Challenges in Determining the Fragilities of Nonstructural Components and Systems
Consulting Structural Engineer
Research Thrust #2: Seismic Retrofit of
Acute Care/Hospital Facilities
The ATC 58 Framework for Predicting the Performance of Nonstructural Systems
Develop Hazard Function for Site
Perform Structural Analysis
Perform Analysis of nonstructural systems
Form Response Functions (e.g. drift)
Form Fragility Functions (when certain performance states or damages states occur as function of the response functions). Probability/reliability based.
Form Nonstructural Damage Functions
Predict overall loss as a function of damage (may also consider downtime as a measure of performance)
Lack of Fragilities of Nonstructural Components and Systems
Difficult to define problem and real need – what really are the performance goals for nonstructural components ?
An immense number of nonstructural system and components – which ones are significant ?
The demands on nonstructural components and systems are complex – function of building response (both acceleration and drift).
Difficult to quantify fragilities of individual systems. Very installation sensitive (both design and installation).
To date, very little fragility data available (supplier proprietary data, SQUG data, some research data)
Example of Nonstructural Fragility Function
Adequate Testing Facilities and Test Protocols for Nonstructural Components
Shake table test facilities constructed to date have focused on ground motions. Building response motions can be much greater than ground motions.
Vertical motions can be significant to nonstructural components
No real consensus testing protocols exist for testing of nonstructural components. Some nonconsensus exist.
shake table testing
drift racking testing (glazing)
component ATC-24 type testing (FM Global)
Lack of Central Trade/Standards Organization for Nonstructural Systems
Because of diverse nature, no central group such as ACI, PCA AISC, AISI, etc.)
No central funding or leadership. No research and development support. No consensus standards.
Not much interest on the part of academia. Difficult to define problem. Does not appear technically challenging or interesting. Or to complex. Little industry support.
Needs someone to personally take on a leadership role to form.
Closest thing we have is the SQUG consortium.