Although the priority from the outset was to make data sets used in the SHRP 2 Reliability research available for many decades, it quickly became apparent that the data would not be understandable or useful without sufficient contextual information. Thus, it was determined that all types of data should go into the Archive, both structured and unstructured. Structured data include comma-separated data, other flat files, and relational data. Unstructured data include data dictionaries, reports, presentations, video, spreadsheets, computer code, and other digital objects. In the parlance of the Archive, structured data are called “Data sets” and unstructured data, “Non—data sets.” The Archive’s home page shows the five main use cases: upload, search, visualize, download, and discuss. Also shown on the home page are different statistics about the Archive and the latest four artifacts that have been uploaded so a user can drill down into the most recent additions. “Artifact” is the term used for each data set and non—data set in the Archive. There are three ways to view the data once in the Archive: (1) a word search; (2) a search of the Archive that shows data sets by geographical location on a map of the country; and (3) a search by SHRP 2 focus area and project listing. The word search is a simple text search. A “search of the Archive” produces a literal spiral of data sets at each location; the user can click on each one and drill down to explore what the data are. A user can also select whether to look at a data set or non—data set and provide further filters for the selection. For example, a user can filter a data set according to whether it has speed, occupancy, and flow data. Finally, a user can search by project by first clicking on a focus area, then clicking a project listed under that focus area. An informative description accompanies each project to acquaint the user with the project objectives, key considerations in undertaking the research, and some of the important research products produced. Every project has metadata, and the metadata for data sets includes a data dictionary. The Archive has substantial visualization capability for a user to preview and evaluate whether a data set is of interest. This feature is only for data sets, not non—data sets. This capability offers three different types of visualization. A user can visualize the first 300 records of a data set. On a map of highway facilities, reminiscent of a GPS navigation map, a user can see the precise location of the traffic detectors used to collect the data. Finally, a user can graph the relationships between different numerical fields in the data set. An example might be a scatter plot of the relationship between speed and flow. The Archive was developed with the flexibility to be a dynamic and living system so that research that is new or related to the original Reliability research could be added to the system. There are significant administrative, maintenance, and operations costs to this capability. As of the end of the Archive contract, only SHRP 2 Reliability-related data had been entered into the system. The Reliability Technical Coordinating Committee recognized early in the formation of the Reliability research program that there would be a significant benefit to the research community, traffic engineers, planners, data managers, and others if the data from all the Reliability-related research projects could be preserved and easily accessed over the next 25 years or more. The inspiration for a repository was the Long Term Pavement Performance monitoring system from SHRP 1. Because it was not clear how to proceed, the TCC determined that first there should be a feasibility study, and if that study found that it was both possible and desirable to develop the Reliability Archive, the Archive would be built. At the outset, the TCC also set aside resources to provide support to enable contractors to populate the Archive with data. The feasibility study strongly suggested that developing the Archive was both desirable and feasible. The study called for the use of open source software and determined that both the software to realize the functional capability of the Archive and the data should be stored in the cloud. The TCC recommended that the Archive be developed. (Author/publisher)
Samenvatting