HR: 1340h
AN: S23A-0222    [Abstracts]
TI: Saving Lives and Money: A Multi-Objective Optimization Approach to the Selection of Structural Retrofits
AU: * Franco, G
EM: gf135@columbia.edu
AF: Columbia University, 500 West 120th Street 610 Mudd Building, MC 4709, New York, NY 10027
AU: Deodatis, G
EM: deodatis@civil.columbia.edu
AF: Columbia University, 500 West 120th Street 610 Mudd Building, MC 4709, New York, NY 10027
AU: Smyth, A
EM: smyth@civil.columbia.edu
AF: Columbia University, 500 West 120th Street 610 Mudd Building, MC 4709, New York, NY 10027
AB: The existence of large numbers of poorly constructed buildings in earthquake-prone areas has made large scale retrofitting campaigns a desirable strategy for reducing the risk of loss of life and infrastructure. Since the retrofitting operation must be, therefore, carried out for numerous buildings, it has become a necessity to find the most attractive retrofitting solution for a given type of buildings that significantly reduces the risk of loss of life while remaining as economic as possible. Each retrofit solution for an existing building carries a set of potential costs and benefits over a given period of time. Some of these costs and benefits can be taken into account in terms of monetary values, whereas others cannot. The value of lives spared or lives lost that may potentially occur over the lifespan of a building as a consequence of choosing a particular retrofit cannot be easily measured in terms of money. In the absence of better methods to incorporate the value of life into economic analyses, a somewhat arbitrary monetary quantity based on personal insurance or personal productivity is often chosen as a proxy to quantify it. These quantities, however, not only fail to capture the otherwise incalculable cost of life, but are also strongly dependent on the economy under study. In this work, the monetary costs of construction and structural damage are clearly differentiated from the loss of life, which is simply measured as the number of potential casualties in a certain earthquake scenario. Finding the best retrofit then becomes a multi-objective optimization problem, whose purpose is to find the cheapest solution that saves most lives. The conflicting nature of the objectives causes the appearance of not only one optimal solution but of a set of most convenient solutions that capture the different levels of trade-off between costs and lives saved. A large number of possible structural retrofits must be considered in order to find the set of best solutions (the Pareto-optimal front) and, thus, the problem quickly reaches an enormous complexity. In this work, a genetic algorithm has been used as an appropriate tool to solve this mathematical problem. Ultimately, the set of best retrofitting options enables one to choose and to evaluate the attractiveness of a given solution from an economy and life saving perspective, without artificially lumping these two together.
DE: 6304 Benefit-cost analysis
DE: 6309 Decision making under uncertainty
SC: Seismology [S]
MN: Fall Meeting 2005