REFIT: Resilient Infrastructure Toolkit

A toolkit to synthesize infrastructure and their interdependencies, model cascading failures, and design and test resilience strategies.

Toolkit Overview

Outages of power (red), water (blue), and combined power and water (purple)

The toolkit was developed to support resilience practitioners and researchers to efficiently invest in protecting their systems against emerging and uncertain hazards. The toolkit has many capabilities including:

  • 1) A synthetic infrastucture model to estimate the layout and characteristics of assets and the network;
  • 2) An interdependency model for joining infrastructure system physical connections;
  • 3) a cascading failure model for estimating how failures progress within and across infrastructure systems;
  • 4) an impact assessment model for capturing the outcomes of failures on people and other infrastructures.

Together, the toolkit supports the design of resilience strategies with sensitivity to limited resources.


Contact Prof. Chester to use the model, or for additional information.
Documentation including publications are provided below.

Mikhail Chester

Capabilities & Methods

The core objective of the toolkit is to describe how failures progress within and across infrastructure systems. A mixed methods approach is used depending on the quality and availability of requisite data. The toolkit starts with a power transmission network, other infrastructure networks (e.g., water distribution, stormwater) and a hazard model that describes how the power network will experience disruption. A probabilistic cascading failure model is run that explores the likelihood of the offlining of transmission lines, and a power system solver is used to balance flows, ultimately describing offlined substations. Millions of iterations are explored to stochastically describe failure likelihoods, and with each iteration the impacts of power outages are captured in other infrastructure. For example, a water distribution model including hydraulic solver is used to describe pump outages and resulting insufficient pressure at buildings. A consequence analysis model is used to capture the effects of power outages on other systems, which can include socio-economic analysis, or transportation signaling outages.

Infrastructure Modeling

The toolkit is flexible in that it can use real infrastructure data, or when limited to no data are available synthetic models can be employed. Power transmission and water distribution are critical infrastructure that can be synthesized for assessment.

Power Transmission

Using population data, demand estimates, and supplemented with publicly available data on local utilities, the toolkit generates a power transmission network including the substations and their service areas, transmission line connections, and transmission line characteristics. The synthetic module produces the necessary information to model the power system in state-of-the-art solvers including OpenDSS and PyPSA.

Water Distribution

Using information on treatment plants, elevation, roadway networks, buildings, and demand, the toolkit generates a water distribution network including pipe location, pipe diameter, pipe initial year of construction, pump location and pump size characteristics. The synthetic module produces the necessary information to model the water system in state-of-the-art solvers including EPANET.


The toolkit connects the water and power distribution models to establish interdependencies between the systems, identifying water system assets and their associations with substations.

Modeling Aspects

REFIT uses standard roadway and building data to estimate water and power distribution networks, including their physical and operational characteristics. Socio-economic data are captured. Interdependencies between power, water, transportation, and building systems is then estimated to capture cascading effects of outages.


Water distribution infrastructure data are often unavailable and rarely capture both infrastructure and operational characteristics. REFIT addresses this challenge by synthetically generating water distribution systems with results useable in simulation packages.


Building data are typically available from assessor databases and are necessary for understanding demand and household characteristics.


Power distribution infrastructure data require both an understanding of the physical layout of infrastructure and how energy is delivered from substations. REFIT estimates these system characteristics and produces results compatible with simulation packages.


Household characteristics are critical for understanding both demand and the implications of failures.


Transportation systems rely on the power system and can be adversely impacted by failure of power and water systems. REFIT connects the transportation system to the other infrastructure both spatially and physically.


REFIT captures the spatial and physical interdependencies of water, power, building, and transportation systems, including the capability to assess cascading failures across the systems.

Capabilities for assessing diverse hazards on infrastructure


Climate | Extreme Events

REFIT is being used to assess extreme heat, flooding, and other climate-related events on infrastructure.

Warfare | Terrorist Attacks

Kinetic attack, EMP, and sabotage scenarios have been assessed with REFIT.


Reliability impacts due to disrepair and their cascading failures, within and across infrastructure, have been assessed with REFIT.

Toolkit Output

The toolkit produces outputs that describe the characteristics of infrastructure systems including physical layout and operational conditions, failure simulations that stochastically describe vulnerable assets, and hazard profiles to assets, and the consequences of failures on interdependent systems. The outputs provide the necessary information to strategically invest limited resources to protect people, services, and assets to hazards with deep uncertainty.

Case Studies

We have developed the toolkit focusing on the Phoenix, New York, San Juan, and Atlanta regions, through National Science Foundation support (UREx, UWIN, Convergence, and RISE projects). The model has been deployed to military installations in support of base resilience planning.

Phoenix, Arizona

Phoenix is one of the fast growing cities in the U.S., with a modern infrastructure at risk to extreme heat, precipitation, and flooding.

New York, New York

New York's infrastructure are some of the oldest in the U.S., supporting a major population and massive economic activity.

Puerto Rico

San Juan's storied infrastructure is in need of major rehabilitation that considers the need for climate adaptation against sea level rise, storm surges, and hurricanes.

Atlanta, Georgia

Atlanta is a major population center in the U.S. Southeast and is subject to major heat, precipitation, and flooding challenges.

Norfolk, Virginia

Nisi elit consequat ipsum, nec sagittis sem nibh id elit. lorem ipsum dolor sit amet of Lorem Ipsum. Proin gravida nibh vel velit auctor aliquet. Aenean sollicitudin, lorem quis bibendum auctor.


Preparing Infrastructure for Surprise: Fusing Synthetic Network, Interdependency, and Cascading Failure Models

Ryan Hoff, Mikhail Chester, Environmental Research Infrastructure and Sustainability, 2023, 3(2), 025009, doi: 10.1088/2634-4505/acd07d.

Read Publication
Identifying Cascading Failures on Synthetic Power Transmission Systems

Ryan Sparks, Ryan Hoff, Nathan Johnson, Mikhail Chester, Arizona State University Digital Repository, 2023.

Read Report
A Synthetic Water Distribution Network Model for Urban Resilience

Nasir Ahmad, Mikhail Chester, Emily Bondank, Mazdak Arabi, Nathan Johnson, Benjamin Ruddell, Sustainable and Resilient Infrastructure, 2022, 7(5), pp. 333-347, doi: 10.1080/23789689.2020.1788230.

Read Publication
A Geospatially-enabled Web Tool for Urban Water Demand Forecasting and Assessment of Alternative Urban Water Management Strategies

Sybil Sharvelle, Andre Dozier, Mazdak Arabi, Brad Reichel, Environmental Modelling & Software, 2017, 97, pp. 213-228, doi: 10.1016/j.envsoft.2017.08.009.

Read Publication
A Methodology for the Creation of Geographically Realistic Synthetic Power Flow Models

Kathleen Gegner, Adam Birchfield, Ti Xu, Komal Shetye, Thomas Overbye, Proceedings of the IEEE Power and Energy Conference at Illinois (PECI), 2016, doi: 10.1109/PECI.2016.7459256.

Read Publication