Choosing the fastest path with the aim of saving time or associated costs, has nowadays play a crucial role in both freight and passenger transportation. With the use of traffic simulation models currently available, travel times in urban streets and highways where average speed primarily depends on traffic volume which is apparently very high most of the time, can be obtained. However, in rural roads, especially in Thailand, where congestion rarely occurs due mainly to very low traffic volumes, travel time is rather dependent on roadway characteristics than traffic flow. As a result, the traffic simulation models available can not be used to predict the travel times in rural roads in Thailand effectively. This research therefore identifies roadway characteristics affecting travel times in rural roads, and subsequently develops the mathematical model, based upon road characteristics identified, for predicting the travel times in rural roads in Thailand, by using multiple linear regression. Data on roadway characteristics and travel time, used to calibrate the mathematical model, is collected onsite from 65 rural roads in Thailand and more than 20,000 vehicles, respectively. The research reveals that speeds and travel times for both passenger and freight transport in rural roads in Thailand are influenced by landuse, horizontal and vertical alignment. Number of bridges and access density are found to have effects on travel times and speeds for passenger transport while travel times and speeds for freight transport appear to be affected by lane width of the road.