Has poor securing of bridges had major repercussions for traffic safety?
Have we had any traffic fatalities due to poorly maintained bridges?
VEME (VG) After the fatal accident in 2005, investigators called for crash barriers throughout Norway to be mapped and improved. ‘What good is it making a recommendation if nothing happens?’ asks the mother of Siri Tobiassen (16).
This is a tough time for Toril Engebretsen (50). On 19 November it was twelve years since she and her husband lost their only child. Siri was just 16 years old.
Advent reminds her of all the hours spent around the kitchen table with her daughter, baking, exchanging gifts and enjoying themselves. And the day after Boxing Day is Siri’s birthday. She would have been 29 years old.
‘Siri is dead. I’ll never get her back, but we'll do everything we can to make sure no one else will have to go through the indescribable pain of losing a child,’ says the mother.
In autumn 2005, Siri was studying to be a children and youth worker at Hønefoss upper secondary school. On Saturday 19 November, Toril Engebretsen drove her daughter and a friend to a party at Tyrifjorden. When the girls got out the car, Toril said just to call if they wanted a lift home.
Just before 11.30 that evening, four young people were sitting in a BMW. At the railway bridge in Ask, events took a turn for the worst. Exactly what happened, no one knows. But the car ended up on top of the safety barrier, and the left side was crushed against the bridge.
Two teenagers survived the accident. Siri and the driver, Henning Vollan Liuhagen (18), were killed.
After the accident, the family had very little information about how it could have happened. They learned that the car had been driving at a high speed, but were never told that accident investigators from the Norwegian Public Roads Administration (NPRA) also listed several other reasons why the accident had such a disastrous outcome: a sharp turn that should have been marked, worn tyres, an inexperienced driver – and a crash barrier that was too short and had an end terminal that was incorrectly constructed.
In order to prevent similar accidents in the future, the accident investigators recommended in 2005 that Norway’s entire road network be reviewed with a view to improving and modifying similar crash barriers.
The accident at Ask is one of 33 fatal accidents where NRPA investigators have established that faults or deficiencies in crash barriers on and adjacent to bridges have contributed to the severity of accidents.
The in-depth analysis of the 33 accidents was first presented to the Road Safety Committee in the Nordic Road Association in January 2016.
It was not until 19 months later that the findings were presented to the public in an article.
For the past year, VG has been charting the damage, maintenance and traffic safety on Norway’s 16,971 national and county road bridges. In the autumn, VG’s journalists started gathering information about inspections and maintenance. That’s when the NPRA started preparing its answers, as shown in e-mails that VG has gained access to.
Has poor securing of bridges had major repercussions for traffic safety?
Have we had any traffic fatalities due to poorly maintained bridges?
Hi Kjell Bjørn
Thanks for the information
You've reminded me of something I have a very bad conscience about - the article about bridge crash barriers that Southern Region wrote based on UAG reports. We had plans to publish the article last summer/autumn, but it was put on the back boiler for a while and was then completely forgotten. Should we take it up again now?
The Directorate wants a proactive media offensive (in the course of the week) for UAG themes related to bridges, crash barriers and fatal accidents.
Will probably be picked up by NRK Morgennytt news programme during the course of the week. Being run independently of the VG situation.
Press Manager Kjell Bjørn Vinje says that the NPRA has published the in-depth analysis completely independently of VG’s bridge project, and writes in an e-mail: ‘The NPRA does not initiate a proactive media offensive ahead of negative items in the media. That’s not our practice, and that applies here too’.
Last week, VG asked Director of Traffic Safety, Guro Ranes, why it took so long to publish the article containing the in-depth analysis:
‘The presentation you are referring to is a series of overheads that was presented at a Nordic forum for traffic safety,’ says Ranes.
She adds that they wanted to present the results in a format that could be shared with more people.
‘We chose an article format – and it’s a format that we’re not very used to for presenting results. It’s taken numerous rounds and much more time than we envisaged to work out the technicalities. It also has to do with the format; it’s quite a challenge trying to write succinctly. We had to get the article down to four pages instead of a more comprehensive and ‘boring’ report, which is the format we usually present results in.’
Do you have any tips to pass on about bridges in Norway?
VG contacted Guro Ranes again after reading the e-mail dialogue where it says that the article was forgotten:
‘The reason why it took so long was two-fold. Firstly, the article format was new to us, and there were therefore more departments involved in the work than usual. Communications staff, UAGs and bridge staff were involved back and forwards. Unfortunately, we lost the thread along the way, and the work on publication stopped,’ writes Ranes.
