«R. SHANKAR NAIR R. Shankar Nair R. Shankar Nair, Ph.D., P.E., S.E. is a principal and senior vice president of Teng & Associates, Inc. in Chicago. In ...»
PROGRESSIVE COLLAPSE BASICS
R. SHANKAR NAIR
R. Shankar Nair
R. Shankar Nair, Ph.D., P.E., S.E. is a principal and senior vice president of Teng & Associates,
Inc. in Chicago. In a career that has focused on structural design of large architectural and civil
engineering projects, he has developed the structural concepts for numerous tall buildings and
many major bridges. This work has received numerous awards, including four AISC/NSBA “Prize Bridge” awards and six Structural Engineers Association of Illinois annual “Most Innovative Structure” awards. Also active as a researcher and leader of professional activities, he has served as chairman of the Council on Tall Buildings and Urban Habitat and is, at present, a member of the AISC Specification Committee and chairman of its Stability Task Committee.
ABSTRACTProgressive collapse is the collapse of all or a large part of a structure precipitated by damage or failure of a relatively small part of it. The phenomenon is of particular concern since progressive collapse is often (though not always) disproportionate, i.e., the collapse is out of proportion to the event that triggers it. Thus, in structures susceptible to progressive collapse, small events can have catastrophic consequences.
After the progressive and disproportionate collapse of the Ronan Point apartment tower in England in 1968, prevention of progressive collapse became one of the unchallenged imperatives in structural engineering, and code-writing bodies and governmental user agencies attempted to develop design guidelines and criteria that would reduce or eliminate the susceptibility of buildings to this form of failure. These efforts tended to focus on improving redundancy and alternate load paths, to ensure that loss of any single component would not lead to a general collapse. But in fact, redundancy is only one of the ways of reducing susceptibility to disproportionate collapse. Improved local resistance for critical components and improved continuity and interconnection throughout the structure (which can improve both redundancy and local resistance) can be more effective than increased redundancy in many instances. Through an appropriate combination of improved redundancy, local resistance and interconnection, it should be possible to greatly reduce the susceptibility of buildings to disproportionate collapse.
INTRODUCTIONOn the morning of 16 May 1968, Mrs. Ivy Hodge, a tenant on the 18th floor of the 22-story Ronan Point apartment tower in Newham, east London, struck a match in her kitchen. The match set off a gas explosion that knocked out load-bearing precast concrete panels near the corner of the building. The loss of support at the 18th floor caused the floors above to collapse.
The impact of these collapsing floors set off a chain reaction of collapses all the way to the ground. The ultimate result can be seen in Figure 1: the corner bay of the building has collapsed from top to bottom. Mrs. Hodge survived but four others died.
While the failure of the Ronan Point structure was not one of the larger building disasters of recent years, it was particularly shocking in that the magnitude of the collapse was completely out of proportion to the triggering event. This type of sequential, one-thing-leading-to-another failure was labeled “progressive collapse” and the engineering community and public regulatory agencies resolved to change the practice of building design to prevent the recurrence of such tragedies.
PROGRESSIVE COLLAPSE AND DISPROPORTIONATE COLLAPSE
Progressive collapse can be defined as collapse of all or a large part of a structure precipitated by failure or damage of a relatively small part of it. The General Services Administration (GSA, 2003b) offers a somewhat more specific description of the phenomenon: “Progressive collapse is a situation where local failure of a primary structural component leads to the collapse of adjoining members which, in turn, leads to additional collapse.” It has also been suggested that the degree of “progressivity” in a collapse be defined as the ratio of total collapsed area or volume to the area or volume damaged or destroyed directly by the triggering event. In the case of the Ronan Point collapse, this ratio was of the order of 20.
By any definition, the Ronan Point disaster would qualify as a progressive collapse. In addition to being progressive, the Ronan Point collapse was disproportionate. A corner of a 22-story building collapsed over its entire height as a result of a fairly modest explosion, an explosion that did not take the life of a person within a few feet of it. The scale of the collapse was clearly disproportionate to the cause.
While the Ronan Point collapse was clearly both progressive and disproportionate, it is instructive to examine other collapses in the same light.
Murrah Federal Office Building The Murrah Federal Office Building in Oklahoma City was destroyed by a bomb on 19 April
1995. The bomb, in a truck at the base of the building, destroyed or badly damaged three columns. Loss of support from these columns led to failure of a transfer girder. Failure of the transfer girder caused the collapse of columns supported by the girder and floor areas support ed by those columns. The result was the general collapse evident in Figure 2.
Fig. 2. Murrah Federal Office Building after 19 April 1995 attack
The Murrah Building disaster clearly was a progressive collapse by a ll the definitions of that term. Collapse of a large part of the building was precipitated by destruction of a small part of it (a few columns). The collapse also involved a clear sequence or progression of events: column destruction; transfer girder failure; collapse of structure above.
But was the Murrah Building collapse disproportional? The answer is not nearly as clear as in the case of the Ronan Point collapse. The Murrah collapse was large. But the cause of the collapse, the bomb, was very large too, large enough to cause damage over an area of several city blocks.
Ultimately, we must judge the Murrah Building collapse “possibly disproportional” only in the sense that we know now that with some fairly modest changes in the structural design (as will be discussed), the damage from the bomb might have been significantly reduced.
World Trade Center 1 and 2
Each of the twin towers of World Trade Center 1 and 2 collapsed on 11 September 2001 following this sequence of events: A Boeing 767 jetliner crashed into the tower at high speed;
the crash caused structural damage at and near the point of impact and also set off an intense fire within the building (see Fig. 3); the structure near the impact zone lost its ability to support the load above it as a result of some combination of impact damage and fire damage; the structure above collapsed, having lost its support; the weight and impact of the collapsing upper part of the tower caused a progression of failures extending downward all the way to the ground.
