https://digitalcommons.calpoly.edu/arce_rpt/7
Date of Award
6-2026
Degree Name
MS in Architectural Engineering
Department
Architectural Engineering
College
College of Agriculture, Food, and Environmental Sciences
Advisor
Anahid Behrouzi
Advisor Department
Architectural Engineering
Advisor College
College of Agriculture, Food, and Environmental Sciences
Abstract
After major seismic events, reconnaissance efforts of various types are made to gather data. Coarse data can be collected with satellite imagery, mid-scale data can be recorded by capturing thousands of images, and granular data is typically documented manually by humans. Post-earthquake reconnaissance teams can be deployed by universities, research organizations, professional associations, and private firms. Academics, engineers, and students with varying degrees of expertise are tasked with collecting building metadata and damage data, typically in a short period of time. This data can be of various types including numerical (i.e. number of stories, height, etc.), categorical (i.e. damage level, lateral system, material, etc.) or binary (retrofitted, base isolated, etc.). Each building has a location on Earth, which naturally results in geospatial data being the most collected and readily available for building response data analysis. Post-processing building damage data collected in a tabular format with varying degrees of granularity can be challenging and time-consuming, especially for organizations that already volunteer their limited time to collect it. This thesis details the implementation of a workflow that starts with post-earthquake raw tabular data including a geospatial location, building damage level, and varying additional data, and generates maps and plots that facilitate quick inference about the most likely causes of damage. The workflow is a step-by-step process in a Jupyter Notebook to be executed within ArcGIS Pro and includes some Python helper files. Using the workflow, a user starts with a tabular database file (e.g. Excel) containing the collected data and ends with layouts in PDF that include maps and plots with applied filters that can highlight which building properties are associated with higher levels of damage.