26 June 2019

"Ordsall Chord - Manchester's Missing Link"

This is the third and last of a set of "souvenir" books I'm featuring which document recent major UK bridge projects.

The Ordsall Chord is quite a different beast to either the Mersey Gateway or the Queensferry Crossing, with the largest bridge span a relatively modest 89m. The Chord is a new railway connection linking Manchester's Piccadilly and Victoria stations, and although there were evidently plenty of bridge works involved, the nature of this book makes clear that it was more of a multi-disciplinary project in nature. The civil engineering construction accounts for only 38 pages, with substantial chapters given over to other topics such as railway signalling, track and overhead electrification.

The Ordsall Chord - Manchester's Missing Link (Mercury Group Limited, 2018, 168pp) was written by members of the project delivery team, and this gives it quite a different slant to either of the other two books I've featured, with a strong focus not just on the project objectives or the construction efforts, but more material on how the project was procured and organised. I think this book is therefore potentially of more interest to professionals than the general public, although I'm sure there are many railway enthusiasts who'd enjoy it.

The foreword to the Mersey Gateway book is by a politician, while the Queensferry Crossing book has multiple forewords from different perspectives. The Ordsall Chord book is introduced by the lead director from the delivery organisation, who introduces a theme that runs throughout the book, "great people working collaboratively". I think it's a interesting that this needs to be highlighted - it suggests that the construction industry is often populated by not-so-great people, not working collaboratively, so that anything else is an anomaly.

The main part of the book opens with a short chapter explaining the history of the site (the project runs right across part of the historic 1830 Liverpool and Manchester Railway), and the need for the new railway line.

The heritage theme continues into a chapter exploring the project's constraints and how the design was developed. This is a better attempt at explaining design issues than in either of the other two books, crediting and naming specific individuals rather than submerging them in corporate anonymity, exploring the challenges of working amongst numerous protected heritage structures, explaining how visual sense was made from a disparate variety of structural forms, and of how the architecture and engineering work in conjunction. The focus is very much on the architecture, but that's a relief after seeing it largely ignored in the other books.

A thorough chapter discusses the Northern Hub Alliance, a contractual partnership which brought client Network Rail together with their contractors to deliver the scheme. My experience is that Alliance arrangements are unusual in UK transport infrastructure, and while I found this chapter very interesting as a professional, I can imagine some readers' eyes glazing over.

The main message I take away from the chapter on civil engineering is the difficulty of building a project of this sort in a constrained urban environment. This is reflected in the way a large number of smaller structural elements were constructed in a "piecemeal" manner, with plenty of off-site fabrication and precasting.

The text never delves into the level of detail that would satisfy a bridge engineer, and I was left with a large number of questions, while recognising that I'm not really the main audience for such a book. The word "success" is used relentlessly, but fortunately leavened with a few short acknowledgements of real problems encountered during the project, such as issues with the stressing of the network arch bridge hangers.

The remaining chapters cover the other railway disciplines and (briefly) the project's outcome and legacy. I'm no trainspotter, but I think I did learn a few new things from reading these.

It's very well-illustrated throughout, with plenty of photographs.

The book is available for £25 plus £5 postage from a dedicated website, or from Amazon.

Further information:

23 June 2019

FIU Bridge Collapse: Designer's analysis

My reader Patrick Sparks has drawn my attention to a presentation from the FIU Bridge designer which makes an interesting read in light of the recent OSHA report.

The linked PDF comprises the minutes of a meeting held to receive a presentation from the bridge designer Figg at 9am on 15th March 2018, just a few hours before the bridge collapsed, plus a series of photos capturing the presentation slides*. Figg were present at a meeting along with the Florida Department of Transportation (FDOT), the contractor MCM, the university FIU and their construction representative BPA. The aim of the meeting was to review cracking which had arisen in the bridge during construction.

(*Is this a normal thing to do in US engineering meetings? It feels very odd to me.)

It is clear that despite the very severe and unexpected cracking which had occurred, nobody present considered it necessary to take any immediate action other than to proceed with re-stressing of the cracked truss diagonal.

