Stamped Doesn’t Mean Safe: Lessons from the Hyatt Regency Collapse

[jar546]

In 1981, 114 lives were lost and more than 200 people were injured when two suspended walkways collapsed at the Hyatt Regency Hotel in Kansas City. The case remains one of the most catastrophic engineering failures in U.S. history — and one of the clearest examples of what happens when assumptions, poor communication, and unchecked revisions go unchallenged.

In a recent video case study, former Missouri Court of Appeals Judge Paul Spinden describes the ethical breakdown that led to this tragedy. The transcript of that video, which I’ve reviewed in full, is a sobering reminder that a professional seal on a set of plans is not a get-out-of-jail-free card for anyone, especially not for Building Code Officials.

What Went Wrong​

Jack Gillum and Dan Duncan, the structural engineers in the Hyatt project, stamped the drawings. However, the engineer gave verbal approval without follow-up calculations when the fabricator proposed changing the hanger rod configuration. That single design change doubled the load on the connection points and was never analyzed or documented. No one did the math. The bridge failed a year after construction, but it was doomed the moment those changes were approved.

The drawings were “sealed,” and yet they were fatally flawed.

Why It Matters to BCOs​

As Building Officials, we aren’t structural engineers. Most of us can't legally re-run calculations or redesign a structure. But that doesn’t mean we blindly accept what’s put in front of us just because it has an engineer’s or architect’s seal.

• We can and should ask for calculations.
• We can request clarification when connections, loads, or alterations raise red flags.
• We can require peer review for unusual or overly complex structural elements.

Plans get value engineered, altered in the field, or changed by shop drawings all the time. And too often, it’s done without the proper communication and oversight.

The Hyatt case showed what happens when nobody follows through.

Bottom Line​

Stamped doesn’t mean safe. As Building Officials, we're the last line of defense. We owe it to the public to speak up, ask questions, and — when needed — push back. Because the next failure won’t be theoretical. It will be on the evening news.

 

[e hilton]

I remember when that happened. Even if they had provided the calcs, how many BCO’s would have the knowledge to understand them? How many would accept the responsibility “i reviewed the calcs and i agree with them”? The EOR signed ff on the change … normally that’s all you need to see.

 

[Yankee Chronicler]

In 1981, 114 lives were lost and more than 200 people were injured when two suspended walkways collapsed at the Hyatt Regency Hotel in Kansas City. The case remains one of the most catastrophic engineering failures in U.S. history — and one of the clearest examples of what happens when assumptions, poor communication, and unchecked revisions go unchallenged.

In a recent video case study, former Missouri Court of Appeals Judge Paul Spinden describes the ethical breakdown that led to this tragedy. The transcript of that video, which I’ve reviewed in full, is a sobering reminder that a professional seal on a set of plans is not a get-out-of-jail-free card for anyone, especially not for Building Code Officials.

What Went Wrong​

Jack Gillum and Dan Duncan, the structural engineers in the Hyatt project, stamped the drawings. However, the engineer gave verbal approval without follow-up calculations when the fabricator proposed changing the hanger rod configuration. That single design change doubled the load on the connection points and was never analyzed or documented. No one did the math. The bridge failed a year after construction, but it was doomed the moment those changes were approved.

The drawings were “sealed,” and yet they were fatally flawed

Actually, the engineer's design drawings were NOT flawed at all. It's more complicated than that.

Once we approve a set of construction documents, a boatload of things get re-drawn (often in considerably more detail) as shop drawings, by the subcontractors who will actually be fabricating and installing various elements and systems. Other than sprinkler layout drawings and hydraulic calculations, in my experience it's virtually unheard of for shop drawings to be submitted to the building department for review and approval. The reality is, most of us don't have enough time to review the original construction plans and keep up with the inspections we have to perform. There's no time and no staff to be reviewing shop drawings.

