In the second part, linear regression was practiced as an AADT estimation process, which was primarily based on known or accepted AADT values on the neighboring state and local roadways, population densities and other social/economic data. In the first part, this thesis performed a study to compile and compare current procedures or methodologies for the estimation of traffic volumes on the roads where traffic counts are not easily available. The findings of this study can lead highway agencies to improve the cost-effectiveness of their short-period traffic counting programs. Introduced adjustment factors can reduce the values of CV for all counts taken during winter months. The application of an adjustment factor to remove the effect of cold and snow from short-period counts is also included in this study. Frequent changes in temperature and other weather events cause significant variation in traffic volume, which results in an increase in CV values for counts taken during winter months. It is also found that the lowest values of CV result for counts taken at commuter sites, and the highest values are observed for recreational sites. The study results indicate that 48-h count seems to be the most cost-effective counting interval during both summer and winter months. The coefficient of variation (CV) is used as the relative measure of deviation for counts of different durations to measure the accuracy of short-period traffic counts. The sample data constitute six different durations of counts (12-, 24-, 48-, 72-, 96-h, and 1 week) taken during summer and winter months. Four types of road sites: commuter, regional commuter, rural long-distance, and recreational sites are studied. The investigation is based on 11-year sample data collected using permanent traffic counters at various locations in Alberta, Canada. The main intent of this study is to investigate the accuracy of short-duration traffic counts conducted during winter months. Nevertheless, these still will have an impact on investment decisions, crash rate calculations, travel demand model validation, and other analyses. With adequate attention, (average) errors in AADT estimates can probably be reduced to the 10 percent level.

These include sampling in spring and summer months (on weekdays), exercising greater caution with counts on multi-lane and low-AADT roadways, pursuing appropriate site assignment to ATR groups, and recognizing the effects of distance to the sampling site. The analytical results of this investigation suggest a variety of recommendations for agencies seeking to reduce and appreciate errors in their AADT estimates. Spatial and temporal (inter-sampling year) extrapolations can further add to such error, in a sizable way. Overall errors at ATR sites are found to be highest (averaging 24.6%) when data come from misclassified sites on weekends. Errors as a function of distance to the nearest sampling site are also studied, using predictions of network travel patterns in Austin, Texas. Variations in AADT estimation errors are investigated across roadway and area types, for both Minnesota and Florida automatic traffic recorder (ATR) sites. This research quantifies the level of uncertainty in AADT estimates and compares these across sampling strategies. AADT values provide a key variable in many models and policy decisions however, these are simply rough estimates of traffic counts along the vast majority of roadway sections.

Mdot traffic count