Genoa bridge collapse

[Mark Hewitt]

I suspect another complication is that any new structure is also probably going to need to be an upgrade rather than just a 1:1 replacement (D2M to D3M?) - with all of the follow-on implications that might have for the rest of the route.

Ideally, if they have the time and budget to do that. But it's more likely to be as quickly as possible?

[Bryn666]

The A10 is in tunnels immediately either side, and they're D2. There's little to no chance of a wider bridge except one with better merge and diverge lanes; the former structure had a 40km/h speed limit on the eastern exit because of the loop to the A7.

[roadtester]

The A10 is in tunnels immediately either side, and they're D2. There's little to no chance of a wider bridge except one with better merge and diverge lanes; the former structure had a 40km/h speed limit on the eastern exit because of the loop to the A7.

Yes - it's a wider problem. Huge amounts of infrastructure in places like Austria and northern Italy is D2M but very long stretches are almost entirely spectacular bridge/tunnel combinations. These were rightly admired as marvels of modern engineering when built but they don't lend themselves to easy widening.

Imagine in the UK context if about 50 per cent of the M4 were made up of old Severn Crossing/Usk/Wye type bridges alternating with Brynglas style tunnels - and that's without the mountainous terrain.

[Owain]

The A10 is in tunnels immediately either side, and they're D2. There's little to no chance of a wider bridge except one with better merge and diverge lanes; the former structure had a 40km/h speed limit on the eastern exit because of the loop to the A7.

IIRC, there is just one short section of D3M on the A10. It is in a tunnel some distance to the west, possibly adjacent to the A26 merge. I remember D3M through a tunnel with no hard shoulder, and I recall thinking when I drove through it that I wouldn't have wanted to be there when all three lanes were full!

Much of the A26 is D3M (including the huge hairpin that leads away from the A10), but none of the other autostrade in the area have anything more than D2M. The A7 only gains a third lane in the Po basin, well on the way to Milan; the A12 is entirely D2M. Hard shoulders are an extreme rarity in Liguria.

If the proposed link from the end of the A12 to the A26 is built as well as a replacement bridge for the A10, I shouldn't think that anything more than D2M would be a good idea, nor even necessary.

[Steve in Cumbria]

The aspect of the photos of the collapsed tower base that concerned me was that I didn't see the amount of of reinforcing bars between the foundation and the tower itself that I would normally expect. This rebar is crucial when it comes to resisting anything other than compressive forces. Given the impact of the falling bridge I would expect to see cracks. What is needed is a forensic examination of the failure to find the origin of the collapse. Waht we can see now is just the result.

Sorry, but I had to edit out the bit of my post you quoted as I found other photos that show the pier base legs are intact.




Found this photo of the cross-section of one of the cable stay A-frames. Hollow brick core was a surprise.


I am still tending to think that the failure started with the SW concrete cable stay.

[BOH]

I suspect another complication is that any new structure is also probably going to need to be an upgrade rather than just a 1:1 replacement (D2M to D3M?) - with all of the follow-on implications that might have for the rest of the route.

At either end of the bridge though it is simply very narrow D2 (so no hard-shoulders) and predominantly all in tunnels too. So no point in having D3 or D3M as this would simply create a permanent delay of 3 lanes filtering into 2. The bridge is around 1km in route and couple this with most Italian drivers thinking they are in a Grand Prix every time they drive and it is a recipe for daily accidents.

D2M would certainly make sense though on any replacement structure.

My Genoa based partner (see my earlier posts) played me the recording of an interview (in Italian) and translated it for me - it is from about 2 years ago. It is from someone who was in Morandi's senior team at the time of construction and now in his 80s. He states that the bridge was never designed for infinite trunk motorway use as the long term plan was for the proper A10 Genoa bypass (named "La Gronda"). The bridge was only to be a temporary trunk motorway route and after the construction of the A10 proper to the north, the bridge was to be relegated to local traffic use.

[Bryn666]

I suspect another complication is that any new structure is also probably going to need to be an upgrade rather than just a 1:1 replacement (D2M to D3M?) - with all of the follow-on implications that might have for the rest of the route.

At either end of the bridge though it is simply very narrow D2 (so no hard-shoulders) and predominantly all in tunnels too. So no point in having D3 or D3M as this would simply create a permanent delay of 3 lanes filtering into 2. The bridge is around 1km in route and couple this with most Italian drivers thinking they are in a Grand Prix every time they drive and it is a recipe for daily accidents.

D2M would certainly make sense though on any replacement structure.

