acalmon/ems-platform

Replication materials for "The Value of Time- and Location-Commitment for Decentralized Emergency Medical Services" (van den Berg et al.). Includes optimization models, simulation code, game-theoretic analysis, and case study data from Flare (Nairobi, Kenya). # Emergency Medical Services Platform Optimization - Nairobi Case Study **Research Paper:** "The Value of Time- and Location-Commitment for Decentralized Emergency Medical Services" **Journal:** Working Paper **Authors:** van den Berg, Calmon, Gernert, Lemmens, Rabinovich, Romero ## Overview This repository contains the data processing pipeline and simulation code supporting the research on decentralized emergency medical services (EMS) platforms in Nairobi, Kenya. The study investigates the relative effectiveness of improving temporal versus spatial commitment in ambulance fleet coordination for better service coverage. **Key Finding:** The coverage provided by ~340 loosely committed ambulances could potentially be matched by fewer than 15 optimally deployed, fully committed units - demonstrating the significant "cost of decentralization." ## Data Sources ### 1. Uber Movement Travel Time Data (2016-2020) **Source:** Uber Movement Platform (now discontinued) **Location:** Nairobi, Kenya **Temporal Coverage:** 2016-Q1 through 2020-Q1 (32 quarterly datasets) **Spatial System:** H3 Hexagonal Hierarchical Geospatial Index **Resolution:** 400 hexagonal zones (hexclusters) covering Nairobi metropolitan area **Zone Size:** Average ~2.94 km² per hexcluster #### Data Description The Uber Movement dataset provides aggregated travel time statistics between geographic zones based on actual Uber ride GPS traces. Each dataset includes: - `sourceid`: Origin hexcluster ID - `dstid`: Destination hexcluster ID - `month`: Month within the quarter - `mean_travel_time`: Average travel time in seconds - `standard_deviation_travel_time`: Standard deviation of travel time in seconds - `geometric_mean_travel_time`: Geometric mean of travel time - `geometric_standard_deviation_travel_time`: Geometric standard deviation Data is split into: - **Weekdays:** Monday-Friday travel patterns - **Weekends:** Saturday-Sunday travel patterns **Privacy & Anonymization:** Data is anonymized a …