Nyono, Harnen Sulistio, Achmad Wicaksono, Ludfi Djakfar
Rapid growth of transportation activities of people live near by railway areas increase traffic frequency rate. And as a result the number of vehicle collisions occurrence at unmanned railway crossing with no guards and gates has also increased. Some factors influence the number of accidents at unmanned railway crossing. Those factors are train engineering, road engineering and environment. The objective of this study is to modeling those factors. The modeling which represents the correlation of collisions at railway crossings with those factors is analyzed with Poisson regression analysis by using Gen-stat statistic program. Some observations were conducted at 33 unmanned railway crossings at DAOP VIII Surabaya, East Java, Indonesia. The results of the last Poison regression modeling show four significant factors influencing number of collisions at unmanned railway crossings. And those are train speed, flashing light, average of daily traffic (ADT) and distance traffic signs to railway crossings. Train speed seems to become the primary contributing factor to collision rates at unmanned railway crossings. However, it is recommended that besides the regulation of the train speed, other factors such as facilities and provision e.g. the distance of flashing light installation and Early Warning System (EWS) should be provided to with stand and hinder crashes at unmanned railway crossing effectively. © IDOSI Publications, 2013.
Department of Civil Engineering, University of Brawijaya, Malang, East Java, Indonesia