That video touches on the cause, but the video is aimed more at the legal aspects rather than the engineering aspects. And the reality is that this collapse completely changed the landscape for how architects and engineers review shop drawings, because prior to the Hyatt Regency collapse the standard of the professions was that review of shop drawings was only for adherence to the design concept. "Approval" of shop drawings (and architects and engineers had stopped using the word "approved" back in the late 1970s of early 1980s) was only for adherence to the design concept, not for adequacy on engineering. The Hyatt Regency changed that.

However, with that said, I disagree with jar546's conclusion: "... a professional seal on a set of plans is not a get-out-of-jail-free card for anyone, especially not for Building Code Officials."

Nothing in that video touched on building officials. I don't think there's any building department anywhere in the country that has on staff licensed structural engineers who rigorously review the calculations for every connection and every detail of a building's structure. And even if they did -- the original design wasn't flawed. If anyone had reviewed the original structural design, it would have passed the review. The problem was created by the change initiated by the steel fabricator at the shop drawing stage. The result of the change was that the load at the points where the upper bridge connected to the steel rod hangers was doubled. The court ruling was that the engineer-of-record had a duty to check that on the shop drawings, and that they failed to do so.

But shop drawings typically aren't submitted to the building department for review and approval. So I don't see any way that this collapse serves as a cautionary tale for building officials.

 

[Yankee Chronicler]

I remember when that happened. Even if they had provided the calcs, how many BCO’s would have the knowledge to understand them? How many would accept the responsibility “i reviewed the calcs and i agree with them”? The EOR signed ff on the change … normally that’s all you need to see.

And I think that's all the law requires us to do. It isn't our job to replicate the engineer's work. Our job is to make a reasonable attempt to determine that the engineer has done his job. Part of that is by verifying that they used the correct codes, that they say they designed for the minimum live loads prescribed by the code, and that they put their seal and signature on the drawings. As an architect, I understand that ALL the seal says is that the drawings were prepared by the engineer or under his/her direct supervision -- the seal is NOT a guarantee that the design meets code -- but it's astonish how many design professionals don't even want to put their seal on the drawings at all.

 

[jar546]

However, with that said, I disagree with jar546's conclusion: "... a professional seal on a set of plans is not a get-out-of-jail-free card for anyone, especially not for Building Code Officials."

The article uses the Hyatt tragedy as an intro to the building department, which was not even mentioned in the video. One of the main reasons for the article, and I have written similar, was to wake up those who pass through plans just because they are signed and sealed by a licensed design professional.

Always ask for the calculations. You will find that in many cases, they were never done. The more resistance from the RDP, the more likely it was that the calculations were never performed. This is especially true when architects perform 'incidental' engineering on their drawings.

So in this case, drawings were stamped, but calculations were never performed. Ask for them.

 

[jar546]

I remember when that happened. Even if they had provided the calcs, how many BCO’s would have the knowledge to understand them? How many would accept the responsibility “i reviewed the calcs and i agree with them”? The EOR signed ff on the change … normally that’s all you need to see.

The calculations were never done after the changes. Asking for them would require them to actually do the calculations.

 

[ICE]

Background information from Grok as to what exactly caused the failure:

The collapse of the walkway at the Hyatt Regency Hotel in Kansas City, Missouri, occurred on July 17, 1981, and was one of the deadliest structural failures in U.S. history. The disaster resulted from a critical design flaw in the hotel's suspended walkways, which were part of the atrium.

The second- and fourth-floor walkways were supported by a single set of steel rods connected to the ceiling, with the walkways hanging one above the other. The original design called for continuous threaded rods to extend from the ceiling through both walkways, with each walkway supported by nuts and washers at different heights. However, during construction, this design was altered for practical reasons. Instead of using continuous rods, the contractor and engineers modified the system to use two separate sets of rods: one set extending from the ceiling to the fourth-floor walkway, and a second set from the fourth-floor walkway to the second-floor walkway. This change effectively doubled the load on the connection at the fourth-floor walkway, as it now had to support both its own weight and the weight of the second-floor walkway below it.

