The relationship between traffic congestion, travel demand, and supply of roadways is investigated by use of the travel rate index, a congestion measure developed by researchers at the Texas Transportation Institute. Data for the top 138 urbanized areas (by population) were assembled for developing a least-squares regression model. The travel rate index was selected as the response (dependent) variable. A variety of explanatory variables were used to address highway and transit supply and travel demand-related factors. The partial regression coefficients measured the effect of each explanatory (independent) variable on congestion (as measured by travel rate index), holding all other independent variables constant. The results of the multiple regression analysis indicated a negative correlation between freeway lane miles and combined travel rate index. Additionally, a strong positive correlation was observed between combined travel rate index and population density and net land area. A positive correlation was observed between combined travel rate index and bus transit service revenue miles. Principal arterial lane miles and rail transit revenue miles variables were not observed to be significant for explaining traffic congestion and were removed entirely during the stepwise regression. The results indicated that the best predictors among the tested variables were freeway lane miles, population density, net land area, and bus revenue miles. When used together, these predictors accounted for approximately 61% of the total variation in the dependent variable, combined travel rate index.
Abstract