Structural robustness

Structural robustness is the ability of a structure to withstand actions or damage, without disproportionate effect on its structural integrity. While current design odes accept structural damage when extreme events occur, it is clearly stated that this should not be disproportionate to the initial cause. Hence, while for the day-to-day operation of a building damage is unacceptable (this is why the Serviceability Limit State – SLS – is considered), for scenarios that might occur quite rarely, such as a strong earthquake, it is acceptable provided that it is not excessive enough to cause (a) significant economic loss, (b) collapse, or (c) casualties.

A factor that was often not considered in conventional design (e.g. residential buildings) is any type of accidents. The reason is that accidents which can be a hazard to structural integrity are quite rare. However, when significant accidents occur, if the buildings have not been explicitly designed against such events, the results are often devastating, leading to progressive collapse of part or the whole structure. Hence, when the overall risk is considered, while the probability is low, the effect is significant, so the risk is considerable.

Progressive collapse

The term ‘progressive collapse’ resistance refers to the inherent ability of a structural system to withstand limited scale failure without disproportionate propagation of the damage to the part which was not directly affected by the cause (Papavasileiou, 2013). For structural engineering purposes, the term ‘collapse’ is not only used to describe the phenomenon with its conventional interpretation, but it includes also significant deformations of the structural members that would be equivalent to collapse. For example, if a 7m-span steel beam deflects about 1.5m, the floor above might not have collapsed, but the deflection would be so large that it would constitute collapse.

In the previous decades, structural design that would accept damage focused mainly on the seismic assessment of buildings. Although the vulnerability of buildings to progressive collapse has been identified since 1968 and the partial collapse of the Ronan Point building (Pearson and Dellate, 2005), researchers started studying the issue mainly during the past decade. Undoubtedly, the triggering event for such an intense investigation was the collapse of the World Trade Centre buildings (Figure 3.1) which caused the death of 2,996 people, while more than 25,000 were injured.

Wikimedia / Public domain

Figure 3.1. Photo of the World Trade Centre twin towers after the attacks.

Figure 3.2 shows the WTC twin towers before and during the collapse.

Pauljoffe at English Wikipedia / CC BY-SA
Figure 3.2. Photo of the World Trade Centre twin towers before and during collapse.

Several investigations have already been published on the topic (Ellingwood and Dusenberry 2005, Starossek and Wolf 2006, Nair 2006, Ellingwood 2006, Sørensen and Christensen 2006, Biondini and Restelli 2008, Izzuddin et al. 2008, DoD 2010, Faber 2011, Gerasimidis et al. 2013, Cavaco et al. 2013). However, a commonly accepted definition has not yet been established (Starossek and Haberland 2010, Faber 2011). Nevertheless, various efforts to provide a means to quantify this ability are reported in literature (Frangopol and Curley 1987, Lind 1995, Ellingwood 2006, Maes et al. 2006, Biondini and Restelli 2008, Izzuddin et al. 2008, Ghosn et al. 2010, Starossek and Haberland 2011, Faber 2011).

Assessment of progressive collapse resistance

A method often used to assess the capacity of a structure to withstand structural damage is the non-linear static analysis under gravitational loads, often referred to in the literature as Pushdown Analysis. An element, usually a column, is removed from the simulated model (see Figures 3.3 and 3.4) and the structure is then required to receive the gravitational loads, which are applied to it progressively. The analysis can be either force-controlled, in order to evaluate the structural response against performance criteria given in the applicable design codes, such as UFC 4-023-03, or displacement-controlled, in order to assess the collapse potential of a building. It is considered to be equivalent to the Nonlinear Pushover Analysis performed at the z-z’ direction. Despite the fact that pushover and pushdown analysis types share a common philosophy, they differ significantly. A major difference between these two simulations is the time instance when the event is considered to take place. Contrary to the Pushover Analysis, which simulates the action itself (i.e. the seismic loads), the Pushdown Analysis takes place after the event has occurred (e.g. an explosion, the collision of a track, or a seismic excitation which significantly damaged the building).

Design of structures against progressive collapse

Progressive collapse previously described the structural failure in the form of a chain reaction, which is the consequence of the damage to a relatively small part of the structure. There is an increase of the internal forces of the elements neighbouring, usually, to a column which has been removed or destroyed, either by human action or by natural causes. If the extra loads cannot be properly redistributed to the undamaged elements, then the damage extends, and a broader or global failure of the structure occurs. It is an unacceptable failure type of a structure, not only because of the disproportionate extension of the damage, compared to its cause, but also because it can take place almost instantly after the failure of the first structural element. Thus, it is necessary to ensure the ability of the structure to receive accidental loads, usually as a result of local failures, without extensive damage. The ability of a structure to sustain small failures without endangering the integrity of the whole system is called structural robustness.

