The Maryland State Highway Administration (SHA) has developed an Enterprise Geographic Information System (e-GIS) framework that provides users, managers and executives the ability evaluate data in a geospatial capacity, providing efficient analysis and promoting data driven decisions. Bringing the Maryland Statewide Transportation Model (MSTM) into this e-GIS platform allows non-modelers to incorporate model data in decision-making and allows linkages with many other data sources available through the e-GIS platform.

The MSTM includes the entire state of Maryland, Delaware, and the District of Columbia, as well as portions of Virginia, Pennsylvania and West Virginia. The MSTM also includes the state’s two metropolitan planning organization (MPO) regions that maintain their own travel demand models. These MPO models are well established and their networks have evolved over the years and are now trueshaped to better align with the existing roadway geometry. This process is known as conflation.

Network conflation, a complex process widely used in transportation modeling to support spatial data unification and information exchange, is the combination of information from one or more network models to create a unified network model in spatial and attribute aspects. The network conflation problem is matching the same roads (links) between two networks and updating/ adding new roads into a destination network from source networks and transferring attributes. A hybrid approach, combining automation and manual conflation, was used to create one statewide True Shape network. Automating the conflation process reduced time and cost, and improved accuracy. Various levels of automation were achieved depending on the similarity of the source and target networks and how close spatially the network links/nodes are to each other.

A set of customized GIS application tools were developed to conflate the MSTM network links/ nodes features to their corresponding links/nodes in the True Shape networks, by creating crosswalk (relationship) tables that recorded matched nodes and links between the target and source networks. This crosswalk approach proved to be effective and efficient in accomplishing the task objectives and resulted in a high percentage of records matched.