Auteur moral

France. Service technique des remontées mécaniques et des transports guidés

Auteur secondaire

Résumé

Ce document en Anglais est un guide technique du STRMTG (Service Technique des Remontées Mécaniques et des Transports Guidés). Il présente les indicateurs de sécurité du métro et leur suivi. Il a été présenté par le STRMTG à la sixième Conférence Européenne de Recherche sur le Transport en Pologne du 18 au 21 avril 2016.

Editeur

MEDDE/DGITM/STRMTG

Descripteur Urbamet

métro ;sécurité ;transport terrestre ;indicateur

Descripteur écoplanete

Thème

Transports

Texte intégral

Available online at www.sciencedirect.com ScienceDirect Transportation Research Procedia 14 (2016) 3342 ­ 3349 6th Transport Research Arena April 18-21, 2016 Underground railways in France: follow-up of safety indicators Amélie Renard, Jérôme Charles a, * a Service Technique des Remontées Mécaniques et des Transports Guidés, 1461 rue de la piscine, 38400 Saint Martin d'Hères, France Abstract In 2003, the French Guided Public Transport Safety Decree updated the current transport legislation. It gave safety missions to the Service Technique des Remontées Mécaniques et des Transports Guidés (henceforth STRMTG), mainly in assessing and controlling the safety level of those systems. Since then, this national safety authority has collected national data on safety for each type of systems. And each system requires a particular follow-up, related to its specificities. The aim of the present paper is to concentrate mainly on underground systems, taking into account their specificities such as tunnels, automatic pilots, platform doors, etc. From those, a classification for the safety indicators was nationally established to identify the information needed by the STRMTG and the related means used to obtain it. The information needed by the STRMTG concerns different types of events with a classification distinguishing those, which have to be immediately reported to the national authority of safety from those, that have to be followed simply by statistics. For all of them, different elements are required to understand the circumstances of the events and their aftermath. Moreover, as there are only very few collective events, the STRMTG decided to ensure a follow-up of other indicators, which don't generate any real incidents but could have, in other conditions : these are precursors of incidents, that also reflect the safety level of the networks. With all this information, the STRMTG constitutes a yearly report, presenting the statistics with those safety indicators and their evolution over a few years. This statistical analysis doesn't aim at comparing networks nor at presenting a safety level classification. The aim of this is first, to have a national and anonymous view on underground events and second, to know which type of configuration is related to the biggest number of events and which aggravating factors are involved. It also gives the possibility to the STRMTG to make analyses on specific issues, to find common events between networks, and to analyse the evolution of these indicators in order to reinforce its attention on those points. This study is only made possible thanks to a number of operators, who regularly transmit all the necessary data to ensure this follow-up. Such follow-up is conducted by different means: annual reports or a national common data-base. On a daily basis, a * Corresponding author. Tel.: +33 4 76 63 78 81; fax: +33 4 76 42 39 33. E-mail address: amelie.renard@developpement-durable.gouv.fr 2352-1465 © 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of Road and Bridge Research Institute (IBDiM) doi:10.1016/j.trpro.2016.05.284 Amélie Renard and Jérôme Charles / Transportation Research Procedia 14 (2016) 3342 ­ 3349 3343 significant amount of work was carried out with those local actors and still goes on to improve the reliability of the follow-up process. After a presentation of the methodology referred to above, we will present the results of this safety indicators' follow-up over the period from 2012 to 2014, based on the data from and on underground networks. © 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license © 2016The Authors. Published by Elsevier B.V.. (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of Road and Bridge Research Institute (IBDiM). Peer-review under responsibility of Road and Bridge Research Institute (IBDiM) Keywords: Underground; safety; indicators 1. Introduction Metro networks (underground railways) in France represent 32 lines, including 2 of the regional express network (RER), within 6 large cities. Several new lines are also in project. As a national safety authority, The Technical Department of the Ski lifts and the public Guided Transport Service Technique des Remontées Mécaniques et des Transports Guidés (STRMTG), which reports to the Ministry of Transport, ensures the mission of safety control of its systems. This mission contains various involvements on each of the metro networks, including safety audits, follow-up of the operation and events which impact on safety. The safety events, based on strict selected criteria, are subject