The methods described in the existing design guidelines (COST TU0601, 2011a; COST TU0601, 2011b, DoD, 2005; DοD, 2010), in order to limit the probability of progressive collapse can be divided into three distinct categories: event control, indirect design and direct design. Event control refers to the safety measures which could be taken in order to prevent the initial damage from occurring and, as such, it is usually not considered to be a matter of concern to structural engineers. Indirect methods aim to put extra inherent strength (robustness) in the structure so that it will be able to withstand the initial damage without extensive failure or collapse. Direct design involves the design of structures taking into consideration either the specific abnormal action or its result on the structure.

The most popular method of direct design against progressive collapse is the alternate load path (ALP) method, in which a structural element, usually a column, is considered to have experienced a destructive event and has failed. The structure is then modelled without the failed element (Figure 3.5), and an analysis under gravitational loads takes place in order to determine its ability to remain stable, as well as the extent of the damages occurring as an indirect result of the initial damage (i.e. elements yielding or failing after the specific member has been removed). It is also referred to as Missing Column Analysis, as the member eliminated is usually a column at the base of the structure, or generally Missing Member Analysis.

© Papavasileiou, 2013 – sketched according to UFC 4-023-03
Figure 3.5. Correct and incorrect modelling for the Alternate Path method - UFC 4-023-03

UFC 4-023-03

In 2005, the United States Department of Defence published the Unified Facilities Criteria (UFC) Design of Buildings to Resist Progressive Collapse (UFC 4-023-03). Its current version, which was revised in 2010, is considered to be one of the well-reviewed documents on progressive collapse in the U.S. In the current version of Eurocodes, there are no detailed provisions for design against progressive collapse. Therefore, engineers who want to design buildings able to sustain a small- scale or even larger-scale damage without local collapse, often incorporate the provisions of UFC 4-023-03 in their design.

UFC 4-023-03 describes three design approaches for buildings, in order to resist progressive collapse: the Alternate Path method (AP), the Enhanced Local Resistance method (ELR) and the Tie Force method (TF). Based on the classifications defined in ASCE 7-10, UFC 4-023-03 classifies the first two design approaches as “direct” methods and TF as an “indirect” method. The main difference between these two classifications is that, while for the “direct” methods a detailed design and analysis is required, the Tie Force method is intended to enhance the integrity of the building by attributing an overall robustness to the structural system.

The Tie Force method was first proposed in 1971, as a response to the collapse of the Ronan Point Building in 1965, by the Institution of Structural Engineers, in the United Kingdom. This method aims to enhance the ductility and continuity throughout the building, creating a network of internal ties. The term “tie” refers to the ability of an element or a group of elements performing as a system, such as beams, columns, slabs, or even rebars, or connections, along a direction to receive tensile forces, limiting this way the deformation of the structure along its direction. In UFC 4-023-03, these ties are classified as longitudinal, transverse, peripheral and vertical ties. Longitudinal and transverse ties are expected to be developed through the catenary behaviour of the floor system, while beams and columns are the elements which provide the peripheral and vertical ties accordingly. Figure 3.6 illustrates these ties on a regular frame structure, as defined in UFC 4-023-03. The arrows denote the result of the tensile force developed by the tie on the building.

© Papavasileiou, 2013 – sketched according to UFC 4-023-03
Figure 3.6. Tie Forces according to UFC 4-023-03.

In order to take into consideration the dynamic nature of the action which causes the loss of a load-bearing element, when a nonlinear static pushdown analysis is performed, current guidelines (U.S.A. Department of Defence, 2005; 2010) suggest the use of a dynamic increase factor (DIF) which multiplies the gravitational loads on the affected bays. Lots of researchers consider the use of a fixed value of 2.0 to be conservative and counter-propose methods of calculation of its value.

Example of the Alternate Load Path method

Let’s assume that we’ve got a 6-storey frame building with 6 bays in each horizontal direction, carrying only gravitational loads, i.e. its self-weight and any imposed load due to typical use. If the building is symmetrical, it can be represented by a plane frame in its main direction, while the surface loads from the slabs are distributed on its beams, as shown below.

The loads from a typical beam, would be transferred to the neighbouring columns and, through the columns, to the foundation and then safely to the ground, as shown below. Of course, in moment-resisting frames (MRFs), the stress distribution in the frame is more complex than that, as moments develop in the structure and cause internal forces to virtually all elements are well. However, this is the direction of the load itself (the load path), while the aforementioned internal forces are its effect on the frame.

If the column under this beam is damaged due to an accident (e.g. a gas tank explosion), then the loads will use a different path to travel to the ground. Part of the load that previously travelled through the column on the left will now use the column to the right of the beam (red load path), while part of it will travel upwards (the beam’s left end will be suspended from the column above) and then follow a similar path (cyan load path). Also, it should be pointed out that the support conditions for this beam ad those above it will change.

