U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
|This report is an archived publication and may contain dated technical, contact, and link information|
Publication Number: FHWA-HRT-07-046
Date: August 2007
Model Minimum Inventory of Roadway Elements—MMIRE
PDF Version (1.09 MB)
PDF files can be viewed with the Acrobat® Reader®
DEVELOPMENT OF MMIRE
MMIRE has been developed through a multistage process involving identification of potential elements, review and comment by data collection and data use experts, modification of MMIRE element listing based on that review, and developing of proposed coding schemes for each element. The following narrative describes that process.
IDENTIFICATION OF POTENTIAL MMIRE ELEMENTS
As noted above, the goal of MMIRE is to define critical safety data inventory elements—those elements needed by State and local agencies to conduct their internal analyses, and those elements required by existing safety analysis tools and resources. However, this effort was not limited to existing data element needs. Thought also was given to critical elements that will be needed for use in future analysis tools and program decisions. For example, pedestrian and bicycle safety are both high-priority areas for both State and local jurisdictions. As in program decisions and knowledge developed for other road users, both crash and exposure data are needed by the analyst. While pedestrian and bicycle exposure data currently are not collected, these counts may be possible by using digital-image-based methods now under development. Such critical, but nonexistent, elements are included in the proposed MMIRE structure. The research team based their choice of proposed elements on five specific existing tools and resources:
These four original inventory element listings were reviewed in detail by the project team staff, and potential MMIRE elements were identified. In addition to elements in these four listings, the project team also included additional elements that are felt to be critical MMIRE variables. These additions were based primarily on four other sources:
This element identification effort produced a listing of over 150 potential MMIRE elements. The listing was converted to a matrix for use in the review workshop described in the section below. For each data element, the project team provided a proposed priority—1st, 2nd, and not recommended. This priority was based on a combination of factors including the requirements of the four major data sources and tools noted above (i.e., MMUCC was added later) and the team’s knowledge of current and expected future analysis and tool needs. In some cases, the difficulty of data collection was considered in this prioritization.
In addition to the draft priority, the matrix also provided information on the presence of or requirement for each data element in the four basic data sources and the "level of priority" provided by each source (e.g., HPMS has both Sample Section and Universe elements; SafetyAnalyst has "mandatory," "optional," and "supplemental" elements).
Finally, the project team searched for each potential data element in inventory files for 20 States that were available to them—the eight current HSIS States and 12 other States. Each State file was examined to determine which of the MMIRE elements being considered were collected by that State. For example, information on "county" is captured in 18 of the 20 State systems examined (i.e., 90 percent). This information was also provided to the reviewers in the workshop to give them some sense of the current presence of each variable in the sample State data systems.
THE MMIRE REVIEW WORKSHOP
A one-day MMIRE review workshop was held August 3, 2006, in conjunction with the International Traffic Records Forum (ITRF). As indicated above, the purpose of the workshop was to have State and local agency data collectors and data users review and provide feedback on the potential data elements in the draft MMIRE matrix described above. To ensure input from State and local DOT data collection managers, State and local data users, roadway safety researchers, and other roadway data experts, the workshop participants include both an invited group of attendees whose travel expenses were funded by FHWA and other ITRF attendees who signed up for the workshop. The workshop included 34 attendees—18 from State DOTs, 2 from local DOTs, 6 from the USDOT, 6 roadway safety researchers, one data-product vendor, and one representative from AASHTO. A full list of attendees is provided in Appendix A.
The above-described MMIRE draft matrix was presented to the attendees, who were asked to:
MODIFICATION OF DRAFT MMIRE ELEMENT LISTING
Following the workshop, the project team incorporated the feedback from the workshop participants into the final proposed MMIRE elements. Changes incorporated included addition of a small number of new variables and deletion of a smaller number of proposed variables, minor variable name changes, and changes in the priorities for some variables. The most significant changes involved (1) reorganization of sections of the matrix, (2) changing what were originally "intersection" descriptors to include additional "junctions" such as mid-block pedestrian and bicycle crossings, and (3) a change in how the "priority" of each element is defined. With respect to the latter, while the draft priorities discussed at the workshop could be affected by the anticipated difficulty in data collection, the discussion there resulted in the final priority being only based on the importance of the element in safety analyses, and a new variable was added estimating the level of difficulty of data collection.
DEVELOPMENT OF PROPOSED CODING SCHEME FOR EACH MMIRE ELEMENT
Following finalization of the proposed listing of MMIRE elements, the project team defined a proposed coding scheme for each element. For existing elements found in any of the five data sources noted above, the team reviewed the proposed coding from each. As would be expected, the coding for a given variable differed to some degree across the sources. The final decision was made based on the expected use in future analyses. Thus, additional weight was given to the schemes now used in both the IHSDM and SafetyAnalyst and anticipated for use in future tools. However, an attempt was made to ensure that the proposed scheme also was compatible with HPMS codes, on the assumption that States may have used that coding when developing their inventory files, thus reducing the amount of recoding that might be necessary in a conversion to the MMIRE formats.