The damage and destruction of homes, buildings, and other infrastructure after a major earthquake makes it necessary for a segment of the population to be evacuated to emergency shelters. However, since areas prone to earthquakes face a major strike at highly unpredictable and infrequent times, disaster and evacuation management can and should be enhanced through the development and effective utilization of useful simulation models. We utilize the Stochastic Pedestrian Cell Transmission Model (SPCTM), a real data-based pedestrian evacuation simulation framework built in collaboration with the National Science and Technology Center for Disaster Reduction (NCDR) in Taiwan to carry out various statistical analyses to gain useful insights for disaster management decision makers. In particular, SPCTM is used to (1) examine how evacuee compliance rate with the current government evacuation protocol (as opposed to SPCTM) affects the total evacuation time (2) compare the results of the evacuation process for different neighborhoods and compliance rates in Da’an district, and (3) determine how allocating different subsets of shelters to be opened affects the evacuation response time. The analysis results are insightful for government officials to formulate effective evacuation guidelines and strategies in the preparedness or response phases.