to an individual follow-up by the STRMTG on the basis of a report transmitted by the operator. In order to have an annual follow-up of these events on the scale of the whole metro networks, relevant safety indicators were set up at national level. Independently of the individual follow-up of these events, the aim of safety indicators is to follow in a multi-annual way the evolutions of the accidents and the trends according to an adapted and invariant classification. This led to the implementation of data collection methods and tools, agreed with all the rail operators. All of these data are then subject to an analysis led by the STRMTG, which compiles all the analyses in a report and publishes them every year. This STRMTG. s report is, of course, sent to the metro operators; moreover, it is public and available online, on its website. 2. Data collection methodology In 2012, in correlation with a database " metro/RER events ", the guide " Processing events relating to metro / RER safety " was drawn up to provide the classification of the operators events and the other indicators relating to the safety of metro networks, followed by the STRMTG (repeated events, critical events). This guide is also available online on its website. 2.1. Safety events & indicators The classification described in this guide shows three event categories and indicators followed-up by the STRMTG, according to the particular attention and seriousness of the event (classified in descending order). First category events: The first category events, which are listed below, must be seized in an individual way in the database metro/RER events , according to their gravity and their occurrence. The number of probable victims and the seriousness of events must be transmitted; these events are followed-up by the STRMTG. 3344 Amélie Renard and Jérôme Charles / Transportation Research Procedia 14 (2016) 3342 ­ 3349 Table 1. First category events. 1. 2. 3. 4. 5. 6. Smoke emission in a train or in a tunnel Derailments Collision Collision with obstacles System affected by the external environment Single person accident, including : falling from the tracks to the platform, falling into trains, falling between a train and a platform, collision by a moving train, electrocution, electrical hazard Various, including : evacuation in inter-stops / inter-train stations that went wrong, railway rolling stock defect, fixe installations defect, any other kind of incident related to the safety 7. The guide specifies the description of every type of event as well as the criteria of the causes and consequences combined, which allows identifying, at best, the classification of the concerned events, as showed in the example below, illustrating the event "Smoke emissions in a train or in a tunnel". Table 2. Excerpt from the Guide "Processing events relating to metro networks / RER safety", for the description of the event "Smoke emission in a train or a tunnel". N° Events Description Cause criteria Consequence criteria Examples - Intervention of emergency Ignition cables, wastes on fire on the track, emission and rescue services of smoke coming from the - limited impact rolling-stock... - Extinguishment by the rescue services and at least one of the below criteria : 1 Smoke emission in a train or a tunnel 1.2 Fire, with large amount of smoke - substantial damage to the installations or to the rolling-stock ; - smoke-filled underground galleries and/or stations/ train stations enforcing an evacuation ; - operational disruption of one hour or more Ignition or fire with a large amount of fumes invading one or many stations / train (e.g. : tires) 1.1 Smoke emissions - Second category events: The second category events, listed below, are of lesser importance. They must be transferred to the state departments, as annual statistics, through the operating report which is yearly established by the operator. The number of potential victims (only slightly affected a priori) involved in the events, must also be provided. Amélie Renard and Jérôme Charles / Transportation Research Procedia 14 (2016) 3342 ­ 3349 3345 Table 3. Second category events. 1. 2. 3. 4. 5. 6. Minor smoke emissions Inter-station / inter- train station evacuations Metro-travelers falling in the train Collision/ entrapment in train doors or in platform doors Other kind of falls outside of trains Irruption Monitoring indicators of the third category events: Finally, considering the low number of events affecting the safety of these systems observed on metro networks, indicator monitoring systems were set up. They must as well be transferred through the operator s annual report. Table 4. Indicators of the third category events. 1. 2. 3. 4. Passing at a closed signal Excessive speed Obstruction on the tracks Switch back to manual driving As stated in the guide, some of these indicators are specific to some systems. For example, excessive speed must only be indicated for specific metro networks, for instance, when nominal mode is manual driving. 