The designer "assured that there was no concern with safety of the span suspended over the road". The client's construction adviser specifically asked if the bridge should be propped but the designer stated that it was not necessary.

The actions discussed were mostly take-aways: the construction adviser would review the designer's analysis over the next few days; the designer was working on a scheme to "capture" the cracked truss node; there was agreement that the engineering peer reviewer should take a look at the situation, but no clear action or timescale for them to actually do so.

A re-stressing procedure was due to take place within a few hours following the meeting, but the designer left site rather than stay to observe, and the client's construction representative had not received details of the procedure prior to the meeting.

Reading the notes from the meeting what strikes me is a total lack of clarity from everyone present. Some actions are implied, but none are clearly recorded. Nobody was present who could provide meaningful comment or review of what the designer had to say. Those present accepted that there was a serious and unexplained issue, but felt it appropriate to push on with re-stressing the truss end diagonal even through the cause of the cracks was completely unknown. There is no sense in the minutes of a forensic approach to the failure.

Some of this is not unexpected - I have sat in similar meetings where there is no clear leadership and the approach to solving a problem appears to be unstructured. It is, however, something for other projects to learn from.

There are inconsistencies between what the designer presented, and what the OSHA investigation reports.

The first of these relates to the stage when the falsework supporting the concrete truss was removed, leaving it sitting on temporary supports before transport to the bridge site. The designer's presentation states there were no significant cracks at this time, but the OSHA report reveals that a cracking or popping sound was heard and several cracks were then visible in exactly the same part of the structure that led to later concern.

Much of the presentation is given over to a re-calculation of vertical transverse bending in the end diaphragms of the bridge. It isn't clear why, as the visible cracks were not characteristic of vertical bending.

A key part of the designer's re-analysis starts on page 27 of the PDF, headed "Total Nodal Shear Stability". This is their check of whether the truss end node can fail by punching out of the end diaphragm. Differences to the OSHA analysis are readily apparent: OSHA calculated shear on a horizontal plane in line with a casting joint, while the designer considered shear on two vertical planes through the diaphragm either side of the node.

OSHA's calculations show the horizontal plane to have had insufficient strength: the concrete area was relatively small, the construction joint was a plane of weakness, the area of reinforcing bar passing through the plane appears to have been small. The designer's calculations showed the vertical plane(s) to be ok: the concrete area was relatively large and doubled (two shear planes either side of the truss diagonal), the area of reinforcement was large, and transverse prestress provided a "clamping" force which enhanced shear strength.

The discrepancy indicates that the designer did not re-calculate shear on the horizontal plane, which was later found to form part of the observed failure mechanism. Both shear planes should have been checked (and any others which might be identified as being potentially weaker).

The designer considered in detail the changes in support of the truss from the initial temporary supports to the permanent supports, with a different shim arrangement, and concluded that it was unclear how this change could possibly create the cracks that were observed. The cracks were described by the designer as "spalls", seemingly not recognising that the end node was in fact in the process of gradually moving and becoming detached from the body of concrete.

The designer's conclusion was as follows:

Paraphrasing this, their calculations showed that the bridge was okay, therefore the bridge was okay. The cracks could not be significant if the computer model did not explain them.

In hindsight, this sounds obviously faulty reasoning, but we should be wary of hindsight bias. With the evidence clearly laid out before us, the significance of the cracks may now seem obvious. However, with the cracks unexplained, the designer was using their familiar tools to try and understand what had happened. Although OSHA has been quick to blame everybody involved in the project for their alleged failures, I think there will be many designers who will wonder whether they would have done any different.

Another cognitive bias may have applied back in March 2018. Here's what Wikipedia has to say about confirmation bias:
Confirmation biases contribute to overconfidence in personal beliefs and can maintain or strengthen beliefs in the face of contrary evidence.
Is this what happened? Were project participants willing to ignore the evidence of progressive failure because it simply did not fit in with their mental model of how the structure should behave? The designer is likely to have had considerable experience of computer modelling, and of structures that behave in a manner that doesn't contradict it. It's easy in that situation to think that the computer modelling is right (in fact, errors and omissions may be made in computer modelling from time to time, but usually structures behave well enough for this not to become evident).

