Building Failures: Avoiding the Mistakes of Yesterday Leads to the Successes of Today

Most readers of the Journal of Architectural Engineering have investigated or performed remedial design for at least one architectural or engineering system failure during their careers. Other practitioners—particularly those who work for firms specializing in investigations, forensics, disaster response, and repair—are more familiar with the variety and extent of building failures as they assist their clients in restoring damaged or deficient buildings.

Your viewpoint of the frequency of the occurrence of failures in our industry likely depends on your definition of a failure. Some would argue that the term building failure applies only to major structural collapses, while others believe that a failure should be defined as any building system or component that does not perform as intended. The latter definition would include myriad building performance problems associated with roofs, facades, mechanical, electrical, and plumbing (MEP) system balancing and control, and so on. From that perspective, a noninclusive list of structural and architectural/building system performance failures consists of the following:

Failures do not discriminate or target specific projects. Construction projects, large or small, can experience failures of a variety of types. The smallest of items can lead to a catastrophic event. Such was the case in July 2006 when the failure of a grouted anchor caused the collapse of a concrete ceiling panel on the Interstate 90 connector tunnel (Boston’s Big Dig), resulting in one fatality. A single sliding connection along an expansion joint resulted in damage claims of approximately $4 million as the result of a partial floor collapse at the David L. Lawrence Convention Center (DLLCC) in Pittsburgh in February 2007.

The cause of the collapse of the I-35W bridge in Minneapolis in August 2007 has been associated with what various investigators have described as undersized gusset plates. If initial reports hold true and no other contributing factors are discovered, the collapse will go down in history as being caused by a single factor. On the other hand, if past failures and collapses are any indicator, the gusset plate may be found to be the triggering mechanism whereas a number of contributing factors will be discovered. Although any one item may be a deficiency, most failures, including the Big Dig ceiling panel and the DLLCC sliding connection, are the result of a combination of factors or events.

One of the most widespread categories of contributors to failures of all types remains human error, that is, the result of a communication or procedural problem. Often referred to as process errors, these problems receive considerable attention after a collapse when the media reveal the cause and place blame. Those who study failures as a means of preventing problems in their own designs recognize that although process errors are not the trigger mechanism of a collapse, they are often responsible for the conditions and events leading up to the failure.

The causes of the 2007 DLLCC partial floor collapse being put forth by investigators and reported to date include problems with the design, fabrication, and installation of a sliding beam connection. Arguments being made relative to the DLLCC in the media by various involved parties sound eerily similar to accounts of the problems associated with shop drawings and connection design that surfaced with the Kansas City Hyatt Regency Walkway collapse in 1981 and, before that, the design and construction responsibility and execution confusion associated with the collapse of 2000 Commonwealth Avenue in Boston in 1971.

The purpose of this editorial is not to solicit failure case history investigations and studies for the JAE. ASCE already has a top-ranked publication edited by Kenneth L. Carper, the Journal of Performance of Constructed Facilities, for that purpose. It is hoped that this editorial will remind the AEC community that the starting point for preventing failures lies with all of us through better technology, coordination, and most important, communication of all types. It is those stories of teamwork, design integration, and coordination at all phases of a project that are the purview of the JAE. I invite anyone who is interested in telling their success stories in the form of a Forum article to submit a manuscript, call, or contact me at [email protected] to discuss ideas in more detail.