2.2. Tools for the data transmission of "events and indicators" All the required data for the follow-up of the events and the indicators of safety are transmitted to the STRMTG through two tools: the database "metros / RER events" and the "operators" annual reports, already mentioned above. The database "Metros/RER events" was created just like the database "Tramway events", pre-existing, with adaptations integrating specificities appropriate to metro systems. This database introduces references to the guide mentioned previously. It is supplied by all the operators during the year N+1 allowing them to compile their own features. The database contains at least an overview of the first category events. Each event of the first category is associated to a memo containing clarifications on: the date, the hour, the place, network characteristics, the circumstances, the facts, the measures taken after the overcome of the events... Improvements were already brought to the database in the aftermath of operators' feedback. These improvements led to precisely identify the events, regulate the practices of the various operators concerning the usage of the database, and to improve its reliability. Since 2003, due to a regulatory obligation, the "operators' " annual reports are passed on to the STRMTG. These reports include elements of the operators (outstanding events, major modifications), the safety figures (indicators) and the accident record detailed analysis (analyses by type of event, and suggested measures). Thus they aim to complete the existing data in the database for the two other aforesaid categories. Among the events, some are subject to immediate information of the Prefect, the department's safety authority, and the STRMTG, without waiting for the transmission of annual reports or for the resource of the database. The criteria concern the seriousness of the events: events causing at least one death or a severely wounded person, fires or important releases of smoke, accidents with important material damages, derailments, collisions; as well as the real or potential newsworthy aspect. The events, according to their types, must therefore be transmitted to the STRMTG, according to various methods summarized below: 3346 Amélie Renard and Jérôme Charles / Transportation Research Procedia 14 (2016) 3342 ­ 3349 Table 5. Table illustrating the different means of data transmission according to the type of events. Immediately Category 1 events Category 2 events Category 3 indicators X (potentially, according to the seriousness and the type of the event) Yearly databases X Annual reports X (potentially, according to the media range) X X 2.3. Collection and data reliability The completed, annual reports and databases, for year N must be sent to the STRMTG before June of the year N+1. This allows the STRMTG to have a sufficient deadline to analyze and to publish its synthesis report before the end of this year N+1. The reliability of analyses and data are based on the rigor and the precision of the operators in data collection accuracy. For that purpose, the guide "Processing events relating to metro / RER safety" was developed by a working group directed by the STRMTG, gathering all the metro operators, involving them to elaborate a classification of the events. The operators were given a deeper understanding of the approach, specifically by the implementation of a tool which can be of use to them for a follow-up of these indicators, both on a national and internal (to their network) level. However, the STRMTG proceeds to data verification on both of the tools (BDD and the annual report) as regards the events that must be immediately stated to the control services. These statements can thus be compared with the filling of the database "Metro / RER events and of the annual report operator . Furthermore, the STRMTG takes the right to bring corrections, if necessary, to the database in order to insure the data reliability. Let us point out that given the recentness of the database, the hindsight on the evolution of safety was not very significant. For periods before 2012, statistics were retroactively completed on certain events, presented in annual reports going back up until 2003, thereby smoothing the observed trends. 3. Results of the safety indicators' follow-up Through these tools, these events and indicators of safety can be subject to a follow-up in time, thus, they are representative of the safety level of these systems. These observed trends can be more clarified in certain cases by the features of these systems or by the circumstances of the event, which allows having deeper analysis of accident rates. In fact, certain characteristics (drivers' presence, platform doors, type of facility, and length of the tunnel in interstation...) or certain circumstances of the event (train in simple or multiple unity), can allow to identify the aggravating factors of the system's safety. 