Groupthink may also have had an influence on the outcome. Here is Wikipedia again:
Group members try to minimise conflict and reach a consensus decision without critical evaluation of alternative viewpoints ...
The minutes of the project meeting show that the designer's viewpoint was challenged, but it seems that challenge wasn't pursued with sufficient firmness - the designer's recommendations (to re-stress the truss diagonal, and not to prop the structure) were followed without giving time for alternative views to inform these decisions. The minutes are unclear who was responsible for the decision anyway: the word "safety" appears only once in the minutes, and it is not clear from the record who took the lead responsibility for safety of the actions taken.

Everyone is vulnerable to confirmation bias, groupthink and similar irrational judgements. The evidence is that cognitive biases in general have a big impact on how individuals and groups operate in many widely varying contexts. Recognition of this has driven the entire field of behavioural economics, but it's not something you hear much about in the engineering context.

I hope that the official incident investigation digs into some of these issues, as I think they are in many ways of wider relevance than issues of calculations and concrete strength, the specifics of which will always vary from project to project. It would be foolish for designers of steel bridges to think that the FIU bridge incident is not relevant to them, for example. The factors (including, often, commercial pressures) which lead to poor analysis of evidence and hence poor decision-making cut across all types of project and many types of error.

I feel the issue of identifying effective behaviours and putting in place processes and a culture which encourage rather than discourage them is the key area where attention should be focused.

20 June 2019

FIU Bridge collapse: OSHA investigation findings

A report has been published regarding the collapse of the FIU Pedestrian Bridge. The bridge collapsed on 15th March last year during construction, resulting in six fatalities and several serious injuries.

The report has been prepared by the US Occupational Health and Safety Administration (OSHA), which I take to be roughly equivalent to the UK's Health and Safety Executive (HSE). A thorough investigation is being carried out by the US National Transportation Safety Board (NTSB), due to be completed later this year, but the OSHA report largely echoes preliminary findings issued by NTSB last November.

There is no direct equivalent to the NTSB in the UK: we have national bodies for investigating aviation and railway accidents, but not for highways, a gap that is long overdue for filling (no doubt, we will have to wait for our own serious incident for this to change, rather than learning from elsewhere).

The bridge was a highly unusual design, comprising two concrete truss spans, each with a single truss along the centre-line of the bridge deck. A pylon and steel pipe stays above the trusses were arranged to give the impression of a cable-stayed bridge, but were largely decorative, with the only reported structural purpose being to reduce pedestrian-induced vibration in the main structural elements. I discussed the design in more detail in March 2018.

OSHA's report into what went wrong appears detailed and thorough, but some caveats are in order.

The report does not hesitate to apportion blame, stating that the designer (Figg), design peer reviewer (Louis Berger, performing an independent check of the entire design), the construction engineer and inspector (BPA, acting as FIU's representative), and the contractor (Munilla Construction Management, MCM) all failed in their duties and missed opportunities to prevent the collapse.

Oddly, less blame is attributed by OSHA to the highway authority who allowed the bridge to be built above live traffic (Florida Department of Transportation, FDOT). I've commented on this previously: in the UK and in some other jurisdictions, a highway authority would undertake a process of technical approval for any new bridge built over or under their infrastructure, to protect their own interests which include both avoiding undue impact on traffic, and protecting the safety of highway users.

FDOT took some interest in the project, having their in-house structural engineering group review the designs, and they were represented at regular site meetings including a fateful meeting on the morning of the collapse. Like all the other participants, FDOT's representative missed the opportunity to call a halt to activity on site, but oddly, this is omitted from the report’s executive summary. It is also left unclear whether FDOT were aware of or agreed to gaps in the scope of the independent peer reviewer, who reportedly did not check the bridge in any construction stages, only in its final state.

OSHA appear to have had extensive access to evidence from before and after the bridge collapse, but their report shows only limited sign of having talked to the main participants. Allegations are made, but any responses received are absent. Figg have rapidly gone on record to criticise the report, stating that they are prevented from saying more publicly while the NTSB investigation continues.