The critical failure occurred at the connection points where the rods were attached to the box beams of the fourth-floor walkway. These beams, made of welded steel channels, were not adequately designed to handle the increased load. On the night of the collapse, a tea dance event drew a large crowd, with people standing on both walkways. The added weight caused the box beam connections to deform and fail, pulling the rods through the beams. This led to the fourth-floor walkway collapsing onto the second-floor walkway, which then fell to the lobby floor below, killing 114 people and injuring over 200.

Investigations, including by the National Bureau of Standards (now NIST), concluded that the design change violated basic engineering principles and building codes. The original design was already marginal, but the modification made it structurally unsound, with the connections unable to support even the walkways' own weight under full load, let alone additional live loads from occupants. Poor communication between the structural engineer, fabricator, and contractor, along with inadequate oversight and failure to review the revised design, compounded the issue.

In summary, the collapse was caused by a flawed redesign of the walkway support system, shifting from a single continuous rod to a segmented system, which overloaded the fourth-floor walkway’s connections beyond their capacity.

Comment:
The original design that had two walkways supported by steel rods just seems wrong. The value engineering that produced that should have required a 200% safety factor at every connection.

When I was assigned to the So, Whittier District office there was a local Structural Engineer that was close to ninety years old. He was in the business of selling his stamp. Some of the strangest thing came over the counter.

An acquaintance was hired at San Bernardino County as an inspector. That County plan checks few projects and if the is an architect or engineer, they just issue a permit without plan checking. He didn't last long as his ability to plan check in the field was not that strong. He could inspect to a plan but not to the code.

Being an inspector, my advice to the inspectors is that you are the last chance to raise an alarm. If you have a concern, you have a duty to speak up. Worst case scenario is that you will piss off a few engineers and that comes with the territory.

 

[Yankee Chronicler]

The calculations were never done after the changes. Asking for them would require them to actually do the calculations.

I understand that the calculations were not done after the shop drawing redesign. The question is: How often does the building department EVER receive and check shop drawings? Between architecture and being a building official, I've been at this for 52 years and I've never heard of any building department wanting to look at shop drawings.

What probably should have triggered some questions by the building department was the fact that the hanger rods for the second story walkway changed from the original (approved) design. Some inspectors might have picked up on that change when they saw the walkways. In this case, obviously they didn't pick up on it.

One other point to consider: Our town counsel (the municipality's staff lawyer) has specifically instructed us NOT to ask for calculations. "Why not?" you ask. For the simple reason that if we ask for them, it can be assumed that we will review them, and that we have someone on staff who is qualified to review them. As an architecture student, I took graduate-level structural engineering classes, but that was half a century ago. I'm not remotely comfortable looking at structures more complicated than simple beams. We had a project three years ago that we considered rather sketchy from a structural perspective. We asked the town engineer if she would review the calculations if we requested them. She declined. She said she and her staff were qualified to engineer roads and simple bridges, not buildings. In the end, we had a second licensed PE review the design -- and, as we had expected, he determined that it wasn't up to code, so revisions were made.

 

[jar546]

I've been at this for 52 years and I've never heard of any building department wanting to look at shop drawings.

The world is much bigger than the one you live in and the world I live in. I have always required shop drawings, even if deferred submittals. Your statement is concerning.

 

[Yikes]

Jar, I’m going to push back a bit. The only reason a building official should need to see a shop drawing is when some aspect of the plan check approval was deferred into the shop drawing - - I.e. design intent was not illustrated in the DPOR’s original design documents to the extent necessary to verify code compliance during plan check on the DPOR’s plans.
Trusses and elevator guide rail supports are typical in this category on my projects.
But if the DPOR showed a stell frame and its connections and provided calcs to demonstrate compliance, IMO there is no compelling need for an AHJ to see that redundant information shown on each individual beam in a shop drawing.

 

[Yankee Chronicler]

So let's take a poll: How many of the building officials on this forum routinely require that all shop drawings (beyond deferred submittals, which are a special case) be submitted for building department review and approval before fabrication may proceed?

