Quantitative Risk Assessment Model for Dangerous Goods Transport through Road Tunnels

A Quantitative Risk Assessment Model (QRAM) to evaluate the risks of dangerous goods transport through road tunnels was jointly developed by PIARC and OECD as part of the ERS2 project (1997-2001).

Since then, the software has been applied to a wide range of tunnels in several countries.


What does the QRA model do and not do?

QRAM - World Road Association PIARC

The software allows users to:

  • assess the societal risks due to dangerous goods transport for a tunnel or a route;
  • compare the societal risk of a tunnel or a route with reference criteria
  • assess the societal risks for each ADR category[1] (A to E) of a given tunnel in order to help choose one. The societal risk of categories A and E corresponds respectively to the societal risk of the tunnel route and of one (or several) alternative transport route(s)

The software can be used to perform a specific risk analysis for dangerous goods transport. This specific risk analysis can be part of the general risk analysis required by the European Directive 2004/54/EC on minimum safety requirements for tunnels on the trans-European road network.

The algorithms, procedures and computer programs of DG-QRAM were developed only for the assessment of risks due to road transportation of dangerous goods through given routes, especially with tunnels. They have been compiled from the best knowledge and understanding available at the initial development date (year 1998), but have important limitations that must be understood and considered by the user. The software package constitutes the QRA-model. For assessment of smoke propagation/ventilation, evacuation etc., there are other, more appropriate tools including 1D and 3D fluid dynamic models. Consequently, users are warned that whilst the QRA model may be used for dangerous goods risk assessment, it must not be the only tool used in the assessment of the acceptability of design proposals to achieve the required global tunnel safety levels.

[1] ADR : "Accord for dangerous goods by road" : European Agreement concerning the International Carriage of Dangerous Goods by Road


Outputs

  • Contour maps of individual risks along tunnels and open routes
  • Societal risks presented as FN-curves, where F is the frequency of N or more fatalities (and/or injuries)

Presentation of the QRA model

The QRAM software is based on Microsoft Excel.

There is a wide range of information that has to be gathered and prepared in order to run the QRAM. Notably, the user has to collect and interpret information and data to describe:

  • routes, defined in sections;
  • tunnel geometry (length, cross-section and vertical alignment), ventilation (specialist help may be required for complex tunnels), drainage, and emergency escape measures (warning systems, spacing of exits);
  • traffic characteristics, including vehicle mix and traffic speeds, defined for each route section and traffic direction;
  • relevant DG transport data (amount, composition of dangerous goods)
  • incident rates
  • population along the route.

For some data the model provides default values, which should only be used on the basis of careful expert judgment (if no specific data is available). Input screens help to guide the user through the process of entering this data and running cases.

The QRAM considers 13 DG incident scenarios which are representative of key dangerous goods groupings.

  1. Heavy Goods vehicle fire with no dangerous goods (20 MW) (back journey after delivering dangerous good)
  2. Vehicle carrying DG from class 4.1 and class 9 (100 MW)
  3. Boiling Liquid Expanding Vapour Explosion (BLEVE) of Liquid Petroleum Gas (LPG) in cylinders
  4. Pool fire of motor spirit in bulk
  5. Vapour Cloud Explosion (VCE) of motor spirit in bulk
  6. Release of chlorine in bulk
  7. BLEVE of LPG in bulk
  8. VCE of LPG in bulk
  9. Torch fire of LPG in bulk
  10. Release of ammonia in bulk
  11. Release of acrolein in bulk
  12. Release of acrolein in cylinders
  13. BLEVE of carbon dioxide in bulk (not including toxic effects)


The QRAM takes into account:

  • accident frequencies (derived from historical datasets);
  • physical consequences of incidents within tunnel(s) and along open routes;
  • escape and shelter effects;
  • effects of hazards (such as heat and smoke) on people.

The results for different routes and traffic are calculated in terms of Societal Risk. The Societal Risk reflects the whole range of possible outcomes of all possible DG incident scenarios for a group of people. There might be a low chance of injuring most of the people in a town, or a higher chance of injuring just a few of them.

This relationship is illustrated by an 'F/N curve', where F is the frequency of N or more fatalities (and/or injuries). An example set of F/N curves is shown here. Each curve can also be evaluated in terms of a single value representing the average number of fatalities per year, called the Expected Value (EV).

The F-N curves can be produced for fatalities and/or injuries, and for road users and/or the local population.

Users support and training

No support is currently available.

Training sessions are scheduled for the 16th, 17th and 18th of December. For the training sessions, please find relevant information here!

This site uses cookies to optimize its operation. They allow you to securely access your personal space and download our publications. You accept their use by clicking on the "Continue" button.