Knowledge of failures is of little value unless it is disseminated and applied to upgrading design and construction practice. Forensic Engineering is progressively becoming an important archive of detailed reports on the fundamental causes of failures. It is these rigorous investigations and analyses that enable us to learn and apply lessons from failures to design, contract management and the maintenance of our infrastructure. The flurry of speculation in the immediate aftermath of a failure usually only serves to confuse and distract from the root causes.
Statistical analysis only becomes possible with a large body of quality data. For this we need to collect information on the less dramatic failures and the much larger number of warning incidents which often precede them. It is essential that we collect and collate this information so that we can obtain a clearer overall picture and have sufficient data for rigorous analysis of risks and trends. Only by drawing on international case studies can we obtain a sufficient breadth of quality data for proper analysis. The reporting of international experience in Forensic Engineering is of particular value. It is to be hoped that two excellent papers from France in this issue will inspire other contributions from around the world.
Forensic Engineering forms part of a growing international network of the research community and practising engineers. Over the coming year three major conferences will provide papers on a wide range of case studies and will be opportunities for discussion. These conferences are ASCE 6th Congress on Forensic Engineering, San Francisco, November 2012; ICE 5th International Conference on Forensic Engineering, London, April 2013; and IABSE Conference: Assessment, Upgrading and Refurbishment of Infrastructures, Rotterdam, May 2013. IABSE provides another international forum through its new Working Group 8 on forensic engineering, chaired by Robert Ratay, and through the Joint Working Commission with JCCS on robustness which has recently reported (Faber et al., 2012).
A theme linking the papers in this issue is the progressive degradation of safety with time arising from changes in loading and/or from structural alteration or material deterioration. There is a presumption in many codes that properties do not deteriorate with time and there is no factor for deterioration in the partial factor on strength. This lacuna needs to be reconsidered. Another lesson from the papers is that lack of attention to small local details and cost cutting in design, materials, construction and maintenance can have disproportionately major consequences.
Delayed ettringite formation (DEF) is one of a family of long term disruptive reactions in concrete (e.g. HAC, alkali aggregate reaction (AAR) and thaumasite) that can create major serviceability problems and, in extreme cases, unacceptable risk of structural failure. Both diagnosis and remedial measures are complicated by the interactions between these deterioration processes and surface degradation from frost and corrosion. The briefing by Ingham (2012) provides a valuable material science review of DEF, which is being identified in an increasing number of structures worldwide.
During construction DEF can be avoided by limiting the peak temperatures in concrete pours to below 65°C. However, if DEF cracking starts to develop, remedial measures can be disproportionately expensive and disruptive. The greatest risks arise in major infrastructure projects where large pours of concrete do not enable the hydration heat rise (11°C per 100 kg/m3 of cement in the mix) to dissipate before the peak temperature is reached. These conditions can arise in major infrastructure projects like bridges, tunnels and power stations, including nuclear, where the consequences are greatest.
Laboratory investigation of long term deterioration phenomena and their interactions is difficult, because of the problems of accelerating the chemical and physical processes, and the combinations of interacting processes. As the deterioration of our ageing infrastructure accelerates, it becomes increasingly important to recalibrate our design and material codes to enhance the durability and sustainability of new construction. This needs the combination of forensic investigations of developing deterioration in mature structures with parallel laboratory studies.
An example of this is the fundamental laboratory studies, such as those by Larive (1998), on AAR, and the forensic investigations of developing AAR damage to bridges and buildings (Wood, 2008). This approach led to the UK guidance on AAR (IStructE, 2010). A similar approach was adopted by RILEM TC ACS in drafting AAR-6.1 (Diagnosis & Prognosis) (Rilem, 2012) and for the forthcoming recommendation AAR-6.2 on assessment.
Breysse (2012) emphasises the value of forensic engineering in reducing risk for the construction industry and to the community. The author's advocacy at conferences and industry forums has done much to alert the French construction industry to the benefits of investigation, dissemination and analysis of failures. The paper summarises three important and well investigated failures of French structures: The Auguste Perret Meteor tunnel collapse in Paris in 2003, the roof collapse at Roissy airport terminal E in 2004 and the progressive collapse of the Saint-Etienne River Bridge on La Reunion in 2007. Breysse's reviews of failures worldwide have enabled him to identify the complex interaction of factors which create risk. We can look forward to further developments from his team at the University of Bordeaux as the body of well-reported rigorously-analysed failures grows.
The investigation of a roof failure in the Falklands by McGregor (2012) demonstrates how rigorous investigation of a minor roof failure with no fatalities can build our understanding of the realities of construction practice that can lead to a failure. Building in a different environment, without proper consideration of local conditions, often leads to problems. After 20 years of resisting winds far more severe than in the UK, this failure was initiated when the rubber washers protecting the roof fixings perished from the severe UV degradation experienced at the edge of the Antarctic ozone hole. If we are to build a sustainable infrastructure, the effect of ageing on building components also needs attention in moderate climates. Structures in extreme conditions provide a good test bed for this.
The whole cycle of forensic practice is well demonstrated in the paper by Godart et al. (2012) on the response to the failure of a corroded pile which closed the 1225 m Larivot Bridge and divided French Guiana. The bridge deck had sagged by 21 cm on its upstream edge and was close to the initiation of a progressive collapse of all 35 spans. The well integrated programme of investigation and structural analysis leading to the development of a supplementary support system enabled the bridge to be reopened after only 4 months. This has been followed through with strengthening and corrosion control measures for the whole bridge and other similar structures.
Lessons from Larivot Bridge are being incorporated in design guidance to ensure proper risk assessment for bridges, including consideration of the consequences of failure on the economy and national transport links. Emphasis is being placed on ensuring that designs are robust so there is not a risk of progressive collapse from the failure of a single element. There is also an emphasis on ensuring that cost cutting does not result in critical elements deteriorating. This should be good practice worldwide.
Unravelling the interaction of contractual and structural factors contributing to the collapse of a 40 m high scaffold into a jumble of twisted wreckage is not an easy task. Andresen (2012) maintains the reputation of the UK Health and Safety Executive in publishing the results of this complex investigation. It is a classic cautionary tale for all those involved with scaffolding. The paper summarises the way in which neglect of many aspects of good practice and poor communications led to the scaffold becoming progressively weakened by successive modifications until, when overloaded locally, it collapsed killing one worker and badly injuring two others. It illustrates how very much more complex real failures are than the over simplified probabilistic code calibrations of ‘reliability’.
The aftermath of a fire presents forensic engineers with the most difficult of challenges in combining the analysis of fire with that of the structural response. For the World Trade Center the initial damage from aircraft impact added another dimension. Chen and Cheng (Torero et al., 2012) have set out their approach to this complex task in response to the paper by Torero (2011). Torero's reply amplifies and clarifies the approach he has developed.
When you have read a paper in Forensic Engineering do not hesitate to express your views on the topic. Discussion contributions can be to obtain clarification, to challenge, to amplify or to draw attention to similar experience or alternative methods. Contributions are welcome and give authors an opportunity to add further information and to clarify their experience and opinions. Good debate leads to progress. Discussion has particular value in encouraging international exchange of experience. If a discussion contribution does not give you enough space to develop your views, the submission of a paper is always welcome.
