# Robust Defibrillator Deployment Under Cardiac Arrest Location Uncertainty via Row-and-Column Generation

**Authors:** Timothy C. Y. Chan, Zuo-Jun Max Shen, Auyon Siddiq  
**arXiv:** [1507.04397](https://arxiv.org/abs/1507.04397)  
**Date:** 2015-07-15  
**Summary page:** https://www.summarizepaper.com/en/paper/1507.04397v4/

## Key points

- Sudden cardiac arrest is a major public health concern
- Timely treatment is crucial for survival
- Automated external defibrillators (AEDs) increase chances of survival when placed in public locations accessible to bystanders before emergency medical responders arrive
- Predicting exact location of future cardiac arrests is impossible, making it necessary to strategically deploy AEDs in public spaces
- Timothy C. Y. Chan, Zuo-Jun Max Shen, and Auyon Siddiq propose a data-driven optimization model that accounts for uncertainty in future cardiac arrest locations
- Their approach involves discretizing a continuous service area into numerous scenarios where the probability of cardiac arrest at each location is uncertain
- They model uncertainty using a polyhedral set calibrated with historical cardiac arrest data and propose a row-and-column generation solution technique that exploits the structure of the uncertainty set
- Using real cardiac arrest data from Toronto, they conduct an extensive numerical study on AED deployment in public locations
- Their findings show that hedging against cardiac arrest location uncertainty can produce AED deployments that outperform intuitive sample average approximations by 9-15% and cut the performance gap with respect to an ex-post model by half
- Accounting for cardiac arrest location uncertainty can lead to improved accessibility of AEDs during emergencies and potentially better survival outcomes

## Summary

Sudden cardiac arrest is a major public health concern and timely treatment is crucial for survival. The use of automated external defibrillators (AEDs) significantly increases the chances of survival when placed in public locations where bystanders can access them before emergency medical responders arrive. However, predicting the exact location of future cardiac arrests is impossible, making it necessary to strategically deploy AEDs in public spaces to ensure accessibility during emergencies. To address this challenge, Timothy C. Y. Chan, Zuo-Jun Max Shen, and Auyon Siddiq propose a data-driven optimization model that accounts for uncertainty in future cardiac arrest locations. Their approach involves discretizing a continuous service area into numerous scenarios where the probability of cardiac arrest at each location is uncertain. They model uncertainty using a polyhedral set calibrated with historical cardiac arrest data and propose a row-and-column generation solution technique that exploits the structure of the uncertainty set. Using real cardiac arrest data from Toronto, they conduct an extensive numerical study on AED deployment in public locations. Their findings show that hedging against cardiac arrest location uncertainty can produce AED deployments that outperform intuitive sample average approximations by 9-15% and cut the performance gap with respect to an ex-post model by half. The authors' research suggests that accounting for cardiac arrest location uncertainty can lead to improved accessibility of AEDs during emergencies and potentially better survival outcomes. This study provides valuable insights into deploying AEDs effectively while considering unpredictable events such as sudden cardiac arrests which can lead to improved survival outcomes.