My Genoa based partner (see my earlier posts) played me the recording of an interview (in Italian) and translated it for me - it is from about 2 years ago. It is from someone who was in Morandi's senior team at the time of construction and now in his 80s. He states that the bridge was never designed for infinite trunk motorway use as the long term plan was for the proper A10 Genoa bypass (named "La Gronda"). The bridge was only to be a temporary trunk motorway route and after the construction of the A10 proper to the north, the bridge was to be relegated to local traffic use.

Yes, precisely - hard shoulders would be a bonus as would decent merge lengths.

1960s autostrade specs, even in mountainous areas where you expect reductions, are frighteningly low compared to today's expectations.

[brombeer]

Never knew that the wider bypass (Gronda di Ponente) had already been contemplated for that long. But with the Gronda already approved -I believe the construction works are in a tender phase now- I would expect that D2M for the new bridge would do.

Ideally it will all be just a bit more spacy than it used to be (and the remainder of the Genoa bypass still is). The tight curvature, lack of shoulder, minimal slip roads etc makes passing Genoa quite an experience. Particularly when you are trying to be sensible by sticking to speed limit and a majority of others doesn't seem to care. Interesting to hear BOH's partner's local take on that situation!

[J N Winkler]

Never knew that the wider bypass (Gronda di Ponente) had already been contemplated for that long. But with the Gronda already approved -I believe the construction works are in a tender phase now- I would expect that D2M for the new bridge would do.

Autostrade appears to have conducted a prequalification exercise for spoils disposal last spring. The TED notice for this project is difficult to find since it is categorized as marine construction works (CPV 45244000) rather than motorway construction (CPV 45233110) or general road construction (CPV 45233000). I ultimately found it by searching on closing date (2018-04-17) and verified it was the correct one by cross-checking buyer (Autostrade) and contract value (€136 million).

It will be interesting to see if the A10 bridge collapse scrambles the schedule for this project. Autostrade apparently intended this €136 million down payment on €5 billion total to signal intent to adhere to a definite schedule for completion of the more substantial works, but large, multiple-fatality disasters have a way of rearranging priorities.

[brombeer]

Could go two ways. M5S, one of Italy's coalition parties, has always been against the Gronda project. And is against major new infrastructure projects generally. As late as 1 August, the transport minister hinted at the project being reconsidered. And Autostrade may become nervous too and seek to buy time until they have certainty about their future.

On the other hand, as "keep things as they are and just do maintenance" no longer is an option, you sense a shift in favour of the big and expensive project.

[Mark Hewitt]

Could go two ways. M5S, one of Italy's coalition parties, has always been against the Gronda project. And is against major new infrastructure projects generally. As late as 1 August, the transport minister hinted at the project being reconsidered. And Autostrade may become nervous too and seek to buy time until they have certainty about their future.

On the other hand, as "keep things as they are and just do maintenance" no longer is an option, you sense a shift in favour of the big and expensive project.

That goes in phases everywhere. The ‘let’s do the big projects properly’ quickly fades and you’re back into saving money and make do and mend.

[roadtester]

Sky News has a special programme about the bridge collapse this evening - I didn't catch the trailer properly but I think it's on at 6.30 and 8.30 this evening.

[owen b]

That's surely foolhardy to even suggest that's possible to do that safely within such tight timescales. A detailed study must need to take place on the state of the rest of the structure; and I can't believe, given the age of it, that any solution to just bridge the missing bit would be value for money...

I frankly doubt this segment of the A10 will be open in five months. What I would do, if I were the Autostrade CEO and had dictatorial control over the company's operations, is to recognize that the bridge has been an enormous maintenance liability from day one, and try to get a complement of long-term local and long-distance detours in place so I could demolish the entire thing and build a low-maintenance replacement with upgraded geometrics.

Now eight months according to BBC radio news just now.

[KeithW]

The aspect of the photos of the collapsed tower base that concerned me was that I didn't see the amount of of reinforcing bars between the foundation and the tower itself that I would normally expect. This rebar is crucial when it comes to resisting anything other than compressive forces. Given the impact of the falling bridge I would expect to see cracks. What is needed is a forensic examination of the failure to find the origin of the collapse. Waht we can see now is just the result.

I am still tending to think that the failure started with the SW concrete cable stay.

Thanks for the photos, that really does look light on rebar there is almost nothing to transmit torsional loads on the tower to the base. What I would expect to see is something like this.
https://www.google.co.uk/search?biw=203 ... X0nGKY5OFM:

[exiled]

Could go two ways. M5S, one of Italy's coalition parties, has always been against the Gronda project. And is against major new infrastructure projects generally. As late as 1 August, the transport minister hinted at the project being reconsidered. And Autostrade may become nervous too and seek to buy time until they have certainty about their future.