Not only isn't that done in my state (even on State projects), it also isn't contemplated in the AIA General Conditions. The standard flow is fabricator to general contractor to architect [to engineer if for an engineering specialty and then back to the architect] back to general contractor and then back to fabricator.

 

[jar546]

I’m not saying Building Officials should be running structural calcs. But we damn sure shouldn’t be rubber-stamping plans and looking the other way when shop drawings roll in that change the design. Let’s be honest, we’ve all seen truss subs or steel fabricators submit shop drawings that don’t line up with the approved plans. If we’re not reviewing and stamping those drawings, how the hell is an inspector supposed to know what they’re looking at in the field? We approve one design, then the contractor swaps it out with a new layout buried in a fresh set of shop drawings, and suddenly the inspector’s standing there trying to make sense of something we never even looked at. That’s not just a gap in process; that’s a liability. The Hyatt disaster wasn’t just a bad design, it was a failure to connect the dots between design changes and field conditions. If we’re going to do this job right, we can’t just trust that a stamp on the front page means the rest of the details are bulletproof. Shop drawings that alter structural intent need to be reviewed and approved by the AHJ, period. Otherwise, we’re setting our inspectors up to fail and putting the public at risk.

 

[steveray]

Jar, I’m going to push back a bit. The only reason a building official should need to see a shop drawing is when some aspect of the plan check approval was deferred into the shop drawing - - I.e. design intent was not illustrated in the DPOR’s original design documents to the extent necessary to verify code compliance during plan check on the DPOR’s plans.
Trusses and elevator guide rail supports are typical in this category on my projects.
But if the DPOR showed a stell frame and its connections and provided calcs to demonstrate compliance, IMO there is no compelling need for an AHJ to see that redundant information shown on each individual beam in a shop drawing.

Agree…but…. I rarely get drawings for permit that show me enough information to build it AND that would meet code…

 

[tmurray]

I just met with one of the largest building departments in my province and they indicated that they accept an engineer-stamped design even when they know it is wrong. When I asked why they don't ask for calculations, they gave a similar push back to the one seen here, "who is going to review them?" When I have asked for calculations, many times I've just gotten a resubmission of the plans. As JAR indicates here, the calculations were never done and when I press the issue and they do the calculations, they find out that they were wrong.

Liability works different in Canada than in the US. We are a lot more exposed. Accepting something that you know is wrong is just... well... wrong.

 

[mtlogcabin]

Always ask for the calculations. You will find that in many cases, they were never done. The more resistance from the RDP, the more likely it was that the calculations were never performed.

That was the common theme taught in the SBCCI code classes I attended in the early 90's.
When we asked for them we usually got them with corrected drawings.

 

[Yankee Chronicler]

I’m not saying Building Officials should be running structural calcs. But we damn sure shouldn’t be rubber-stamping plans and looking the other way when shop drawings roll in that change the design. Let’s be honest, we’ve all seen truss subs or steel fabricators submit shop drawings that don’t line up with the approved plans. If we’re not reviewing and stamping those drawings, how the hell is an inspector supposed to know what they’re looking at in the field? We approve one design, then the contractor swaps it out with a new layout buried in a fresh set of shop drawings, and suddenly the inspector’s standing there trying to make sense of something we never even looked at. That’s not just a gap in process; that’s a liability. The Hyatt disaster wasn’t just a bad design, it was a failure to connect the dots between design changes and field conditions. If we’re going to do this job right, we can’t just trust that a stamp on the front page means the rest of the details are bulletproof. Shop drawings that alter structural intent need to be reviewed and approved by the AHJ, period. Otherwise, we’re setting our inspectors up to fail and putting the public at risk.

The proper way to handle it if a subcontractor or fabricator wants to change the design is for the RDP to submit amended construction documents, per IBC 107.4. This is different from deferred submittals, which are things that haven't yet been designed when the construction documents are submitted for permit.

Shop drawings that alter structural intent need to be reviewed and approved by the AHJ, period. Otherwise, we’re setting our inspectors up to fail and putting the public at risk.

