Centergy Parking Deck Collapse
Atlanta, GA - June 2009
Ben Borden, Penn State University B.A.E. 2013

Figure 1: View of collapsed bay from fifth floor. Photo Courtesy of Dr. Jochen Teizer


During lunchtime on June 29, 2009 four floors of a bay which was part of a six story precast concrete parking garage collapsed destroying 45 cars (building can be seen on its site in Figure 2). The Garage received its occupancy certificate in 2004 and left no cause for suspicion, as it had not received any complaints. Garage use was mainly for occupants of the Centergy building, but spaces were also leased to a number of tenants including some employees of Georgia Tech, which resides across the street. After a spandrel beam became dislodged, the floors collapsed one on top the other until they rested in the basement. Fortunately there was no one injured in the collapse.

Figure 2: Arial View of Centergy Parking Deck After Partial Collapse. Photo courtesy of Dr. Jochen Teizer.

Key Words

Precast, Parking Deck, Siding Connections, Progressive Collapse, Construction Error

Events Leading Up to the Collapse

After receiving an inspection, a building permit was issued for the Centergy parking deck in December of 2001. (Stafford, 2009) The parking deck was opened in 2004 and no subsequent inspections were performed. Common practice in the city of Atlanta is that additional inspections are performed only when a complaint or request has been filed. Rarely do garages collapse after construction is completed except when they begin to near the age to which they were designed to last. Most garages that collapse do so during construction or a numbers of years later due to how the effects of corrosion weaken the structure. However, the Centergy Parking Deck was only 5 years old. It appeared to be in good condition and there was no cause for subsequent inspections.

Figure 4: Spandrel beams lying on grass adjacent to the collapsed portion of the garage. Photo courtesy of Dr. Jochen Teizar.

Figure 3: Column where spandrel beam rested before collapse. Photo courtesy of Dr. Jochen Teizer.

Around lunchtime (12:30 PM) Matt Jones was driving his car through a precast concrete parking garage in Atlanta, when he saw large masses of concrete and cars falling in front of him. Leaving his car, he got out and ran to safety (Action News, 2009).

Since the incident occurred during lunchtime, there were a number of people in the garage at the time. However, no one was in the collapsed bay when it fell.

Figure 5: Ledge where t-beams rested, note neoprene pad and spalling where beam slid off the ledge Photo Courtesy of Dr. Jochen Teizer

First, the connection holding the fourth floor spandrel beam to the column (shown in Figure 4) broke. This caused the beam to slide away from the column. The beam moved away from the garage far enough that the T-beams composing the main floor of the parking deck were bearing on a very thin edge on the ledge of the beam. The concrete of the ledge shown in Figure 5 spalled and as the t-beam fell it pushed the spandrel off of the structure (Figure 4 shows the detached spandrel lying on the ground). The fourth floor of the deck fell on top of the floor beneath. The weight of the falling floor overwhelmed the capacity of the floor beneath, and initiated a progressive collapse of the entire bay beneath, which was only stopped when the floors had reached the
basement (see Figure 1) (Engineering News Record, 2009).

Investigation & Cause of Failure

David Tyndall, who was head of the company who owned the parking deck reported that an exterior beam extending from one column to another had popped out. (Yang, 2009) Observers suspected that it was a factor which led to the collapse. However, the reason for the beam's movement took somewhat longer to determine.

While they waited for engineers to determine the cause of the collapse, citizens speculated as to what they believed was the cause. There were initially two main theories as to what caused the spandrel beam to become dislodged:

One theory was that the secondary bolts for resisting torsion in the spandrel were inadequate to resist the forces imposed on them by the floor. (Boone, 2009)
Another theory was that bolts connecting the spandrel to the column were installed improperly. (Insurance Journal, 2009)
Figure 6: End of spandral beam which attaches to column shown in figure 3. Photo courtesy of Dr. Jochen Teizer.

Figure 7: Closeup of bolt which appears to have been sheared off. Photo courtesy of Dr. Jochen Teizer.

Ultimately engineers determined that a threaded rod supporting the exterior wall had failed (Bailey, 2010). The joint which failed has a slotted connection to allow for thermal expansion of the structure (See figures 6 and 7). However, the way in which the joint was assembled did not allow the joint to move as was intended. Metromont's (the precast manufacturer for the Centergy parking deck) website has details for all of their current typical precast connections which shows similar connection from 2011 exhibiting identical precast components with only the absence of the nut seen in the figure 7 above. The thread rod was therefore not able to resist the forces because of its improper installation and appears to have sheared off. The large hole in the column shown in figure 3 is where the rest of the thread rod should be. This picture was taken within days after the collapse, the rust on the sheared end of the bolt suggests that it may have already failed prior to the collapse and the beam eventually popped off.

