. In Telematika pro regionální dopravu Brno: KPM Consult, 2008, s ISBN [4] Baudyš, K. - Janoš, V.: Relation between Modal-Split and transport planning. In Udržitelná příměstská doprava. Praha: ABF a.s., 2007 [5] Strategic Plan of public transport of Ústecký kraj: dokumenty2.asp?id_org=450018&id= &p1= [6] research report: Research and development project "Configuration the IFIT-nodes in the Czech rail network"; Czech Technical University ISBN: