A near-miss is an unplanned event that can precede events in which a loss or injury could occur. Therefore, it is an indicator leading to an accident. Near-miss analysis does not look at what happened but look into what could have happened. Serious conflicts are the result of a breakdown in the interaction between the road user, environment and vehicle, which leads to traffic accidents. As the similarity between accidents and serious conflicts is striking, accidents can basically be prevented by isolating and handling the conflicts potentials. The Swedish Traffic Conflict analysis of near-miss accidents is adopted in this study to improve the traffic safety of a channelized junction with U-turn at LentengAgung, Jakarta. The junction is a two-way junction with a wide median and an island. The location has been indicated as an accident-prone location with high conflict rates. A set of traffic video-recordings employing a number of surveyors at different points of observations were carried out on-site to obtain real near-miss accidents. Prior to the survey, surveyors practiced judging vehicles running speeds until they reached a certain maximum error of judgment. Evasive actions such as braking, swerving and accelerating were recorded, and actions were classified into serious and non-serious conflicts based on the time-to-accidents and speeds. The results of the analysis show that almost all of the total recorded conflicts fall in the category of serious conflicts. An improvement scheme of the junction with reduced potential traffic conflicts is proposed, which can be expected to lower the accidents occurrences.

Investigate and measure the relationship between accident frequencies and the consistency of road geometric designs. Methodology: Several accident prediction models from straight section, curve section, and road segment data from variables that combine design consistency. Poisson and negative binomial regression approaches are used for model development. The model obtained can later be used as a quantitative tool for evaluating the impact of design consistency on road safety. From the model obtained, from all the results of the model, it was decided to use the model of negative binomial regression because the model has proven to be undispersed, so that the model results obtained are better than poisson regression. Obtained the final model with exponential and power forms. In addition, the model of the combined data results represents more existing field data, this is caused by several factors that can affect these results. Then, it was found that the model obtained could consider design consistency so that the model could identify inconsistencies better and accurately reflect the resulting impact on safety. Many road conditions in Indonesia are still incompatible with existing geometric road design standards. Consistency in the geometric design of the road must be considered, given that all elements in the geometric design of the road have an influence on the number of accidents that occur on the road. So more consideration is needed for new roads that are designed according to existing standards so as to produce a good road, hoping to reduce the number of accidents that occur significantly.

Pedestrian-vehicle crashes are the results of a combination of infl uencing factors including the driving enviroment. This paper looks into the driving environment factors in pedestrian crashes injury levels on road links in Jakarta and the surounding cities which contribute to the city traffi c generation. The vehicle-pedestrian accident data used were obtained from the 2016 Indonesian national police accident database covering 4,646 pedestrian accidents on road links from Jakarta, Depok, Tangerang Selatan, Tangerang and Bekasi, Indonesia. Various factors were analyzed including crash level severity, month of occurrence, weather condition, lighting condition, road function, road class, road type, road surface condition and road status. As injury levels were categorized into slight injury, severe injury and fatal injury and it was assumed that the dependent variables which were crash injury levels could not be perfectly predicted from the independent variables, Multinomial logistic regression (MNL) was used in the analysis to predict the probability of different categories of dependent variables. It was found that the relative risks of pedestrian accident risks factors changed with different categories both in terms of fatal and severe injuries. One of the fi ndings shows that the risk of having severe injuries would decrease by 40.2% on national roads, by 70.5% on provincial roads and by 53.5% on urban roads. The results can be expected to be referred to in the improvement of pedestrian safety level and in the development of related measures.

No abstract

A median is required for a two-way road to separate the opposite traffic and prevent head-on collisions. In road capacity calculation, the median factor contributes in terms of its existence regardless of the difference in median types. Road capacity is determined by a number of geometric factors such as road types, width of carriageway, shoulder/curb characteristics, and the presence/absence of medians, etc. The contributions of these factors are represented by the coefficients in the capacity calculations. Despite the different types of medians, the Indonesian Highway Capacity Manual (IHCM) does not adopt different coefficients to accommodate the effects in the capacity. The aim of this study is to obtain an adjustment factor for road capacity calculation based on median types. The method of this study adopts video recordings of real traffic moving along three different types of medians: raised medians, fenced medians, and line medians. As it is assumed that the effects of different median types are expressed in the vehicles' safety distances from medians, the capacity of the road will also vary by types of medians. The adjustment coefficients for roads with raised medians, fenced medians, and line medians obtained are: 0.79, 0.78, and 0.81, respectively. The results of this study confirm that in addition to the presence of the medians, the types should essentially be considered in calculating the road capacity. The result of this study will contribute to the enrichment of the road capacity calculation in the IHCM.