On the other hand, as "keep things as they are and just do maintenance" no longer is an option, you sense a shift in favour of the big and expensive project.

M5S is having the lessons of the reality of power in the worst possible way. It is trying to shift the blame to Autostrade per l'Italia but is discovering the internet does not forget. Especially when the commercial press in Italy is often owned by concerns that have investments in Autostade per l'Italia's parent company.

[Steve in Cumbria]

Thanks for the photos, that really does look light on rebar there is almost nothing to transmit torsional loads on the tower to the base. What I would expect to see is something like this.
https://www.google.co.uk/search?biw=203 ... X0nGKY5OFM:

There is a decent blogpost here with links in it to some articles about the bridge construction (in Spanish) and repair works in the early 90's.

[RichardA35]

The aspect of the photos of the collapsed tower base that concerned me was that I didn't see the amount of of reinforcing bars between the foundation and the tower itself that I would normally expect. This rebar is crucial when it comes to resisting anything other than compressive forces. Given the impact of the falling bridge I would expect to see cracks. What is needed is a forensic examination of the failure to find the origin of the collapse. Waht we can see now is just the result.

I am still tending to think that the failure started with the SW concrete cable stay.

Thanks for the photos, that really does look light on rebar there is almost nothing to transmit torsional loads on the tower to the base. What I would expect to see is something like this.
https://www.google.co.uk/search?biw=203 ... X0nGKY5OFM:

That may be so, but if the design cases used for analysis didn't expect these loads to be present then there would be no reinforcement to resist them and none of us know whether there should have been more reinforcement there than there was.
The knowledge and standards of the time were fairly primitive compared to today's knowledge and I would expect the investigators to be looking far further back up the design chain and examining the standards to which the structure was designed and constructed, the load cases used in analysis, how the bridge has been maintained or strengthened over the years with e.g. higher axle weights and what features would have been present in the bridge built to today's standards that were missing in the design.

A couple of questions come to my mind,
How was the integrity of each of the single cables or anchorages to be guaranteed?
What would be the outcome if a cable or anchorage failed?

I would suggest they will examine closely the arrangements surrounding what appears to be a bridge with several possible single points of failure at the cable anchorages and how the owning authority managed the inspection, maintenance and renewal to ensure that any one of these failures was not triggered.

It is interesting to see how generally the cable stayed designs have evolved away from these large single cables to multi anchorage cables (e.g. Millau Bridge, Dartford Crossing) where the loss of anchorage on one cable is unlikely to lead to complete collapse of the structure but (merely) a degradation in serviceability.

[Big L]

I came over the Erskine bridge on Friday, which seems to be single cable cable-stayed.

[RichardA35]

I came over the Erskine bridge on Friday, which seems to be single cable cable-stayed.

Indeed, opened some 20 years ahead of Dartford.

[J N Winkler]

As the engineering investigation gets underway, I expect many pathologies in common with the original I-35W Mississippi River bridge to be found, specifically the following:

* Both bridges were long recognized by their respective owners as problem children, and as such each underwent a sequence of studies and rehabilitation projects that were designed to address possible points of failure

* The actual cause of collapse will be found to be something that could have been observed in the course of a thorough inspection, and was likely photographically documented, but whose significance was not appreciated at the time

* The engineers responsible for maintaining the bridge will have been oblivious to the mechanism of failure because it is something that is typically "assumed away" in statical analysis

* There was a long-term intention to get rid of the bridge through replacement that was constantly deferred because other priorities were deemed more immediately urgent

Bridge 9304, the original I-35W bridge, had a public-sector owner and its construction and maintenance was financed through fuel taxes, though the other entities involved in its design and construction were largely in the private sector. Meanwhile, the Polcevera viaduct had a private-sector owner with owner and designer for the latest rehabilitation project (the one that was in progress at the time of collapse) forming part of the same business group: Atlantia owns both Autostrade and SPEA, which performs the bulk of Autostrade's engineering design work. However, I don't see either the distinction between public and private or, in the case of Polcevera, the horizontal integration of owner and designer contributing to the collapse in any meaningful way. Managing a problem-child bridge is an inherently technically difficult enterprise.

Some figures in the governing Italian coalition have bloviated about how the collapse represents betrayal of the private-sector promise of efficiency. To my eye this is both fatuous and ahistorical. The Italian motorway network is largely in the hands of private-sector concessionaires because, in the early 1950's when the Legge Romita was being hammered out, the Italians were unwilling to commit to an extensive free network in terms either of finance or policy. Instead, they chose to leverage public subsidy by giving concessionaires, who were largely owned by automakers, a chance to double-dip by taking profits from both the cars and the roads built for them.