Regional risk and resilience analysis of businesses and supply chains

Nocera, Fabrizio

Permalink

https://hdl.handle.net/2142/115710 Copy

Description

Title

Regional risk and resilience analysis of businesses and supply chains

Author(s)

Nocera, Fabrizio

Issue Date

2022-04-18

Director of Research (if dissertation) or Advisor (if thesis)

Gardoni, Paolo

Doctoral Committee Chair(s)

Gardoni, Paolo

Committee Member(s)

• Chen, Xin
• Konar, Megan
• Bocchini, Paolo

Department of Study

Civil & Environmental Eng

Discipline

Civil Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Interdependent critical infrastructure
• Modeling resolution
• Business interruption
• Supply chain disruption

Abstract

The well-being and economic prosperity of societies depend on critical infrastructure and their provision of goods, services, and resources to communities. Critical infrastructure enable individuals to achieve valuable states and activities. For instance, while having access to energy and being mobile are directly reliant on the performance of the power and transportation infrastructure, food security and business activities could be indirectly affected by the reduction in the performance of critical infrastructure. Past events highlighted the vulnerability of infrastructure to disruptions caused by natural or anthropogenic hazards. There are also complex interdependencies among infrastructure. Such interdependencies can cause disruptions to propagate across infrastructure, resulting in multi-fold catastrophic consequences at several levels (e.g., individuals, households, and communities). Much research has been devoted to assessing the performance of individual infrastructure components such as bridges, electric substations, and water pipelines and individual infrastructure such as transportation, power, and potable water infrastructure when facing a natural hazard. However, there has been limited effort directly linking failures and disruptions in critical infrastructure to failures and disruptions in systems of systems like businesses and their supply chains. There is a need for a mathematical formulation to predict accurately the impact of hazards on critical infrastructure and the cascading effects on businesses and supply chains. Such mathematical formulation must be probabilistic to account for the uncertainties inherent in the entire process. A comprehensive formulation must consider aleatory uncertainties in the characteristics of the hazards, epistemic uncertainties due to scarcity of data typical of low probability high consequence events, limitations in current models of critical infrastructure, and their interdependencies. This dissertation addresses some of the fundamental challenges in developing realistic models to study the infrastructure’s behavior under disruptive events and the cascading effects of infrastructure failures on the performance of businesses and supply chains. Specifically, the contributions of this dissertation are in four main areas that are (i) how to classify and then model infrastructure interdependencies, (ii) how to systematically select the modeling resolution of infrastructure, (iii) modeling and quantifying the likelihood of business interruption during the immediate aftermath and through the recovery of a natural hazard, and (iv) modeling the resilience of supply chains. This dissertation develops a novel classification of infrastructure interdependencies and presents a general mathematical formulation for modeling interdependent infrastructure. Specifically, the developed classification partitions the space of infrastructure interdependencies based on their ontological and epistemological dimensions. Under the ontology dimension, infrastructure interdependencies are classified into chronic and episodic. Under the epistemology dimension, infrastructure interdependencies are classified according to their mathematical modeling. The proposed classification enables understanding and mathematically modeling several classes of infrastructure interdependencies. The dissertation then develops a mathematical formulation to select the modeling resolution of infrastructure. The novelties of the proposed formulation are (i) how to systematically define equivalent simplified infrastructure models at the desired level of resolution, and (ii) how to select the appropriate level of resolution. In principle, the formulation adaptively increases the model resolution starting from a low-resolution infrastructure model until we reach the desired tradeoff among accuracy, simplicity, and computational efficiency. To define such a tradeoff, novel metrics are introduced to measure the level of agreement among estimates of the quantities of interest based on different resolution levels. This dissertation proposes a mathematical formulation to model and quantify the likelihood of business interruption with a ground-up approach by incorporating the dependency of business operations on physical structures, infrastructure, and social systems. The proposed formulation starts by estimating the direct physical damage to the business properties, the impact on the functionality of the supporting infrastructure, and the changes in the social systems. Then, the formulation integrates the effects of the individual causes that may lead to business interruption in a matrix-based system reliability method to estimate the likelihood of business interruption. Similarly, this dissertation develops a mathematical formulation to estimate the duration of business interruption. The formulation is developed using a combination of a multinomial logistic model and a random field. The multinomial logistic model predicts the business owner (or manager) class profile. For instance, depending on the type of business, certain business owners (or managers) are inclined (or have the need) to resume business activities as quickly as possible (e.g., even before all infrastructure services are restored). Similarly, other business owners (or managers) may prefer to resume business activities only when all the needed components have fully recovered (e.g., access to infrastructure services and the building where business activities are performed). Then, the random field model predicts the duration of business interruption given that a business experiences an interruption and the predicted class profile of the business owner (or manager). The random field formulation starts with predicting the duration of business interruption based on the duration of infrastructure service disruption and the time to recover the building where the business operates. The formulation improves the model accuracy by introducing a correction term. The model also includes a zero-mean field to capture the spatial correlation in the duration of business interruption among different businesses and an error term that captures unexplained uncertainty in the model. Finally, the dissertation also develops a novel mathematical formulation to model and quantify the impact of natural hazards on the supply chain by modeling the dependency of supply chain operations on physical structures and infrastructure and their potential damage. In the proposed formulation, supply chains are modeled using graph theory as networks where the physical components of the supply chain (e.g., hubs) are nodes of the network while the links represent the exchange of materials, information, and money. In a supply chain, hubs play a major role in the movement of goods because they usually connect many sources with many destinations. Typically, we can classify hubs into two main categories: distribution logistics hubs like warehouses and transport logistics hubs like ports or freight stations. Distribution logistics hubs are facilities where businesses hold their inventories and from which they receive the goods to sell in their retailers. Given the major role of hubs in a supply chain, in the proposed mathematical formulation, a particular focus is on the impact of natural hazards on hubs and the resulting cascading effects on businesses. The metric used to quantify the performance of transport logistics hubs is the number of goods displaced, while the metric used to quantify the cascading effects on businesses is the cumulative loss of profit of the retailers.

Graduation Semester

2022-05

Type of Resource

Thesis

Copyright and License Information

Copyright 2022 Fabrizio Nocera

Owning Collections