Skip to content
Technical Report

Characterizing Washington State’s Supply Chains

Publication: Transportation Northwest Regional Center X (TransNow)
Publication Date: 2012

The University of Washington (UW), Washington State University (WSU), and Washington State Department of Transportation (WSDOT) recently developed a multi-modal statewide geographic information system (GIS) model that can help the state prioritize strategies that protect industries most vulnerable to disruptions, supporting economic activity in the state and increasing economic resilience. The proposed research was identified after that project as an important step in improving the model’s ability to measure the impact of disruptions. In addition to developing the model, the researchers developed two case studies showing the model’s capabilities: the potato growing and processing industry was chosen as a representative agricultural sector and diesel fuel distribution for its importance to all industry sectors. As origin-destination data for other freight-dependent sectors is added to the model, WSDOT will be able to evaluate the impact of freight system disruptions on each of them. Moving forward, it is not cost-effective to develop case studies in the manner used for these case studies, therefore, the state is currently supporting activities at the national level that will provide methods for collecting statewide commodity flow data. However, this commodity flow data will still lack important operational detail necessary to understand the impacts of transportation changes. This research will begin to fill that gap by developing a transportation-based categorization of logistics chains. The goal is not to capture all of the complexity of supply chain logistics but to identify approximately 15-20 categories within which supply chains behave similarly from a transportation perspective, for example, in their level of scheduling and methods for route selection. Researchers will use existing publicly available data, conduct an operational survey, and analyze GPS data collected for WSDOT’s freight performance measures project to identify the categorization.

Authors: Dr. Anne Goodchild, Andrea Gagliano, Maura Rowell
Recommended Citation:
Goodchild, A., Gagliano, A., & Rowell, M. (2012). Characterizing Washington State’s Supply Chains (No. TNW2012-13).
Technical Report

Development and Analysis of a GIS-Based Statewide Freight Data Flow Network

Download PDF  (4.92 MB)
Publication: Washington State Department of Transportation
Publication Date: 2009
In the face of many risks of disruptions to our transportation system, this research improves WSDOT’s ability to manage the freight transportation system so that it minimizes the economic consequences of transportation disruptions.
Faced with a high probability that major disruptions to the transportation system will
harm the state’s economy, the Washington State Department of Transportation
(WSDOT), in partnership with Transportation Northwest (TransNow) commissioned
researchers at the University of Washington and Washington State University to
undertake freight resiliency research to:
  • Understand how disruptions of the state’s freight corridors change the way
    trucking companies and various freight-dependent industries route goods,
  • Plan to protect freight-dependent sectors that are at high risk from these disruptive
    events, and
  • Prioritize future transportation investments based on the risk of economic loss to
    the state
To accurately predict how companies will route shipments during a disruption,
this research developed the first statewide multimodal freight model for Washington
State. The model is a GIS-based portrayal of the state’s freight highway, arterial, rail,
waterway and intermodal network and can help the state prioritize strategies that protect industries most vulnerable to disruptions.
The report features two case studies showing the model’s capabilities: the potato growing and processing industry was chosen as a representative agricultural sector, and diesel fuel distribution for its importance to all industry sectors. The case studies are found in sections 5.2 and 5.3 in the report and show how the statewide freight model can:
  • Predict how shipments will be re-routed during disruptions, and
  • Analyze the level of resiliency in various industry sectors in Washington State
The two case studies document the fragility of the state’s potato growing and processing
sectors and its dependence on the I-90 corridor, while showing how the state’s diesel
delivery system is highly resilient and isn’t linked to I-90.
As origin-destination data for other freight-dependent sectors is added to the model,
WSDOT will be able to evaluate the impact of freight system disruptions on each of
them. This will improve WSDOT’s ability to develop optimal strategies for highway
closures, and prioritize improvements to the system based on the relative impact of the
This research addressed several technical areas that would need to be resolved by any
organization building a state freight model. First, the researchers had to decide on the
level of spatial and temporal detail to include in the statewide GIS freight model. This
decision has significant consequences for data resolution requirements and results.
Including every road in Washington would have created a cumbersome model with a
large number of links that weren’t used. However, in order to analyze routing during a
disruption all possible connections must exist between origin and destination points in the model. While the team initially included only the core freight network in the model,
ultimately all road links were added to create complete network connectivity.
Second, as state- and corridor-level commodity flow data is practically non-existent, data
collection for the two case studies was resource intensive. Supply chain data is held by
various stakeholders and typically not listed on public websites, and it isn’t organized by
those stakeholders for use in a freight model. In most cases it’s difficult to assure data
quality. The team learned that the most difficult data to obtain is data on spatially or
temporally variable attributes, such as truck location and volume. So they developed a
method to estimate the importance of transportation links without commodity flow data.