The new conditions imply that:

1. the beams should be designed for increased bending moment and shear to avoid failure due to this event,
2. the columns on the right should be designed for increased compression (due to the excess load) and additional bending moment from the beams to avoid failure due to this event,
3. the columns on the left might be in tension, an additional state that needs to be considered,
4. the beam-column connections will need to be designed for the new loading cases, and
5. the beams above the damaged bay need to be appropriately designed to limit their deflection within the admissible limits.

As one can see, a number of additional requirements need to be addressed in structural design to allow for the structure to withstand this damage scenario without undesirable results.

Activities

References 

Biondini, F., and Restelli, S. (2008). Damage propagation and structural robustness. In F. Biondini, and D. Frangopol (Ed.), Life-Cycle Civil Engineering: Proceedings of First International Symposium on Life-Cycle Civil Engineering (IALCCE’08) (pp. 131-136). London, UK: Taylor andFrancis.

Cavaco, E., Casas, J., Neves, L., and Huespe, A. (2013). Robustness of corroded reinforced concrete structures – a structural performance approach. Structure and Infrustructure Engineering: Maintenance, Management, Life-Cycle Design and Performance, 9 (1), 42-58.

CO.S.T. TU0601. (2011a). Structural Robustness Design for Practicing Engineers. (T. Canisius, Ed.) T. D. Gerard Canisius: European Cooperation in Science and Technology.

CO.S.T. TU0601. (2011b). Theoretical Framework on Structural Robustness. (J. D. Sørensen, Ed.) European Cooperation in Science and Technology.

Ellingwood, B. (2006). Mitigating risk from abnormal loads and progressive collapse. Journal of Performance of Constructed Facilities, 20 (4), 315-323.

Ellingwood, B., and Dusenberry, D. (2005). Building design for abnormal loads and progressive collapse. Computer-Aided Civil and Infrustructure Engineering, 20, 194-205.

Faber, M. (2011). Robustness of structures – Final report of COST action TU0601. COST (European Cooperation in Science and Technology) .

Frangopol, D., and Curley, J. (1987). Effects of damage and redundancy on structural reliability. Journal of Structural Engineering, 113 (7), 1533-1549.

Gerasimidis, S., Bisbos, C., and Baniotopoulos, C. (2013). A computational model for full or partial damage of single or multiple adjacent columns in disproportionate collapse analysis via linear programming. Structure and Infrastructure Engineering, 1-14.

Ghosn, M., Moses, F., and Frangopol, D. (2010). Redundancy and robustness of highway bridge superstructures and substructures. Structure and Infrustructure Engineering: Maintenance, Management, Life-Cycle Desing and Performance, 6 (1-2), 257-278.

G.S.A. ‘Progressive Collapse Design Guidelines Applied to Concrete Moment-Resisting Frame Buildings’, General Services Administration, Nashville, Tennessee. 2004.

Izzuddin, B., Vlassis, A., Elghazouli, A., and Nethercot, D. (2008). Progressive collapse of multi-storey buildings due to sudden column loss - Part I: Simpified assessment framework. Engineering Structures, 30, 1308-1318.

Lind, N. (1995). A measure of vulnerability and damage tolerance. Reliability Engineering and System Safety, 48 (1), 1-6.

Maes, M., Fritzsons, K., and Glowienka, S. (2006). Structural robustness in the light of risk and consequence analysis. Structural Engineering International, 16 (2), 101-107.

Papavasileiou, G. S. (2013). Optimized seismic design and retrofit of collapse-resistant steel-concrete composite buildings. Nicosia, Cyprus: University of Cyprus. (PhD Thesis)

Pearson, C., and Dellatte, N. (2005). Ronan point apartment tower collapse and its effects on building codes. Journal of Performance of Constructed Facilities, 19, 172-177.

Sørensen, J., and Christensen, H. (2006). Danish requirements for robustness of structures: Background and implementation. Structural Engineering International, 16 (2), 172-177

Starossek, U., and Wolf, M. (2006). Design of collapse-resistant structures. In Proceedings of JCSS and IABSE Workshop on Robustness of Structures, Building Research Establishment. Garston, Watford, UK.

Starossek, U., and Haberland, M. (2010). Disproportionate collapse: Terminology and procedures. Journal of Performance of Constructed Facilities, 24 (6), 519-528.

Starossek, U., and Haberland, M. (2011). Approaches to measures of structural robustness. Structure and Infrastructure Engineering: Maintenance, Management, Life-Cycle Design and Performance, 7 (7-8), 625-631.

U.S.A. Department of Defence. (2005). Design of Buildings to Resist Progressive Collapse, UFC 4-023-03.WashingtonD.C.: Unified Facilities Criteria.