In order to allow for expansion the t-beams are not welded to the spandrel but rest on its ledge separated by a neoprene pad (shown in Figure 5). When the Spandrel slid out far enough the t-beams dropped off of the ledge and onto the floor below. Because the fourth floor was not designed for the impact of the floor above, the weight of the falling floor overwhelmed the subsequent floors beneath until the entire bay beneath had collapsed.


The designers, builders, and owners of Centergy had several opportunities to protect their parking deck from its unfortunate outcome. A highly discussed issue after the collapse, was the lack of inspections performed on the garage after its construction. The parking deck had been inspected for issuance of a permit for the entrance in August of 2002. It had not been inspected since since it was complete. According to Catherine Woodling, a spokeswoman for Atlanta's presiding mayor, this was consistent with the standard practice of the city of Atlanta at the time of the collapse (Stafford, 2009). A routine inspection, though not historically performed after construction without a complaint, may have caught the problem before it resulted in a partial collapse.

Maintenance should also consist of inspections throughout the life of the structure. The Precast Concrete Institute distributes a manual for the maintenance of precast parking structures which they created in 2004, long before the Centergy incident (Korkosz et al., 2004). Their manual suggests inspection of several structural components including the joints. Had the owners of Centergy performed these recommended maintenance inspections, they would have had several additional chances to avert a disaster.

The problem with the garage was caused by a flawed installation of the bolts holding the spandrel beam. Construction professionals could trained to better understand the dynamics of the structure and hopefully be able to speak up when things are not making sense.

Lessons Learned

The Centergy collapse did not change any codes or permanently mandate inspections, however it was cited as impetus for better and more frequent inspections in the International Code Council's building safety Journal (Bailey, 2010).

Many of the parking decks throughout the city of Atlanta were given inspections as direct result of the collapse. Two days after the collapse the Georgia Institute of Technology announced it was going to reinspect all of its garages. (Matteucci, 2009). The Classic Center parking facility in Athens GA was shut down for inspections due to suspicion of flaws with the connections similar to that of the Centergy deck. The deck was also constructed by Metromont, the same precast contractor who worked on Centergy(Shearer, 2009). This inspection revealed that a different contractor's failure to mix air bubbles into the concrete resulted in tiny cracks in sections due to thermal expansion and contraction (Aued, 2011).

Because of nondisclosure agreements (Garner, 2011) it is difficult to find specific details of what happened, and therefore many lessons that could have been learned may not be unless someone who was involved with the investigation is able to share the results.

Related Cases

Many of the reports which surfaced immediately after the collapse, noted that Harden Construction, who had been the General Contractor on the construction of the garage, had recently been fined by the Occupational Safety and Health Administration after the collapse of the Botanical Garden Elevated Walkway in 2008, which resulted in 18 injuries and 1 fatality. The two cases were related because they were both collapses and because Harden was the GC. Harden was rehired for the repair in both cases (Stafford, 2009).

The David L. Lawrence Convention Center also failed due to malfunctioning sliding connections. Unlike Centergy however, the Convention Center was in a much colder region and the resulting change in temperature was a factor that contributed to the failure. The Convention center connections were also constructed using steel.

Recently the Woodridge Crossing Parking Deck collapsed in Woodridge, NJ. Woodridge is still under investigation however, pictures reveal that the spandrel beam is laying next to the structure and a progressive collapse was initiated similar to Centergy.


The Centergy parking deck was rebuilt at a cost of about $1,000,000 and open just over six months after the collapse. This time they took extra precautions to ensure the safety of the structure. Each similar joint was strengthened, and every bolt and weld was double checked. A third engineering firm was brought in to help evaluate the soundness of the structure and its completed repairs. According to Tyndall the deck is likely the safest parking deck in the country because of the scrutiny they went through to ensure its safety (Mayerle, 2010).

Litigitation between a number of insurance companies continued through June of 2011 (Garner, 2011)

Wally Bailey's conclusion sums it up pretty well:
"We should recognize that parking decks present unique structural safety issues we need to address through maintenance programs and/or regular inspections. Periodic structural safety inspections are as important as periodic life safety inspections. We give immediate attention to life safety deficiencies (faulty exit lights, fire extinguishers or door hardware), but protection of property and life safety are at risk when structural deficiencies are not identified and repaired, particularly in precast concrete parking structures. We must diligently insure that structural integrity is maintained in all buildings." (Bailey, 2010)

Hopefully we will follow Bailey's advice, remember Centergy and provide proper inspections to structures to help ensure their safety.


Action News (2009) "Partial Parking Deck Collapse In Midtown Atlanta." <> (Oct. 4, 2012)

  • This contains a statement from Matt Jones who abandoned his car after he saw the building collapsing in front of him.

