A Predictive Modeling Approach to Estimating Seismic Retrofit Costs

Area of Science:

• Civil Engineering
• Structural Engineering
• Risk Management

Background:

• Accurate seismic retrofit cost estimation is crucial for managing risks associated with aging infrastructure.
• Existing methods can be time-consuming or lack the scalability needed for large building portfolios.

Purpose of the Study:

• To develop a cost-effective methodology for estimating seismic retrofit expenses.
• To utilize historical data for predictive modeling of retrofit costs.
• To quantify the uncertainty associated with these cost estimations.

Main Methods:

• A generalized linear model (GLM) was developed using historical seismic retrofit cost data (FEMA 156).
• Building characteristics were used as predictors for retrofit costs.
• Prediction error was analyzed to compare different modeling choices and cost distributions.

Main Results:

• The GLM provides a practical approach for generating rapid seismic retrofit cost estimates.
• The methodology offers a quantifiable measure of prediction uncertainty.
• The choice of cost distribution significantly impacts the final cost estimates.

Conclusions:

• The proposed methodology offers a valuable tool for decision-makers managing extensive building portfolios.
• This approach balances accuracy with ease of application for preliminary seismic retrofit cost assessments.
• Understanding the influence of distribution choice is key for refining cost predictions.