So take anything that follows with a pinch of salt – despite the report's impression of thoroughness, it may be partial and unfair. We'll have to wait until later this year to see what the NTSB have to say.

OSHA start by repeatedly drawing attention to the fact that the bridge was expected to be an aesthetically attractive landmark structure. No connection to the collapse is made explicit, but they seem to want to create the impression that the circumstances of procurement contributed to failure. Otherwise, why mention the topic? Personally, I think the FIU were entitled to rely upon the professional integrity and judgement of their various advisers and contractors, and cannot be blamed for wanting a nice bridge.

The main truss span was cast off-site in four stages on a falsework assembly, before being picked up by a self-propelled modular transporter (SPMT), driven to site, and then set down on its final supports.

Prior to the SPMT stage, all truss members which would experience tension either during construction or in service were prestressed using stressing bars. Most stressing bars were grouted in, except for the two end diagonals (truss members 2 and 11), as these required prestress to be adjusted during construction.

The first sign of trouble came when the temporary falsework was removed, leaving the truss span supported on temporary trestles at each end. Operatives reported hearing a "loud popping sound", and soon discovered cracks at the bases of diagonals 2 and 11, where they connect to the bridge's lower deck slab. In this configuration, the end diagonals were in compression, so any prestress applied should have been minimal. Photographs of the cracks were shared with the designer, who OSHA state showed no real concern. The image on the right shows one of the larger cracks at this stage, at the bottom of diagonal 11.

Over a week later, the structure was transferred from the temporary end supports onto the SPMT ready for transport to site. The SPMT carried the bridge on the lower truss nodes one bay in from the end supports. At this stage, diagonals 2 and 11 would have been in tension, and prestressing would have been required. Once placed on the permanent supports, the intention was then to de-stress the bars in these members.

According to the OSHA report, no further cracks appeared and the initial cracks did not develop further during transportation and installation of the truss span. Member 2 was then de-stressed without incident, but further signs of trouble emerged rapidly during de-stressing of member 11. Multiple cracks appeared at the lower node of this member, leading one prestressing operative to report that "it cracked like hell".

Two days later, pictures of the cracks were sent to the designer for comment. Construction operatives later stated that the cracks had already grown larger, and photographs taken at this stage appear quite alarming in what they show, with evidence that the truss node was separating from the lower slab. Nonetheless, the designer stated that "this is not a safety issue", proposing that member 11 be re-stressed to return it to its previous state. The designer did not plan to attend site during this operation.

Photographs of the cracks taken over the next two days look steadily more alarming. A key meeting was held 5 days after installation, attended by FIU, the designer, contractor, construction inspector, and FDOT. The designer had now inspected the cracks (which continued to grow) in person, but stated that there was still no safety concern.

Nonetheless, they emphasised the need to proceed with construction of the second span (which, when made continuous with the first span, would reduce loads on member 11), and stated that they would examine methods by which the progressively cracking node could be "captured" or "restrained". According to OSHA, no reference was made back to the peer reviewer at any stage.

On the face of it, the recommendation to re-stress diagonal 11 was very strange. The stressing bars were provided for the temporary condition when the bridge was supported on an SPMT, a situation which no longer existed. Once the bridge was erected on the permanent supports, re-stressing the diagonal added to load in the diagonal and its connecting nodes, rather than reducing it.

Re-stressing commenced as instructed, and was nearly complete when the bridge collapsed.

OSHA describe the failure as a "concrete blow-out", but an appropriate term for it is a shear or punching shear failure: the upper part of the truss node sheared clear out of the lower deck slab, partially along the lines of construction joints created during the original phased construction pour.

The image below shows the node after failure, looking towards the end of the deck. The large empty pocket was originally solid concrete; the rough area in the centre of the bottom edge of the photo was originally where the truss diagonal connected into the deck. The various vertical ducts appear to be for lighting cables or similar.

The next photo shows the same area looking straight onto the end face of the deck slab after it was removed from the collapse site. The empty pocket at the top is the same empty pocket visible in the previous image.

