Reducing data dimensions for systems engineering and risk management of transportation corridors

The agencies responsible for transportation corridors tend to hold large volumes of data that can be relevant for both operations and planning. Meanwhile, it is a challenge for these agencies to prioritize their investments addressing risk, benefits, and costs. The agencies recognize an opportunity to improve project selection and programming with a centralized database of performance measures that will aid a consistent application of evaluation metrics. To support the identification, planning, and selection of highway transportation projects, this research has used a data structure known as dynamic segmentation for cross-referencing heterogeneous data sources including projects, traffic, safety, bridge and pavement conditions, etc. The result is a multiscale method for agencies to utilize big-data analytics and multicriteria decision analysis to assemble evidence for project selection and prioritization. The paper describes an approach to (1) visualize multiple attributes along linearized road corridors to identify future projects, minimize conflicts with current projects, and identify potential project synergies and (2) prioritize projects to the Top-20 with multiple performance factors. The methods are demonstrated at several geographic scales. The effort supports a fast, repeatable, and evidence-driven prioritization of projects and provides a complement to the use of electronic map data that has been overwhelming the available computing resources.