Third, the freight model identified the shortest route, based on travel time, between any
origin and destination (O/D) pair in the state, and the shortest travel-time re-route for
each O/D pair after a disruption. The routing logic in the model is based on accepted
algorithms used by Google Maps and MapQuest. Phase III of the state’s freight
resiliency research was funded by WSDOT and will result in improved truck freight
routing logic for the model in 2011.
The two case studies showed how the state’s supply chains use infrastructure differently,
and that some supply chains have built flexibility into their operations and are resilient
while others are not, which leads to very different economic consequences. The results
of these case studies significantly contributed to WSDOT’s understanding of goods
movement and vulnerability to disruptions.
In the future, Washington State will need corridor-level commodity flow data to
implement the research findings and complete the state freight model. In 2009, the
National Cooperative Freight Research Program (NCFRP) funded development of new
methodology to collect and analyze sub-national commodity flow information. This
NCFRP project, funded at $500,000, will be completed in 2010 and provide a mechanism for states to accurately account for corridor-level commodity flows. If funds are available to implement the new methodology in Washington State, the state’s freight
model will use the information to map these existing origin destination commodity flows
onto the freight network, evaluate the number of re-routed commercial vehicles, and their increased reroute distance from any disruption. This will allow WSDOT to develop
prioritized plans for supply chain disruptions, and recommend improvements to the
system based on the economic impact of the disruption.
This report summarizes 1) the results from a thorough review of resilience literature and resilience practices within enterprises and organizations, 2) the development of a GIS-based statewide freight transportation network model, 3) the collection of detailed data regarding two important industries in Washington state, the distribution of potatoes and diesel fuel, and 4) analysis of the response of these industries to specific disruptions to the state transportation network.
The report also includes recommendations for improvements and additions to the GIS model that will further the state’s goals of understanding the relationship between infrastructure availability and economic activity, as well as recommendations for improvements to the statewide freight transportation model so that it can capture additional system complexity.
Authors: Dr. Anne GoodchildDr. Ed McCormack, Eric Jessup, Derik Andreoli, Kelly Pitera, Sunny Rose, Chilan Ta
Recommended Citation:
Goodchild, Anne V., Eric L. Jessup, Edward D. McCormack, Derik Andreoli, S Rose, Chilan Ta and Kelly Pitera. “Development and Analysis of a GIS-Based Statewide Freight Data Flow Network.” (2009).

Using a GIS-based Emissions Minimization Vehicle Routing Problem with Time Windows (EVRPTW) Model to Evaluate CO2 Emissions and Costs: Two Case Studies Comparing Changes Within and Between Fleets

Publication: Transportation Research Board 90th Annual Meeting
Publication Date: 2010

Growing pressure to limit greenhouse gas emissions is changing the way businesses operate. A model was developed in ArcGIS to evaluate the trade-offs between cost, service quality (represented by time window guarantees), and emissions of urban pickup and delivery systems under these changing pressures.

A specific case study involving a real fleet with specific operational characteristics is modeled as an emissions minimization vehicle routing problem with time windows (EVRPTW). Analyses of different external policies and internal operational changes provide insight into the impact of these changes on cost, service quality, and emissions. Specific considerations of the influence of time windows, customer density, and vehicle choice are included.

The results show a stable relationship between monetary cost and kilograms of CO2, with each kilogram of CO2 associated with a $3.50 increase in cost, illustrating the influence of fuel use on both cost and emissions. In addition, customer density and time window length are strongly correlated with monetary cost and kilograms of CO2 per order. The addition of 80 customers or extending the time window 100 minutes would save approximately $3.50 and 1 kilogram of CO2 per order. Lastly, the evaluation of four different fleets illustrates significant environmental and monetary gains can be achieved through the use of hybrid vehicles.

Authors: Erica Wygonik
Recommended Citation:
Wygonik, Erica and Anne V. Goodchild. “Using a GIS-based emissions minimization vehicle routing problem with time windows (EVRPTW) model to evaluate emissions and cost trade-offs in a case study of an urban delivery system.” Proc., 90th Annual Meeting of the Transportation Research Board, Transportation Research Board, Washington, DC.
Student Thesis and Dissertations

Finding a (Food) Way: A GIS Modeling Approach to Quantifying Local Food Transportation Systems

Publication Date: 2017

In recent years the focus on and prioritization of the notion of local food, food access and sustainability has been increasing throughout the U.S., especially in urban areas. The rising demand and growing preference for local produce in turn leads to changes in how we transport food. The supply chains found in urban areas are already complicated and costly, and as demand changes this poses a challenge if the local food movement is to be accommodated in our cities. A new initiative seeks to mitigate these challenges through the introduction of a mobile application that allows users to order local produce online. Logistics modeling was conducted as a case study to support this effort. The goal of the research was to be able to inform and support decision-making on the logistics to support agricultural development and equal food access. The research found that there is opportunity for improvement to how local food is accessed, and that these mobile applications have the possibility to address food accessibility, economic vitality and sustainability, with a lower negative impact on the transportation environment.

Recommended Citation:
Bovbjerg Alligood, Anna (2017). Finding a (Food) Way: A GIS Modeling Approach to Quantifying Local Food Transportation Systems, University of Washington Master's Degree Thesis.
Thesis: Array