Supply network stress-testing of food security on the establishment-level

Recent events exemplified the fragility of national and international supply networks (SNs), leading to significant supply shortages of essential goods, such as food and medicines. Severe disruptions propagating along complex SNs can expose entire regions or countries to these risks. A lack of data and quantitative methodology has hitherto prevented an empirical quantification of the vulnerabilities of populations created by SN disruptions. Here, we propose supply network stress-testing (SNST) as a new data-driven methodology to quantify product-level supply losses of administrative districts that result from cascading supply disruptions between establishments. We demonstrate SNST on a large fraction of the Austrian food SN – composed of the pork production network from farms to meat processors and the distribution network of large food retailers – containing 23,001 establishments, 44,730 supply links, and 116 administrative districts. We rank all establishments with respect to their systemic criticality for the population using a novel systemic risk index, $ ESRI_i^{crit} $ ESRIicrit. We identify 28 facilities that – in case of failure – are expected to cause severe supply shortages to up to 20% of the population. SNST enables governments and industry to stress-test national supply networks of critical goods, to identify their weak spots and make them more resilient to future crises.