Aued, B. (2011). "Downtown deck repair costs double." Online Athens, <> (Dec. 20 2012).

  • An article concerning repairs discovered because inspections resulting from Centergy deck collapse.

Bailey, W. (2010) "An Ounce of Prevention Can Keep a Parking Structure Intact" building safety journal, <> (Oct. 4, 2012).
  • A comparison of the circumstances of the Centergy Collapse to another garage which narrowly avoided collapse due to a similar issue.

Boone, C (2009) "Bolt welds may have caused Midtown parking deck collapse." The Atlanta Journal-Constitution, <>

  • This article is a premature statement that secondary bolt welds were the cause for the failure of the spandrel beam.

Engineering News-Record (2009) “Officials Are Investigating Partial Collapse of Atlanta Parking Deck” <> (Oct. 4, 2012).
  • A short synopsis of the Centergy Collapse including a description of progressive collapse
  • Cost and square footage values from project

Garner, M. K.(2011) "Lawsuits to close Centergy parking deck collapse Sega." The Atlanta Journal-Constitution, <> (Oct. 4, 2012).

  • Information on litigation and mention of a nondisclosure agreement.

Insurance Journal. (2009) "Georgia Garage Collapse Blamed on Improper Bolt." <> (Dec. 5, 2012).

  • Centergy's owner claims the collapse was caused by the improper use of the bolt connecting the spandrel to the column.

Korkosz, W. et al. (2004) "Maintenance Manual for Precast Parking Structures" PCI Committee on Parking Structures." <> (Dec. 5, 2012).

  • This Parking Structure Maintenance Manual was drawn up by the Precast Concrete Institute in 2004. Though it is not officially adopted by the city of Atlanta, it does acknowledge the need for periodic inspections of connections.

Mayerle, J. (2010). "Centergy Parking Deck Opens 6 Months After Collapse." CBS Atlanta News.
<> (Oct. 4, 2012).
  • This news excerpt shows the collapse, gives a brief description of the cause of collapse and interviews of people who use the building after the garage has been reopened.

Metromont (2011). "Engineering Details." <> (Dec. 5, 2012).

  • Metromont was the precast manufacturer for Centergy. This page links to Engineering Details for various precast connections.

Shearer, L. (2009). "Deck closed after inspection." Online Athens, <> (Dec. 5 2012).

  • Centergy collapse resulted in several inspections of parking decks in the city of Atlanta. This one was closed while more extensive inspections could be performed.

Stafford, L. (2009). "Garage builder was fined in deadly walkway collapse." Atlanta Journal Constitution, <> (Dec. 5, 2012).

  • This article contains a statement from Catherine Woodling, a spokeswoman for the mayor, concerning the inspections that the Centergy deck had received in context with the cities general inspection policy.

Teizer, Dr. J. (2009). "Analysis of the Partial Collapse of the Parking Garage on Spring Street, Atlanta, GA." RAPIDS Laboratory, <> (Dec. 5, 2012).

  • Real-time Automated Project Information and Decision Systems (RAPIDS) Laboratory created an exercise for students concerning the collapsed parking garage. Dr. Teizer took very detailed photographs of the failed components from the garage. For more photographs and the pertaining exercise visit their site above.
  • RAPIDS Laboratory concentrates on real-time pro-active safety warning and alert technologies, equipment blind spot measurement, operator visibility tracking, wireless 3D real-time resource location tracking, 4D (building) information modeling and processing, site layout management, and an inference management framework with major focus on real-time pro-active safety, health, and work activity monitoring and sampling.

Yang, Z. (2009) "Centergy Parking Deck Collapses, No Injuries." Technique [Georgia Tech]. 95(4), 1, <> (Oct. 4, 2012).

  • This article describes in more detail the collapse and the circumstances surrounding it.

Additional Resources and References

Georgia Department of Community Affairs Planning and Environmental Management Division (2007) "Georgia State Amendments to the International Building Code (2006 Edition)." <> (Dec. 5, 2012).

  • This is the Current Georgia Building Code which was first required on January 1, 2007, after construction but two and a half years before the collapse.

Mercan, B. (2011). "Modeling and behavior of prestressed concrete spandrel beams." University of Minnesota. ProQuest Dissertations and Theses, 194.<>

  • Dessertation on the effects of lateral forces on prestressed spandrel beams similiar to the one which caused the Centergy collapse.

Prestressed Concrete Institute (1999). PCI Design Handbook. Fifth Edition, Chicago, Illinois.
  • Design guide detailing prestressed concrete design. This guide is now obsolete and has been removed from availability. Current design guides must be referenced.