The cracks observed prior to failure are consistent with this, including longitudinal splitting seen on diagonal 11, which was presumably caused by the longitudinal movement of the bottom node, splitting occurring due to the different anchoring positions of the upper and lower stressing bars within the concrete.

With the node destroyed, the truss was no longer fully triangulated at all points, and the full load of the bridge could only be carried by bending of the top or bottom truss members (the roof canopy and the deck slab), a condition which neither member could ever have been designed for. This is precisely what is meant when the design is described as lacking redundancy – there was no secondary load path. In addition, there was limited ductility – shear failure can occur much more suddenly than a bending failure.

Here is the node detail as shown in the OSHA report, viewed from the side – the position of the construction joint is not clear but the report states elsewhere that the truss diagonals and verticals were cast after the deck slab, so the joint runs horizontally somewhere through the middle of the diagram.

OSHA's calculations show that the combined capacity of reinforcing bars crossing the construction joint, plus friction caused by the vertical component of axial load in diagonal 11, were insufficient (in accordance with the design codes) to resist the horizontal component of axial load in diagonal 11: the shear demand was calculated to be 22% higher than the shear capacity. Re-stressing the bars made the situation worse, with demand then 45% higher than capacity. I have some diagrams below which perhaps make this clearer.

The potential punching shear failure is shown in this diagram in the OSHA report, viewing the node from above with the force from member 11 acting towards the bottom of the diagram:

The report doesn't spell out why failure did not occur when the truss was first landed on its permanent supports: the loads and prestress state at this time were the same as immediately prior to the actual failure.

The progressive nature of the cracking may be relevant here: the re-stressing was applied to a structure which had already severely fractured, showing evidence of slip along the shear planes of the construction joints. In this situation, resistance would have been greatly reduced below what the design code suggests. The fracturing and slip itself may have been caused by an initial failure of the concrete surface upon installation, with the progressive cracking then caused by plastic deformation of the reinforcing bars passing across the joint.

The visible evidence of the punching shear failure also indicates the failure to have been 3-dimensional and the 2-d drawing of the node is therefore misleading. What particularly bothers me about the node detail is that it is highly unclear how the compression in diagonal 11 is transferred into the tension system of the post-tensioned bridge deck. The OSHA report draws attention to this, and it seems to be the case that a lack of adequate provision to transfer these forces laterally from the plane of the truss into the plane of the deck contributed to the failure.

In the OSHA analysis, the truss node was therefore under-designed for the loads it had to carry; it had not been checked in this condition by the independent checker (who reportedly considered only the final situation once the two spans were made continuous); and there was no redundancy in the design, such that if any one truss member failed, bridge collapse was inevitable. To add to the bridge's vulnerability, vertical ducts were located either side of the node, potentially weakening the whole area – these are visible in one of the photographs included above.

It may be partial at this stage, but the picture that OSHA sets out is consistent and compelling.

Their report is lacking in diagrams which clearly show the lines of force, the positions of the cracks, and the eventual failure, so I've had a go.

The first diagram shows the compression in diagonal 11 (red), which can be resolved into a vertical compression force and a horizontal shear force where the diagonal connects to the deck (both in grey), with the horizontal force balanced by tension in the deck (blue).

The second diagram shows the position of cracks visible prior to failure (yellow), and the direction of movement of the upper part of the node corresponding to these cracks (white).

The final diagram shows very roughly the planes along which the concrete failed to cause collapse. On the left hand edge of the section highlighted in black, a block of concrete punched clean out, while on the right hand edge of the same area, failure was in pure shear through a construction joint between the diagonal and the deck.

Note that the construction drawing shows shear reinforcement passing through that part of the construction joint – failed reinforcement can be seen in some of the OSHA photographs. The report doesn't state the diameter of these bars, but from the photographs they appear to be very small compared to other visible reinforcement.

OSHA concludes (a) that the truss node was under-designed for the forces it had to carry, and (b) that a series of competent professionals failed to take adequate precautions when clear evidence of problems with the structure came to light. OSHA addresses the "what", but not the "why".

