Abstract
Railways are critical for economic integration and regional development, yet many
emerging economies face uneven infrastructure performance and limited tools for monitoring efficiency.
This study develops a replicable framework to measure the efficiency of Uzbekistan’s railway sector
using Data Envelopment Analysis and clustering methods. Three DEA models were constructed to
capture complementary aspects of efficiency, linking population and investment inputs with outputs such
as railway length, organizational activity, industrial production, and retail turnover. The results reveal
strong regional disparities: Tashkent City and Karakalpakstan consistently perform above average, while
Khorezm and Sirdarya lag behind. Efficiency trends further show that infrastructure availability does not
always align with economic activity, underscoring mismatches in investment allocation. Clustering
analysis grouped regions into three categories—high performers, emerging regions, and lagging areas—
providing a practical framework for policymakers to design differentiated interventions. Beyond the case
of Uzbekistan, the study demonstrates how open-source data and transparent Python-based DEA can be
used to support transport policy in data-limited contexts.