Fig. 3. World Trade Center 1 and 2 on 11 September 2001
Clearly, this was a “progressive collapse” by any definition. But it cannot be labeled a “disproportionate collapse.” It was a very large collapse caused by a very large impact and fire.
And unlike the case with the Murrah Building, simple changes in the structural design that might have greatly reduced the scale of the collapse have not yet been identified.
Observations on “Progressive” and “Disproportionate” Collapse
Prevention of progressive collapse is generally acknowledged to be an imperative in structural engineering today. But in fact, virtually all collapses could be regarded as “progressive” in one way or another, and a building’s susceptibility to progressive collapse should be of particular concern only if the collapse is also disproportionate. Indeed, the engineering imperative should be not the prevention of progressive collapse but the prevention of disproportionate collapse (be it progressive or not).
CODES AND STANDARDS
Since the progressive collapse of the Ronan Point apartment tower in 1968, many codes and standards have attempted to address the issue of this type of collapse. A complete survey of these efforts is beyond the scope of this paper, but a small sampling of current and recent provisions related to progressive collapse will provide an indication of the alternative approaches being considered and the direction in which these efforts appear to be evolving.
The American Society of Civil Engineers Minimum Design Loads for Buildings and Other
Structures (ASCE, 2002) has a section on “general structural integrity” that reads thus:
“Buildings and other structures shall be designed to sustain local damage with the structural system as a whole remaining stable and not being damaged to an extent disproportionate to the original local damage. This shall be achieved through an arrangement of the structural elements that provides stability to the entire structural system by transferring loads from any locally damaged region to adjacent regions capable of resisting those loads without collapse. This shall be accomplished by providing sufficient continuity, redundancy, or energy-dissipating capacity (ductility), or a combination thereof, in the members of the structure.” Clearly, the focus in the ASCE standard is on redundancy and alternate load paths over all other means of avoiding susceptibility to disproportionate collapse. But the degree of redundancy is not specified, and the requirements are entirely threat-independent.
ACI 318-02 The American Concrete Institute Building Code Requirements for Structural Concrete (ACI,
2002) include extensive “Requirements for structural integrity” in the chapter on reinforcing steel details. Though the Commentary states that it “is the intent of this section … to improve … redundancy” there is no explicit mention of redundancy or alternate load paths in the Code. The Code provisions include a general statement that “In the detailing of reinforcement and connections, members of a structure shall be effectively tied together to improve integrity of the overall structure” and many specific prescriptive requirements for continuity of reinforcing steel and interconnection of components. There are additional requirements for the tying together of precast structural components. None of the ACI provisions are threat-specific in any way.
GSA PBS Facilities Standards 2000
The 2000 edition of the GSA’s Facilities Standards for the Public Buildings Service (GSA,
2000) included the following statement under the “Progressive Collapse” heading in the “Structural Considerations” section: “The structure must be able to sustain local damage without destabilizing the whole structure. The failure of a beam, slab, or column shall not result in failure of the structural system below, above, or in adjacent bays. In the case of column failure, damage in the beams and girders above the column shall be limited to large deflections.
Collapse of floors or roofs must not be permitted.” This is an absolute and unequivocal requirement for one-member (beam, slab, or column) redundancy, unrelated to the degree of vulnerability of the member or the level of threat to the structure.
GSA PBS Facilities Standards 2003 The 2003 edition of the GSA’s Facilities Standards for the Public Buildings Service (GSA, 2003a) retained the “Progressive Collapse” heading from the 2000 edition, but replaced all of the words reproduced above with this short statement: “Refer to Chapter 8: Security Design.” The structural provisions in Chapter 8 apply only to buildings deemed to be at risk of blast attack. For such buildings, the chapter provides general performance guidelines and references to various technical manuals for study of blast effects. This represents a complete change of approach from the 2000 version of the same document.
GSA Progressive Collapse Guidelines 2003
The GSA Progressive Collapse Analysis and Design Guidelines for New Federal Office Buildings and Major Modernization Projects (GSA, 2003b) begins with a process for determining whether a building is exempt from progressive collapse considerations. Exemption is based on the type and size of the structure (for instance, any building of over ten stories is nonexempt) and is unrelated to the level of threat. Typical non-exempt buildings in steel or concrete have to be shown by analysis to be able to tolerate removal of one column or one 30 -ft length of bearing wall without collapse. Considerable detail is provided regarding the features of the analysis and the acceptance criteria.
In some ways, these guidelines appear to be a throw-back to the GSA’s PBS Facilities Standards of 2000 in that their central provision is a requirement for one-member redundancy, unrelated to the degree of vulnerability of the member or the level of threat to the structure.
METHODS OF PREVENTING DISPROPORTIONATE COLLAPSEThere are, in general, three alternative approaches to designing structures to reduce their
susceptibility to disproportionate collapse:
• Redundancy or alternate load paths
• Local resistance
• Interconnection or continuity Redundancy or Alternate Load Paths In this approach, the structure is designed such that if any one component fails, alternate paths are available for the load in that component and a general collapse does not occur. This approach has the benefit of simplicity and directness. In its most common application, design for redundancy requires that a building structure be able to tolerate loss of any one column wi thout collapse. This is an objective, easily-understood performance requirement.
The problem with the redundancy approach, as typically practiced, is that it does not account for differences in vulnerability. Clearly, one-column redundancy when each column is a W8x35 does not provide the same level of safety as when each column is a 2000 lb/ft built -up section.