U.S.A. Department οf Defence. (2010). Design of Buildings to Resist Progressive Collapse, UFC 4-023-03.WashingtonD.C.: Unified Facilities Criteria.

Further reading

Arup (2011). Review of International Research on Structural Robustness and Disproportionate Collapse. Department for Communities and Local Government. [Available at www.communities.gov.uk/publications/planningandbuilding/robustness]

Brooker, O. (2008). How to Design Concrete Buildings to Satisfy Disproportionate Collapse Requirements. TCC/03/45. Camberley: The Concrete Centre.

Carpenter, J. (2007). SCOSS reports on progressive collapse & robustness. The Structural Engineer 85(5), 5.

Harding, G. and Carpenter, J. (2009). Disproportionate collapse of ‘Class 3’ buildings: the use of risk assessment. The Structural Engineer, 87(16–18), 29–34.

Institution of Structural Engineers (1971). RP/68/05: The Resistance of Buildings to  accidental Damage. London: IStructE.

Institution of Structural Engineers (2010). Practical Guide to Structural Robustness and Disproportionate Collapse in Buildings. London: IStructE.

Ministry of Housing and Local Government (1968a). Circular 62/68: Flats Constructed with Precast Concrete Panels. Appraisal and Strengthening of Existing Blocks: Design of New Blocks. London: HMSO, 15 November.

Ministry of Housing and Local Government (1968b). Report of the Inquiry into the Collapse of Flats at Ronan Point, Canning Town. London: HMSO.

Office of the Deputy Prime Minister (2004). The Building Regulations 2000 – Approved Document A: Structure. A3 – Disproportionate Collapse. NBS, RIBA Enterprises Ltd, 2004 edition incorporating 2004 amendments.

Secretary of State for Communities and Local Government (2006). Building Act 1984 – Section 16(10)(a). Determination of compliance with Requirement A3 (‘Disproportionate Collapse’) of the Building Regulations 2000 (as amended) in respect of building work to add a storey and lift shaft to an existing four storey block of flats. Ref. 45/1/225. Department for Communities and Local Government, 15 November.

Secretary of State for Communities and Local Government (2008). Determination of compliance with Requirement A3 (Disproportionate Collapse) in part A (Structure) of Schedule 1 to the Building Regulations 2000 (as amended), in respect of mezzanine rooms within the roof space of a new block of 28 flats/maisonettes. Ref. 45/1/232. Department for Communities and Local Government, 17 March.

Way, A. G. J. (2005). Guidance on Meeting the Robustness Requirements in Approved Document A (2004 edition). SCI Publication P341. Ascot: Steel Construction Institute.

Related Standards and Codes

BSI (1985). The Structural Use of Steelwork in Building. Part 1: Code of Practice for Design in Simple and Continuous Construction: Hot-Rolled Sections. London: BSI, BS 5950-1:1985.

BSI (1997). Structural Use of Concrete. Part 1: Code of Practice for Design and Construction. London: BSI, BS 8110-1:1997 incorp­orating Amendment Nos. 1, 2, 3 and 4, 15 March 1997.

BSI (2000). Structural Use of Steelwork in Building. Part 1: Code of Practice for Design – Rolled and Welded Sections. London: BSI, BS 5950-1:2000 incorporating corrigenda Nos. 1 and 2 and Amendment No. 1, 15 May 2001.

BSI (2002a). Structural Use of Timber. Part 2: Code of Practice for Permissible Stress Design. London: BSI, BS 5268-2:2002 incorp­orating Amendment No. 1, 14 March 2002 and 2007.

BSI (2002b). Eurocode: Basis of Structural Design. London: BSI, BS EN1990:2002+A1:2005 incorporating corrigenda December 2008 and April 2010, 27 July 2002 and 2010.

BSI (2004). Eurocode 2: Design of Concrete Structures. Part 1–1: General Rules and Rules for Buildings. London: BSI, BS EN1992-1-1:2004 incorporating corrigendum January 2008, 23 December 2004 and 30 June 2008.

BSI (2005a). Code of Practice for the Use of Masonry. Part 1: Structural Use of Unreinforced Masonry. London: BSI, BS 5628-1:2005 incorporating corrigendum No. 1, 8 December 2005 and 2009.

BSI (2005b). UK National Annex for Eurocode: Basis of Structural Design. London: BSI, NA to BS EN1990:2002+A1:2005 incorpor­ating National Amendment No. 1, 15 December 2004 and 2009.

BSI (2006). Eurocode 1: Actions on Structures. Part 1–7: General Actions: Accidental Actions. London: BSI, BS EN1991-1-7:2006.

BSI (2010). Background paper to the National Annexes to BS EN1992-1 and BS EN1992-3. London: BSI, PD 6687-1:2010.