Prediction of public transportation occupation based on several crowd spots using ordinary kriging method

Abstract

It is known that traffic congestion problems had occurred more frequently and had become much worse nowadays, not only in big cities but also in smaller cities. Therefore, we are required to find a tool to help us developing an analytical model that can be used to govern a public policy regarding traffic management. In the process of building the system there are several issues that have to be addressed. One of those issues is mostly related with how to determine the public transportation routes, how to determine the type of public transportation, and how to determine the optimal amount of public transportation needed for each route. In this research, we limit our study of public transportation system only in Bandung City area. The problem of determining the route of public transportation system can be solved in several ways, for example, by determining the required amount of public transportation based on the estimation of its occupancy in each route. In this research, we use ordinary kriging method based on spatial analysis, to predict occupancy of the public transportation. We conducted daily observations on weekdays and weekend, to collect the traffic data. The result of our research is a 94 Sri Suryani Prasetiyowati et al. prediction of occupancy of public transportation for each crowd spots area. On the weekdays Gaussian theoretical semivariogram is chosen, whereas on the weekend is spherical. Based on the distribution pattern of the occupancy of public transportation as represented in contour-display figure, we conclude that the average occupancy of public transportation in Bandung City is not optimal. Amount of public transportation mode in Bandung exceed standard of transportation public requirement.