As part of MCEER's research on acute care facilities, two decision support platforms are being developed by a multidisciplinary group of researchers.
One system is envisioned for use by a hospital's board of directors, who would use the software to help make decisions on which capital improvements would most effectively make the facility more hazard-resistant. Toward this goal, the research team is investigating a fragility approach in which multiple structural and nonstructural systems can be integrated within a common framework. Primarily developed by Professor Mircea Grigoriu and graduate student Cagdas Kafali at Cornell University, this platform has been embodied into a Rehabilitation Decision Analysis Toolbox (RDAT) through a user friendly MATLAB interface.
The overall objective of the fragility-based decision support system is to enhance the seismic resiliency of a given acute care facility. The hospital system resilience is measured by (1) the probability that it does not meet design specifications, or in other words, its fragility, (2) the consequences of inadequate performance, and (3) the system recovery time following a seismic event.
The input to RDAT consists of a reference time, seismic hazard information, geotechnical, structural and nonstructural systems properties, performance criteria, rehabilitation strategies, retro-fit and repair costs, loss of use, and loss of life, as well as some potential monetary benefits of retrofit.
RDAT rates distinct rehabilitation strategies for structural and nonstructural systems, using estimates of life cycle losses, and consequently, allows the selection of an optimal rehabilitation strategy. The RDAT output will be made compatible with STRATACAP, a widely used capital allocation decision analysis software. It is available now from the MCEER User's Network.
The second MCEER decision support platform investigates the use of an evolutionary framework for the design of structural systems incorporating advanced protective technologies in an uncertain seismic environment. It has been embodied into PC-based Evolutionary Aseismic Design and Retrofit (EADR) software. Professor Gary Dargush and graduate student Yufeng Hu of the University at Buffalo are the primary developers.
The EADR framework was successfully applied to a series of multistory steel moment-resisting frame retrofit examples, incorporating several advanced protective technologies including metallic, viscous and viscoelastic dampers. The initial version of the software was developed for a single processor by utilizing a genetic algorithm for discrete optimization under uncertain seismic environments. Within this approach, each structural evaluation is performed via a nonlinear transient dynamic analysis using phenomenological models of various advanced protective technology. Recently, efforts were directed toward enhancing the functionality, robustness and efficiency of this evolutionary approach. For example, a more realistic seismic environment was incorporated that is compatible with the seismic hazard model developed by the U.S. Geological Survey.
Recent work on EADR has focused on improving computational efficiency in order to make this approach a viable option for the engineering community. This has been accomplished by introducing new multi-level techniques and by migrating the framework to a desktop PC environment. A beta-version of the software has been made available to a selected group of MCEER's Industry partners.
The EADR software is part of a larger decision support system under development which will include non-engineering organizational and socio-economic constraints. This system is being developed to enable planners and public policy decision makers to evaluate, for example, the impact that proposed seismic related legislation may have on a geographically distributed network of hospitals within a given area.