A household car ownership modeling is crucial in understanding the impact on an individual's or a family's travel behavior in traveling demand analysis. Trips or tours as a unit of analysis can be used in the modeling of car ownership demand for analyzing travel needs. Machine learning is widely used to describe a car owner's decision since the machine learning model was specifically designed to give more accurate predictions through a variety of mechanisms. This research presents car ownership modeling using two types of machine learning models, including decision trees and neural networks. The impacts of socio-demographic attributes on household car ownership demand are discussed and compared against these two models after adding the main attributes of variables from tour-based models. Data was collected from 2,015 households surveyed in Khon Kaen Province, Thailand, conducted in 2015. The outcomes indicate that the machine learning model can be used to predict household car ownership. It also found that when using the default parameters across all datasets, whereas the neural networks provide a more accurate result than the decision tree algorithm. However, in cases where the household car ownership prediction from the dataset with add attributes of the key variable used in tour-based models. In that case, the neural networks algorithm would give a prediction accuracy that corresponded to results as found from the prediction using a dataset with only the household's socio-demographic attributes.

Abstract. This research aims to study a correlation between a pavement skid resistance and wet-pavement related accidents in order to determine the minimum friction threshold so called 'Investigatory Level (IL)' for the roads in Thailand. An accident database, a skid resistance database and a traffic volume of road network from the department of Highways, totalling 19 routes and 386kilometers, were used in the analysis of this study. In the analysis, 500-meter subsection intervals are used to determine a correlation between an average pavement skid resistance and wet-pavement accident rate using a non-linear regression analysis model. It was discovered that the pavement skid resistance has a major influence on the accident rate, depending on various types of road geometry. Moreover, the preliminary investigatory level is determined by using a past accident rate information as a reference for an investigatory level basis. We discover that the single carriageway (non-event) road is at highest threat due to low pavement skid resistance. The recommended investigatory level for each of the five road geometries all site, single carriageway, dual carriageway, horizontal alignment and curve road categories are 035, 0.50, 0.30, 0.30, and 0.40, respectively, which are in accordance to the investigatory level values in other countries.

This article aims to analyze the effect of median U-turns and estimate the capacity of primary highways in Thailand using a traffic micro-simulation model. Six-lane and four-lane primary highways were selected for the study. The base condition results determined that the maximum capacity of a six-lane primary highway was 2,130 passenger cars/hour/lane, while the four-lane primary highway capacity was recorded as 2,194 passenger cars/hour/lane. Both results were slightly higher than those of the HCM2010 approach. Under prevailing conditions, both sections exhibited lower capacities than the HCM results by approximately 33.7% and 19.8% for the six-lane and fourlane primary highways, respectively, causing the impact of the median U-turn and highway characteristics in Thailand to directly affect traffic and driving behavior. Using the micro-simulation results, an equation was also regressed for estimating the capacity resulting from the impact of the median U-turns and heavy vehicles. These results may be used as guidelines for the design and analysis of multilane highways in Thailand.

This article purposed to present the maximum capacity and to develop the equation in the capacity estimation of 4 types of four-lane highways by using the micro-simulation model. Regarding the analysis, the factors affecting the capacity include access-point, heavy vehicles and median u-turn. According to the study, it was found that the maximum capacity of the four-lane highways in type 1 is 2194 passenger car/hour/lane. In the descending orders, the maximum capacity of the highways with four lanes in type 2, 3, and 4 are 2161, 2094 and 2017 passenger car/hour/lane, respectively. At the same time, the maximum capacity of the prevailing condition in the study is 1300-1600 vehicles/hour/lane, which is different from the HCM2010 method for 20-30%, due to the Thai's highway characteristics directly affect the traffic and driving behavior. Median u-turn affects the four-lane highways in type 1 and 2 the most, while access-point factor has the most influence towards type 3 and 4. In addition, the author has developed the equation models for capacity estimation, which the result derived from the relationship between the capacities and affecting factors. It was aimed at using as the guidelines in the capacity assessment of Thai's four-lane highways in the future.