Preliminary Study for Modeling Train Accident in Indonesia Using Swiss Cheese Model

Abstract

Train is one of the mass transportation that famous in Indonesia, which continues to evolve and become a good choice for all class, including middle class and upper class. Train has special advantages which have high security, but it does not guarantee the least number of accidents. Seeing this fact, it is necessary to put more attention in finding a solution to prevent the accident by identifying the root causes of the accident using accident model. Swiss Cheese Model, one of the basic models of accident, acts as a framework for accident investigation. The train accident investigation would lead as a great example to have a better understanding of how the train accident could be happened. In recent years, the most common train accidents that occurred in Indonesia were slip incidents and train-to-train or train-to-another vehicle crashes. This research aims to implement the Swiss Cheese Model of train accident for developing countries. The developed country, in this case study was UK, that already implemented the model would be compared for the sustainability approach of the implementation in Indonesia. The result was showing that Swiss Cheese Model could be implemented in developing countries, especially in Indonesia. Based on this train accident, there are four factors that should be considered for understanding the accident: maintenance system, shift system, crossing system between rail and other road vehicles, and warning system. However, it was difficult to obtain the confidential and overall information regarding the train accident because of the readiness of the overall system to the accident model implementation. The layers of Swiss Cheese Model should be determined to understand the overall system of train in Indonesia. The recommendation of this research was the basic accident model, such as Swiss Cheese Model should be entirely defined to portray the whole situation of train accident. Moreover, to improve the level of understanding, the model could be developed into the newer system such as Accimaps, STAMP or HFAC.