I think speculation should be limited and those involved should have the opportunity to explain their actions and identify any flaws in OSHA's argument. However, I think there are a number of questions which should be asked:
  1. How did the agency responsible for highway safety ensure that an acceptable process was in place for design and construction, and did they review or agree any of the methods of design to be used, especially for critical or unusual details? In the UK, the design methodology would be presented in advance through a Technical Approval process, reviewed and accepted by the relevant infrastructure authority. This process was introduced into the industry following notorious bridge failures in the 1970s. Do the US authorities apply anything similar, and if not, is there a failure to learn from international experience?
  2. Who determined and who accepted the peer reviewer's scope? This was an unusual design, with significant changes in load at different construction stages, yet apparently the peer reviewer entirely ignored key construction stages, including the stage where collapse occurred. Why was this not part of their scope?
  3. What calculations did the designer undertake? Were key calculations missed, or incorrect, and if so, what errors were made? The OSHA report presents the conclusions of the agency's own calculations but says nothing about what the designer actually did or why in their view the node design was adequate.
  4. Given the structure's clear lack of redundancy, how did the designer conceptualise critical details? Were simple shear rules applied or were complex details such as the node modelled in more detail? I would have expected a finite element model of a detail like this, unless simple calculations showed it to be highly robust.
  5. Why did the designer state that the cracking was not a safety issue? With hindsight, it seems clear that a serious developing failure should have been apparent, with the cracking demonstrating insufficient strength. It should also have been clear that once the concrete cracked in the manner it did, the strength of the node was immediately reduced.
  6. Was it reasonable for the other project participants to rely entirely on the designer's advice? OSHA's view is that they did not act reasonably – that they had sufficient expertise to stop work until further precautions were in place. I'm far from convinced about that: the designer was clearly the structural engineering expert and from one perspective it's hardly unexpected for others to defer to that.
  7. What external pressures existed, relating to cost or programme, or to a desire not to disrupt traffic further?
  8. Was this the most appropriate design, particularly with the choice of materials? It seems difficult to believe that this failure would have occurred in a steel truss design.
OSHA's report is very direct and reductive in its analysis of failure. The truss node failed in punching shear. The designed capacity of the node to resist this failure mode was inadequate. The designer, and everyone else involved in the project, failed to recognise the seriousness of the initial cracking. Nobody took any steps to prevent the cracking from progressing, or to safeguard highway traffic. The action taken, re-stressing the prestress bars, made things worse rather than better.

This is a clear causal chain, and in the Swiss cheese model of systemic failure, all the holes in the various layers of cheese lined up. However, I'm not convinced that it is sufficient simply to identify human error: mistakes are inevitable, and those responsible for the system should put in place mechanisms both to identify error (such as independent checking), and to minimise its impacts (such as adoption of robust materials and structural forms).

If the OSHA report is accurate, then serious issues are apparent, and from conversations with others in the industry, these issues are not unique to the FIU pedestrian bridge project. Some of the key issues appear to be behavioural: reliance on advice without challenge; the assumption that competent people do everything right; a reluctance to take the evidence of today and imagine what it will result in tomorrow; the suggestion that because an occurrence is not yet explained, no action should be taken.

Many of these can be attributed to cognitive biases and even if they were found not to have caused the failure of the FIU bridge, those involved in other projects should still reflect on their own safety culture, the way in which teams respond when challenged, and what can be done to ensure appropriate behaviours are rewarded rather than discouraged. Engineers should also reflect on whether the divisions of responsibility in their technical governance and assurance processes are fit for purpose. Given that errors will occur on some project at some time, what are we doing to prevent them resulting in catastrophic outcomes?

My experience is that structural engineers are often prone to simply following normal process, standards and methods. Risk assessment too often concentrates on construction and operational hazards or even just on commercial risks. It is rare to see a structural engineering plan of work arising directly from a meaningful analysis of the risks of structural failure: often, it's just assumed that codes and standards are sufficient to address those particular risks. The best way to work out how to make a structure stand up is first to work out how it would fall down, a lesson that could have been learned from many disasters but is still not embedded properly in regular practice.