Access restoration is an extremely important part of disaster response. Without access to the site, critically important emergency functions like search and rescue, emergency evacuation, and relief distribution, cannot commence. Frequently, roads are opened in descending order of functional class. While this may work in simple cases, it does not take into account the urgency of the needs at the demand nodes, which represent population and economic centers or critical infrastructure that need access restored quickly. The key to optimal allocation of resources is proper prioritization. Such prioritization is possible only with state-of-the-art computing power that supports a Decision Support System (DSS) based on Commercial Remote Sensing (CRS), given the large area coverage, the complexity of the decision process, the need for rapid action, and the spatial nature of the impacts. This project has developed a DSS based on the following principles: (1) that CRS is key to estimating both network conditions and disaster impacts; (2) that Access Restoration Plans (ARPs) are more effective when they explicitly take into account priority rules/metrics and the resource constraints faced by responders; (3) that such priority rules/metrics must consider the impacts on population, economic centers, and critical facilities; (4) that cutting-edge optimization algorithms—using CRS inputs and priority rules/metrics—are the best way to reach sound decisions; and (5) that the proposed DST, with modifications, could play a key role during various phases of the disaster cycle, including response and effective recovery. The DST is comprised of 4 modules. The first module assesses the impacts on the transport network, using a CRS multi-modal data collection/processing at its core. The second module identifies the resource constraints (e.g., number of trucks available at time t). The third module specifies the rules/metrics that define the level of importance of the competing needs. Finally, an optimization procedure uses the other modules’ outputs to estimate the optimal Access Restoration Path. The CRS technologies developed as part of this project implemented algorithms that use road network shapefiles and CRS data (lidar for debris, imagery for flooding) to automatically detect obstructions to the roadway. The methodology implemented in the project to quantify debris and floods allows debris volumes and water depths to be automatically estimated from the output of the detection algorithms. The outputs from these algorithms (debris volumes and water depths) are converted into GIS shapefiles for easy viewing and integration with the Access Restoration module. This project has also developed a mathematical model that solves the problem of access restoration (AR). The mathematical model incorporates the impact of the allocation decisions using social costs (SC) as the objective function, although other priority metrics can be considered, such as population, time, cost and deprivation cost. The mathematical model integrates scheduling and capacity constraints into the process of AR. The model recognizes the temporal evolution of needs and availability of resources in the inclusion of varying parameters of capacity in different time periods along the planning horizon. This allows for integration of arriving resources into disaster response operations throughout the response. From the computational perspective, the complexity of the model makes it unfeasible to use commercial software to find optimal solutions to large instances. In response to this limitation, the team developed a heuristic procedure able to solve large instances in short execution times. Throughout the course of this project, the team wanted to ensure that the research and associated DSS tools could be smoothly transitioned into practice, fully validated, and useful to responders. It was important to ensure that the DSS met the expectations of the end users, in terms of ease of use, quality of results, and usefulness. To accomplish this, the research team created a Technical Advisory Council, and conducted outreach, validation and training to assist with transitioning the research into practice. All of the components of the solution developed are integrated in a web-based software application. The team has created a project website that provides an overview of the solution components, and provided access to the web application to stakeholders that have been trained in the use of the software. The project was a collaboration of the Rensselaer Polytechnic Institute (RPI), the Rochester Institute of Technology (RIT) the New York City Department of Transportation (NYCDOT), Ohio University and Emprata LLC. The RPI team worked on: the DST’s core mathematical models; research to produce priority rules/metrics; coordination and outreach with public-sector practitioners; project management and system integration. RIT was in charge of the development of the CRS algorithms. NYCDOT provided guidance on implementation, contributed datasets and expertise, and pilot-tested the DST.