And where does the IBC give me any authority to require that shop drawings be submitted?

 

[steveray]

The proper way to handle it if a subcontractor or fabricator wants to change the design is for the RDP to submit amended construction documents, per IBC 107.4. This is different from deferred submittals, which are things that haven't yet been designed when the construction documents are submitted for permit.

We typically look for the RDP’s “Approved Shop Drawing” stamp or it is no bueno…

 

[Yankee Chronicler]

We typically look for the RDP’s “Approved Shop Drawing” stamp or it is no bueno…

Look for it -- when? Does your office routinely receive submittals of shop drawngs, or are you referring to looking at the field copies of the shop drawings when on site?

 

[steveray]

Look for it -- when? Does your office routinely receive submittals of shop drawngs, or are you referring to looking at the field copies of the shop drawings when on site?

Which is why we get the deferred submittals listed on the con docs so we can tell them what is deficient and what we are going to need shops on…IBC 107.3.4.1

 

[Yikes]

I agree that all deferred submittals should be listed on the plans and reviewed by the AHJ.
But I would like to clarify that not all shop drawings are deferred submittals, addressing code information omitted on the original plans.

Shop drawings that are deferred approvals:
Here in California, when public buildings are under the plan check jurisdiction of the Division of the State Architect, and the shop drawings contain essential code compliance information, the DSA requires the DPOR to either design everything themselves or to stamp the shop drawings. This is true for trusses, fire sprinklers, etc. They also require it at time of initial plan check -- they do not allow submittal to be deferred until after permit is issued.
So that is a shop drawing that is not a deferred approval.

Shop drawings that are not deferred approvals:
On the other hand, let's look at something like residential kitchen cabinets in a public housing project. Code requires that they have 50% accessible storage, and that their handles not require tight grasping or pinching. Our plans will show the initial cabinet elevations as proof-of-compliance. I do not expect to submit the cabinet maker's shop drawings for approval by the AHJ, because we've already proven our code compliance.

 

[jar546]

And where does the IBC give me any authority to require that shop drawings be submitted?

For me, its right here:

[A] 107.1 General. Submittal documents consisting of construction documents, statement of special inspections,geotechnical report and other data shall be submitted with each permit application in accordance with FloridaStatute 553.79. The construction documents shall be prepared by a registered design professional where requiredby Chapter 471, Florida Statutes & 61G15 Florida Administrative Code or Chapter 481, Florida Statutes & 61G1Florida Administrative Code. Where special conditions exist, the building official is authorized to require additionalconstruction documents to be prepared by a registered design professional.

 

[tmurray]

I would get shop drawings and submittals of critical items. Unfortunately, a lot of the time they forgot to give them to me in a timely manner. I've had a lot of stuff ripped out of buildings that could have been avoided if they had simply given them to me earlier. Some of the submittals/shop drawings that needed to be ripped out had those "reviewed by engineer" stamp on them.

 

[redeyedfly]

For me, its right here:

[A] 107.1 General. Submittal documents consisting of construction documents, statement of special inspections,geotechnical report and other data shall be submitted with each permit application in accordance with FloridaStatute 553.79. The construction documents shall be prepared by a registered design professional where requiredby Chapter 471, Florida Statutes & 61G15 Florida Administrative Code or Chapter 481, Florida Statutes & 61G1Florida Administrative Code. Where special conditions exist, the building official is authorized to require additionalconstruction documents to be prepared by a registered design professional.

Shop drawings aren't prepared by RDPs.

I build all over the country and I have never once had an AHJ ask for shop drawings in 30 years.

 

[Yikes]

Another example:
Here is a detail of a metal stair anchorage to foundation, as provided by the SEOR. (Stringer was called out on a different plan.)


This was approved on the original plan check. There was no listing of deferred approvals for steel, and the AHJ did not require later review of the steel shop drawings. In other words, the AHJ felt that the structural plans were already sufficient to verify code compliance.

Here's what came back on the shop drawing:

Do you feel that AHJ review of this shop drawing should have been required?