This paper explored the direct and indirect causal effects of adverse weather and winter maintenance actions on mobility as measured by traveling speed and traffic volume. Using Structural Equation Modeling (SEM) with the particular Categorical Variable Methodology (CVM), the paper analyzed the weather, maintenance and traffic data from 2001 to 2004 in the State of Iowa. Separate structure models were fit simultaneously to subgroups, which are disaggregated by road classification, Annual Average Daily Traffic (AADT) and speed limit. The analysis results suggest that despite winter maintenance operations that might slow down traffic during the hour when they are performed, winter maintenance operations have significant positive effects on improving speed, and the positive effects have been fully mediated through road surface conditions. Compared to the effects of plowing and applying chemicals, sanding is the least influential operational method to improve speed. Also the analyses suggest that the influences are different across road classifications, speed limits and different levels of AADT.
Samenvatting