‘This was obviously unfortunate. But when we discovered that the article had not been published, we picked up the thread again and quickly finished it. The article was then posted in the usual way – through our own channels, and through the media if we manage to do that.’
The analysis concludes that there is a need for a national survey of how many bridges have such problems.
Siri’s mother, Toril Engebretsen, says it’s shocking to hear that the article was forgotten.
‘It makes me both angry and sad. This is far too important to be shut away in a drawer for so long,’ she says.
Engebretsen has now read a copy of the in-depth analysis where her daughter’s accident is discussed. In total, 37 people have died in 33 fatal accidents.
‘It’s unbelievable. They themselves say that it has cost 37 lives. Thirty-seven! Imagine how many thousands of people are affected by these accidents. I feel that the NPRA has failed the road users,’ Toril says.
‘After the accident, they conclude that the entire road network should be reviewed, and such crash barriers should be improved. But what’s the point in making recommendations if nothing happens!’
‘It’s maybe possible to understand why not all crash barriers have been properly fixed yet?’
‘No, not really. If they can save just one life, then it must be done immediately. The NPRA needs to get an overview of the situation and make sure that politicians understand that this must be a priority,’ says Toril Engebretsen.
VG has mapped how many bridge crash barriers have faults or deficiencies that can affect traffic safety. The NPRA’s bridge register Brutus shows that 711 bridges have severe damage to the crash barriers. A total of 1085 bridges have vulnerable crash barriers, most of them failures to meet current safety requirements. In total, this concerns about 1745 individual bridges.
1745 bridges with a vulnerability
or severe damage to the crash barrier
These figures are probably incomplete. Even Roads Director Terje Moe Gustavsen has described the shortcomings in Brutus as severe.
In November, VG revealed extensive breaches of regulations as regards the registration and inspection of Norway's 17,000 national and county road bridges, and several road departments admitted that they have only used the Brutus registration system to a limited extent.
The revelations led to the Roads Director being reprimanded by the Minister of Transport and Communications, Ketil Solvik-Olsen. After the meeting, Gustavsen said that he had expressed that ‘we’re not on track with the registration’. The Roads Director has promised to sort out Brutus, and that complete figures for 2017 will be in place by 1 February next year.
The Road Supervisory Authority (RSA), whose role is to ensure that the NPRA complies with laws and regulations, has also pointed out extensive registration shortcomings in the regulatory work in Northern Region and Western Region.
Despite the deficiencies that have been uncovered in Brutus, it’s this system that the Director of Traffic Safety, Guro Ranes, refers to when VG asks if the NPRA now has a complete overview of how many bridges have faulty or deficient crash barriers.
Reports and analyses have been calling for this situation to be mapped for more than ten years.
‘We’ve discussed and reviewed the situation with the section for bridges at the NPRA. Our assessment is that the registration of crash barriers and vulnerabilities in Brutus provides a good overview of the problem and scope,’ Ranes said last week.
She adds that she is confident that those who manage the road network have a good overview – and that shortcomings will be followed up and reported to the appropriate parties.
An internal e-mail that VG has gained access to, reveals that Bridge Director, Børre Stensvold, does not have the same confidence in the registration in Brutus being complete.
In the e-mail dated 21 September this year, Stensvold writes that the deviations uncovered by the RSA in Brutus are ‘very disturbing’:
[...] These are very disturbing instances of non-conformance, which in itself would be a good enough basis for a tabloid feature with fierce criticism of our procedures and follow-up systems. I didn’t think this was possible after working with such a strong focus on the technical side of bridge inspections for many years. This has also had both an external and internal focus, so I was confident we had better control. [...]
Conclusion: We can only show humility here and take full responsibility.
VG contacted NPRA’s Director of Traffic Safety, Guro Ranes, again after summing up the criticism of the registration shortcomings in Brutus.
‘My comments related to the task of conducting a systematic, general and technical assessment of crash barriers and roadside terrain,’ writes Ranes, and refers to the plans adopted in the National Transport Plan.
‘I didn’t mean to comment on errors and deficiencies in Brutus per se. The NPRA has competent experts in the regional units, and I am confident that they have a good overview of the situation, even though VG has uncovered deficiencies in the registration of inspections in Brutus,’ she adds.
A light permanently burns on the grave of Siri Tobiassen (16). Summer and winter, her parents visit the grave every three days and change the grave light. Toril shakes her head and says she knows it’s not really necessary: There’s already light where Siri is now. There was always joy and light around her daughter.
