• Skip to primary navigation
  • Skip to main content

Center for Infrastructure, Transportation, and the Environment (CITE)

Rensselaer Polytechnic Institute SEARCH
School of Engineering
Center for Infrastructure, Transportation, and the Environment (CITE)
  • About Us
  • Team
  • News
  • Training & Outreach
  • Research
  • Software & Tools
  • CoE-SUFS
  • Contact Us

Disaster Response

Remote Sensing Decision Support System for Optimal Access Restoration in Post Disaster Environments

October 16, 2019 By admin

Remote Sensing Decision Support System for Optimal Access Restoration in Post Disaster Environments

START YEAR:

COMPLETION YEAR: 2017

TOPIC(S): Disaster Response

PRIMARY CONTACT(S):

  • José Holguín-Veras

PARTNER(S):

  • New York City Department of Transportation (NYCDOT)
  • Rochester Institute of Technology (RIT)

SPONSORS/FUNDING:

  • U.S. Department of Transportation
Remote Sensing Decision Support System for Optimal Access Restoration in Post Disaster Environments Project

OVERVIEW

The project’s main goal is to develop a state-of-the-art Decision Support System (DSS) that, using network condition and disaster impact assessments provided by Commercial Remote Sensing (CRS), will compute optimal Access Restoration Plans (ARP). This will help responders optimally use their scarce resources to orchestrate the road openings, road repairs, and other similar actions; and subdivide the work so that multiple groups could seamlessly cooperate to reach maximum effectiveness. The DSS will process multi-modal temporal data feeds (GIS, social media feeds, etc.) to update estimates of disaster impacts, and will use modern optimization procedures to update the ARP as new data becomes available. To achieve the overall goal, the team will collaborate to:

  • Develop a Commercial Remote Sensing (CRS) Module that allows emergency response personnel and planners to produce rapid and accurate assessments of network conditions and disaster impacts by integrating multi-modal/multi-temporal data
  • Develop an Optimization Module and the Decision Support System (DSS) to produce an Optimal Access Restoration Plan using appropriate priority metrics
  • Create a DSS that is smoothly transitioned to practice, fully validated, and useful to responders. These tasks will ensure that the DSS meets the expectations of the end users, in terms of ease of use, quality of results, and usefulness. This will be accomplished by means of creating a Technical Advisory Council, a vigorous process of outreach and validation, and a solid process of training and transition to implementation
  • Develop appropriate procedures for integration of multiple responder groups. This objective seeks to facilitate the integration of multiple responder groups (such as nearby DOTs that send equipment like plow trucks and loaders to help the effort) to the overall effort of access restoration

KEY TASKS

  • Assessment of the CRS technologies for use on debris and flood classification
  • Development of algorithms for location, classification, and quantification of debris/flood occurrences
  • Development of procedures to integrate multi-modal/temporal data to assess disaster impacts
  • Development of algorithms and scripts to geo-locate estimates of network conditions and disaster impacts
  • Review, evaluate, and select applicable priority metrics
  • Develop and improve the Optimization Module
  • System Integration of the CRS and Optimization Modules to create the Decision Support System (DSS)
  • Creation of Technical Advisory Council (TAC)
  • Review current practices

KEY FINDINGS

  • Developed CRS technologies to implement algorithms to detect obstructions to the roadway
  • Considered 5 different metrics (population, private cost, time, deprivation time and cost, and social costs)
  • Recognize that each metric could lead to different results
  • Suggest procedures to let the users of the DSS select the most appropriate metric

KEY PRODUCTS

  • Final Report
  • Access Restoration Planning (ARP) Software

ADDITIONAL PRODUCTS

CONTRIBUTING TEAM MEMBERS

  • Felipe Aros-Vera

RELATED PROJECTS

Cyber Enabled Discovery System for Advanced Multidisciplinary Study of Humanitarian Logistics for Disaster Response

October 16, 2019 By admin

Cyber Enabled Discovery System for Advanced Multidisciplinary Study of Humanitarian Logistics for Disaster Response

START YEAR: 2012

COMPLETION YEAR: 2017

TOPIC(S): Disaster Response, Network Modeling

PRIMARY CONTACT(S):

  • José Holguín-Veras

PARTNER(S):

  • University of Delaware
  • Virginia Polytechnic Institute

SPONSORS/FUNDING:

  • National Science Foundation Information and Intelligent Systems (NSF-IIS)
Remote Sensing Decision Support System for Optimal Access Restoration in Post Disaster Environments Project

OVERVIEW

This project is concerned with the development of an integrative “Cyber Enabled Discovery System for Advanced Multidisciplinary Study of Humanitarian Logistics for Disaster Response.” As part of the work, transportation, computer, mathematical, and social scientists will collaborate to:

  • Create new paradigms of humanitarian logistic (HL) models that explicitly consider two key aspects not studied by current techniques: deprivation costs (DC), and material convergence (MC)
  • Develop appropriate models to represent human suffering as a DC
  • Explicitly consider DC in the key HL decision models
  • Develop analytical models to quantify/influence the amount, type, and arrival patterns of donations
  • Gain insight into the links between media framing of needs and MC
  • Define mechanisms to modify donor behavior
  • Develop algorithms and heuristics to solve the formulations developed

The goals of this project are:

  • To develop a new generation of computable HL models capable of explicitly considering the impacts of delivery actions on DCs, and able to integrate the real time estimates of MC produced by the cyber enabled discovery system in the definition of proper control procedures
  • To predict—on the basis of real (or quasi) time analysis of media reports—the flow of goods to the disaster site. Predictions would integrate all media data (e.g., news, websites, social networks), and a predictive model based on responses to previous similar disasters
  • To qualitatively and quantitatively explain the relationship between media framing of needs and MC, and suggest response strategies to better react/influence, media-driven MC

KEY TASKS

  • Incorporation of DC in HL models
  • Identification of the linkages between media framing and MC, both qualitatively and quantitatively
  • Development of cyber tools to estimate donation amounts, types, and arrival patterns
  • Definition of control procedures to influence donation behavior
  • Integration of MC estimates and DC into HL
  • Conceptual validation of the models
  • Educational activities/Outreach to practitioners/Curricular changes

KEY FINDINGS

KEY PRODUCTS

ADDITIONAL PRODUCTS

CONTRIBUTING TEAM MEMBERS

RELATED PROJECTS

Copyright © 2023 Rensselaer Polytechnic Institute (RPI)

Media Policy    |    Web Privacy Policy    |    Student Consumer Information    |    Title IX Policy    |    Accessibility