3.1. Example of results This section presents various examples of analysis which are possible to conduct with this approach of follow-up indicators. The first example consists in having a national vision of the first category events that have taken place on metro networks between 2007 and 2013. Amélie Renard and Jérôme Charles / Transportation Research Procedia 14 (2016) 3342 ­ 3349 3347 Fig. 1. Excerpt from "annual report on the networks, the traffic and the events of metro operation events taking place in the metro networks by classification. showing the distribution of the number of Analysis: the individual events are mainly collision / entrapment in the doors of the train or in the platform doors, falls in trains, falls on the railway, falls between trains and platforms. Let us notice that the completely automated systems and provided with platform doors remain exempt from these first two types of cited events. Between 2007 and 2013, the distribution of the events remains generally stable. Only the collision/entrapment into doors decreases in a continuous way in time. The installation of platform doors on the line 1 and 13 discouraging the users to "get in late or to get off late" into the train could be an explanation to this. The second example consists of a sharper analysis of a precise event, in particular the collision/entrapment in the doors of the train or the platform doors. Indeed, this analysis distinguished the events that have taken place on networks having platform doors from those not having them. 3348 Amélie Renard and Jérôme Charles / Transportation Research Procedia 14 (2016) 3342 ­ 3349 Fig. 2. Excerpt from "annual report on the networks, the traffic and the events of metro operation" illustrating the collision/entrapment with or without platform doors, by number of millions of travels. Analysis: whereas the collision / entrapment on lines without platforms doors remain globally stable these last years, these events are down substantially on lines having platform doors, which are almost exclusively automatic lines. These two examples illustrate the use of these indicators to follow the level of safety of all the metro networks by trying to explain these trends. 3.2. Use of these data The use of these safety indicators allowed justifying the need for implementation of the database. Indeed, this approach was first of all seen as a constraint, for some operators already having tools to follow the safety system level. However, this method showed its advantages: set up means of follow-up computer system allowing the use of data to those who were not endowed with it and to harmonize the practices between the various operators, by sharing the same safety indicators. Thus, today all the operators use or inquire the same tool as the STRMTG which they validated with the latter, and which helps them better monitor the safety of their systems. This tool also allows the national authority of safety to achieve statistics on the safety of metro networks at national level. Therefore, the operators can relate to each other by comparing data of their own network with national level data formalized by the STRMTG in the "annual report on the networks, the traffic and the events of metro operation". Indeed, the annual report published by the STRMTG presents, in a fully anonymized format, the statistics in terms of accident rates of all the networks, by putting these statistics in perspective with those of the previous years. The annual reports of the STRMTG are also presented and discussed within an authority, gathering all the operators of the working group "Rex-Metro/RER". It is also the most adapted place to share feedback experience of the usage of the database and to make continuous improvement therefore. It is possible to work on a specific accident rate issue, for example, when it is impacting on the safety of the systems and when the network features are not enough to explain it with simple data. Amélie Renard and Jérôme Charles / Transportation Research Procedia 14 (2016) 3342 ­ 3349 3349 4. Conclusion The STRMTG's metro networks safety control mission is being strengthened by this follow-up of indicators at national level, boosted by the database and the guide of the "Processing Events relating to metro networks / RER safety" which are in operation since 2012. These allow an overall analysis, yet a sharp one, of accident rates of these networks, supplemented by elements from the operators' annual reports. The success of this approach originates substantially from the participation, the understanding and from the commitment of the profession's operators, as a whole, in the implementation of this follow-up of indicators. Indeed, the necessary data can only be transmitted by the operators, who were convinced to set up this follow-up. Thus, National level data are now analyzed and published by the STRMTG, which reports it in in a working group appointed "Rex-Metro / RER". In 2012, the implementation of these indicators was a first step for a better follow-up of the metro networks' safety level, homogeneously, within the follow-up of the tramway systems. This first step required a phase of data reliability; nevertheless, in its continuous improvement approach and in a multiannual hindsight, the safety indicators could certainly be improved. References STRMTG, 2012, "Métros et RER (hors RFN), Traitement des événements intéressant la sécurité" STRMTG, 2015, "Rapport annuel sur le parc, le trafic et les événements d'exploitation des métros et du RER (hors RFN) 2013"