Shortly after the accident at Ask, Siri's father, Tormod Tobiassen, and Henning's father, Erik Liuhagen, started Siri and Henning’s memorial fund – which for the past twelve years has campaigned for preventive road safety.
‘It was important for my husband to find a meaning in the meaningless. Talking to the youth community here, so that at least we do whatever we can to ensure that this never happens again.’
‘If we can help to save even just one family from the grief we have suffered, then it's worth fighting for the rest of our lives,’ Toril says.
Director of Traffic Safety Guro Ranes says she has a lot of respect and understanding for what Siri’s family have gone through and the grief they've endured.
‘All of these fatal accidents are far more than just statistics to us. It's important for me to be able to say that. Every accident is one we want to avoid’.
‘The accident report concludes that the entire road network should be reviewed and crash barriers improved. Siri's mother asks what's the point of a recommendation if nothing is done?’
‘It’s perhaps difficult for the family or relatives of someone who died in an accident to see how the analysis from that single accident has helped improve traffic safety, and I understand that. But those of us who work with traffic safety know that what has come from each analysis, and not least the sum of them, means that we’re now far better equipped to propose good measures and point out where we still have challenges,’ says Ranes.
‘I realise this is of no comfort to this family, but the number of fatalities has halved since 2005, and that’s not just a coincidence.’
‘Siri’s mother says it's hard to understand that this situation has not been rectified yet, and that the NPRA needs to obtain an overview of the situation and make sure that politicians understand that this must be a priority. Do you have such an overview and have you done enough to ensure that this is being prioritised?’
‘I think we have a reasonably good overview of this and many other challenges within traffic safety. But it’s always complicated having to explain how big the challenge is or how important it is. There are many challenges in the road network and many competing goals,’ says Ranes.
Recommendations from accident reports are not automatically backed up with funding. Today, the reports do not recommend concrete action points as they did in 2005, but instead point out safety issues and proposals that are evaluated by those responsible for our roads.
Ranes emphasises that the cost of making a bridge or crash barrier safer for traffic varies considerably:
‘There are huge variations, from relatively simple repairs or replacing crash barrier end terminals, to building a new bridge or new road. The mapping work that started in the early 2000s, when they began reporting non-conformance of the standard requirements of the day, has meant that over time, a relatively large number of crash barriers on and adjacent to bridges have been replaced and improved,’ she says.
1. The NPRA’s database is called Brutus. This is a bridge management system that the NPRA itself has developed. Each bridge has its own health status journal.
2. In March, VG was given access to the database via PDF files for all of the 16,971 bridges on Norway’s national and country roads, where the NPRA is responsible.
3. Based on these data, VG has visited 104 bridges in 15 counties.
4. Analysis of the reports and VG’s own observations show that the NPRA systematically breaks the rules they are required to follow in order to ensure that Norwegian road bridges are safe.
Published: 18 December 2017
Below, VG has summarized the report for :
When a damage is discovered on a bridge, it is evaluated with a severity grade, a consequence grade and a damage type.
Severity grade is stated on a scale from 1 to 4, where 4 is most critical.
Consequence grade is also stated on a scale from 1 to 4, where 4 is most critical.
Along with the consequence grade, a letter shall indicate what kind of consequence is in question. There are four consequence types: load capacity, road safety, maintenance and environment.
VG has received a statement of all damages on load capacity or road safety, with severity grade or consequence grade 3 or 4.
Eventually, the severity grade and consequence grade are multiplied into a priority on a scale from 1 to 16. The most critical damages have priority 16B (load capacity, bæreevne) or 16T (road safety, trafikksikkerhet).
A vulnerability is a known weakness, which is not a damage.
A common example is an old railing which does not fulfill today’s requirements. Before the vulnerability category was introduced, old railings were often logged as damages, even if they were not damaged.
Main inspection: Extensive check of the entire bridge above water, normally every five years. Shall be logged.
Basic inspection: Less thorough, visual inspection of all bridge elements above water. Normally every year, but can be omitted in the year of a main inspection. Shall be logged.
General inspection: Visual control of road stretches. Can be carried out from a car at slow speed, with stops at certain objects. Every week or every second week. Are not logged.
In this presentation, we mark as a deviation where too much time has passed since the last inspection.
Brutus is also used as a planning tool for large and small action points on the bridges. When an action point has been carried out, it remains in the list, but it changes status from planned to carried out.
In this listing we mark as a deviation if an action point has an execution date in 2016 or before, and still has the status ‘planned’. But the date is only a suggestion, so this is not a violation of any rules.