The Hyatt Regency walkway collapse occurred on July 17, 1981, when two suspended walkways inside the hotel’s atrium in Kansas City, Missouri, fell to the atrium floor. The National Bureau of Standards investigation, now catalogued by the National Institute of Standards and Technology (NIST), identified insufficient load capacity in the box beam and hanger rod connections as the most probable cause, with the initiating failure at the east end of the fourth-floor walkway’s middle box beam. The Hyatt collapse is also a project-process failure. A change in how the walkways were hung roughly doubled the load on the upper walkway connections, and the project never clearly established who owned the design of those connections or communicated the change through the proper channels.
This article separates the physical mechanism from the decisions that produced it, then sets out the accountability and lessons that followed.
What happened on July 17, 1981
Two suspended walkways crossed the atrium of the Hyatt Regency Hotel. The fourth-floor walkway fell onto the second-floor walkway, and both reached the atrium floor. A third walkway, offset from the other two, was independently suspended and did not share the same collapse path. The sequence began with the fourth-floor structure, which means the failure was not simply a chain reaction that started at the bottom.
The NBS investigation, published in 1982 as BSS 143, combined site inspections, laboratory tests, and analytical studies. NIST’s current incident summary draws on that work and is the most direct official statement of the probable cause.
#1 Best Overall
How the walkways were hung
Each walkway was a steel assembly with box beams carrying the deck, and hanger rods suspended those beams from the roof structure above. The critical question is how the load reached the roof. Two arrangements matter here: the one the engineers originally conceived, and the one that was actually built.
| Arrangement | How the hanger rods carried the walkways | Load on the upper walkway connection |
|---|---|---|
| Original concept | Continuous hanger rods ran through the upper walkway’s box beams, carrying that walkway and the walkway below from the roof. | Not stated as a numerical baseline in the sources reviewed; ASCE describes it only as the arrangement that was later abandoned. |
| As built | An offset double-rod arrangement. One rod hung the upper walkway from the roof. A second rod connected the lower walkway to the upper walkway. | Doubled, according to the 2007 ASCE account by Tara Hoke. |
The practical consequence is a change in the load path. In the built arrangement, the connection at the upper walkway’s box beam had to carry not only the upper walkway but also the lower walkway hanging from it. That is why a connection that had been adequate in the concept could be overloaded in the field.
The structural failure mechanism
NIST’s wording is deliberately cautious. It identifies insufficient capacity in the box beam and hanger rod connections as the most probable cause, not as a single broken part. The damage pattern points to the initiating connection at the east end of the fourth-floor walkway’s middle box beam. The failure was therefore a connection failure: the joint between the steel that carried the load and the rods that transferred it could not sustain the doubled demand.
Rank #2
Describing the event as “the rods snapped” misses this. The official finding concerns the connection as a whole, and ASCE’s account explains why the modified arrangement placed more demand on that connection than the design intended to handle.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →How the design decisions unfolded
ASCE’s 2007 account by Tara Hoke reconstructs the sequence of decisions that led to the built arrangement. The steps below are drawn from that account, and they show how each decision depended on an assumption about who was responsible for the next one.
- The connection drawing was produced as a concept. The engineer treated the original connection drawing as conceptual, not as a complete design.
- The fabricator treated the same drawing as complete. The fabricator read it as finished engineering that could be built as shown.
- The continuous-rod arrangement proved impracticable. When the fabricator could not build the continuous-rod connection, it sought a change.
- The change was requested by telephone. The request was not followed by a written submission intended for formal approval.
- The detailer assumed the connection had been designed. Believing the connection was already settled, the detailer did not calculate it.
- Review and approval proceeded without a complete independent check. The connection calculations were not fully checked before the walkways were approved.
Each party acted on a reasonable-looking assumption, but no single party held the complete picture. The fabricator had a changed design, the detailer assumed a design existed, and the reviewers were not checking the connection that had changed.
Why ASCE calls it a process failure
ASCE’s 2011 retrospective by Tara Hoke, then ASCE general counsel, identifies two fundamental procedural failures. In her words, they were “a lack of clarity as to which party bore the ultimate responsibility for the design of the connections and a failure to communicate vital information to the various parties involved in the project.”
Those two failures explain why the technical error was not caught. Responsibility for the connection was ambiguous, so no one was assigned to verify the altered load path. Information about the change was not communicated in writing to everyone who needed it, so the people checking the design were working from an outdated picture.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
ASCE says it responded with a 1985 policy assigning structural design responsibility for steel structures, including connections, to the engineer of record. Its later quality-assurance guidance, including ASCE Manual 73, extends that focus to how owners, designers, and constructors share responsibility for quality.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Accountability: three separate processes
Several bodies examined the collapse, and their findings address different questions. Keeping them apart avoids conflating technical cause, professional ethics, and legal licensing.
The technical finding: NBS and NIST
The NBS investigation and NIST’s incident summary address the physical question: why the structure failed. Their finding is the probable-cause statement about the box beam and hanger rod connections. It is a technical determination about the structure, not an ethics or licensing ruling.
The professional ethics finding: ASCE
ASCE’s Committee on Professional Conduct found the engineer of record responsible for all elements of the structural design. The Board of Direction did not agree with the full extent of that ethics finding and imposed a three-year membership suspension. This was a decision by a professional society about its own members, based on the society’s code of conduct.
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Clear out junk files and repair common Windows errors3Fix the driver behind crashes, sound loss and screen glitchesBest Value
- Used Book in Good Condition
The code language the committee relied on reads: “Engineers shall hold paramount the safety, health, and welfare of the public in the performance of their professional duties.”
State licensing action
Missouri took licensing action against engineers involved in the project. A state licensing proceeding is a separate process from ASCE membership discipline, and it carries different consequences. The sources cited for this article do not describe the outcome of that proceeding, so no result is stated here.
Casualty figures and why they differ
The sources give different casualty counts, and the differences are not reconciled in the records reviewed. When citing a figure, attribute it to its source.
| Source | Date | Deaths | Injuries |
|---|---|---|---|
| NIST incident summary | Page created 2011, updated 2025 | 113 | 186 |
| ASCE, Tara Hoke account | 2007 | 114 | Not stated |
| ASCE, Tara Hoke retrospective | 2011 | 114 | More than 200 |
Practical lessons for engineering and construction teams
The Hyatt collapse matters less as history than as a checklist of failure points that recur on projects with complex suspended or connected structures. The following practices address the specific gaps identified above.
- Name one responsible engineer for every connection, in writing, before fabrication begins.
- Mark any concept drawing as conceptual when it is not a complete design, and state who must complete it.
- Require a written change request, with a recalculated load path, for any field change to a load-carrying connection, even when the change is discussed by phone first.
- Confirm that the detailer knows whether a connection has been calculated before detailing it.
- Ensure the independent check covers the connections that changed, not only the drawings that were originally submitted.
- Distribute changes in load path to every party that relies on the design, including those who approve and inspect it.
Further reading
ASCE points readers to Norbert J. Delatte’s Beyond Failure: Forensic Case Studies for Civil Engineers for further technical discussion of failure investigations. For quality-assurance practice, ASCE Manual 73, Quality in the Constructed Project: A Guide for Owners, Designers, and Constructors, covers how responsibilities are shared across a project. Neither work replaces the NBS investigation as the primary technical source.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
