SIRA

https://github.com/GeoscienceAustralia/sira
Release: 1.0.1

SIRA stands for Systemic Infrastructure Resilience Analysis. It represents a methodology and supporting code for systematising vulnerability and risk analysis of lifeline infrastructure assets to natural hazards.

The impact assessment incorporates the fragilities of individual system components and their topological configuration within the infrastructure system under study. This analysis is supplemented by an assessment of the post-hazard system functionality through network flow analysis, and consequent restoration times and costs.

SIRA has been used to study responses of infrastructure systems, both facilities (e.g. power generation plants, high voltage substations) and geospatially dispersed networks (e.g. electricity transmission networks), to earthquake hazards. The methodology is implemented in an open-source Python package.

At the time of writing this documentation, the software has been tested to work with earthquake hazards only. However, the methodology and class structure have been developed to make the modelling process hazard agnostic. The hazard specification process and infrastructure models are designed to allow for incorporating different hazards and arbitrary infrastructure sectors/assets.

SIRA has been developed in Geoscience Australia (GA) in support of the agency’s strategic priority to contribute to enhancing the hazard resilience of communities in Australia and its region.

Features

Open Source: Written in Python, avoids dependency on proprietary tools, platform agnostic.

Flexible Infrastructure Model: The data model is based on graph theory. All infrastructure systems are represented as networks. This allows an user to develop arbitrarily complex custom facility models – for a facility or a network – for impact simulation.

Scalability: The code can detect and leverage the host system capabilities to enable parallel processing capabilities on high performace computing platforms for very large scale simulations (tens of millions of hazard scenarios and model with thousands of components), while being able to run on midrange laptops for smaller hazard sets.

Extensible Component Library: User can define new instances of component_type (the building blocks of a facility or network) and link it to existing or custom fragility algorithms.

Component Criticality Analysis: Scenario analysis tools allow users to identify the cost of restoration for chosen scenarios, expected restoration times, and options for targeted component upgrades that yield greatest improvements to system resilience.

Restoration Prognosis: SIRA can assess restoration times for a system based on assigned priority levels for specific functions (or outputs) of the system, the recovery functions of components, and repair streams. The restoration curves used are taken from literature or developed with support from industry. ‘Repair streams’ indicate how many repair task can be undertaken in parallel, which is a proxy for the resources available for repair tasks.

User Documentation

• 1 Introduction
• 2 Model Concepts
  • 2.1 The ‘System’ Model
  • 2.2 System Loss Modelling
  • 2.3 System Restoration Model
• 3 Simulation Setup
  • 3.1 Simulation Configuration File
• 4 Infrastructure Model Setup
  • 4.1 Basic System Metadata: system_meta
  • 4.2 List of Components: component_list
  • 4.3 Connections between Components: component_connections
  • 4.4 Configuration of Supply Nodes: supply_setup
  • 4.5 Configuration of Output Nodes: output_setup
  • 4.6 Component Type Damage Algorithms: comp_type_dmg_algo
  • 4.7 Definition of Damage States: damage_state_def
• 5 Installation and Usage
  • 5.1 Requirements
  • 5.2 Recommended Process for Building a SIRA Environment
  • 5.3 AWS and Docker
  • 5.4 Running a Simulation with SIRA
  • 5.5 Parallel Execution
• 6 Component Fragility Attribution

• Copyright
• References
• Glossary