Framework for Evaluating Economic Impact of IT based Disasters on the Interdependent Sectors of the US Economy

Jalal Ali and Joost Santos

The US economic system has become highly dependent on the Information Technology (IT) sector in the past several years and classified it as one of the critical infrastructures. The IT sector is a conglomerate of Internet services, Software industry, Computer design based infrastructures, and Information and data processing systems. Like every economic sector it is susceptible to natural and manmade disasters that cause disruptions to the production and delivery of services essential to other economic sectors, which are interdependent on each other within this economic system. This paper focuses on such perturbations caused by Denial-of-Service (DoS) attacks on IT infrastructure, and their consequences propagated in the form of inoperability and amplified losses as a result of these economic sector interdependencies. It analyzes the effects of such a scenario on the recovery behavior and indirect economic losses to other sectors of the economy. The Dynamic Inoperability Input Output model (DIIM) is utilized to identify the highly affected economic sectors based on two parameters: 1) the overall daily average economic loss, and 2) the average inoperability within the sectors. A modification to the model is proposed to accommodate variable inoperability over multiple periods. The paper utilizes Bureau of Economic Analysis (BEA) statistics to simulate the effects of an IT disaster scenario using a DoS attack example. This research provides policymakers a framework for estimating the consequences to the US economy of disruptions to the IT sector through a decision tool they can use for strategic planning, resilience management, and risk mitigation strategies across the IT-dependent economic sectors.

Stochastic Modeling of Manufacturing-based Interdependent Inventory for Formulating Sector Prioritization Strategies in Reinforcing Disaster Preparedness

Joanna Resurreccion and Joost Santos

The production of commodities and provision of services in today's infrastructure and economic sectors have become heavily driven by their intrinsic interdependent relationships. Disruptive events such as natural and man-made disasters have been known to render a number of these sectors inoperable. With a minimum level of inventory, the disrupted sectors are unable to deliver the expected product parts and/or services that fulfill the total production input requirements of other dependent sectors. Hence, the disaster consequences are propagated across these interdependent sectors, ultimately leading to amplified losses and diversified inoperability to the entire system. This research investigates the levels of inventory of the manufacturing sectors and how they impact system capability to absorb reductions in input requirements due to the inoperability of various sectors. A unique contribution of this research is the formulation of a stochastic model of interdependent inventory to provide more reliable estimates of economic losses and sector inoperability. System recovery may be improved and economic losses reduced through the implementation of inventory-enhanced policies to critically disrupted sectors. Inventory modeling and simulation are utilized using empirical cumulative distribution functions of inventory levels for the manufacturing sectors. To generate such distribution functions, we utilized the inventory-to-sales database from the Bureau of Economic Analysis, which spans 14 years and comprises a total 168 observations for each of the 21 manufacturing and trade sectors considered in the study. Results of the study are incorporated into the Dynamic Inoperability Input-Output Model to provide insights in formulating sector prioritization policies with economic loss and sector inoperability as metrics. Based on these two metrics, a Dynamic Cross Prioritization Plot categorizes the sectors to identify the prioritized set of critical sectors for further inclusion in inventory-enhancement planning. Risk assessment models with and without factoring inventory yield different sets of critical sectors. The latter case was found to have overestimated total economic loss by an average of 21.86% or $136.27M for a moderate intensity hurricane scenario in Virginia. Inventory modeling strengthens the choice of the critical sectors as it presents a closer replica of the real system. To complement the manufacturing-based inventory enhancement study explored in this paper, further work is recommended to evaluate the resilience of the service sectors.