Estimasi Matrik Asal Tujuan Ddari Data Lalu Lintas dengan Metode Estimasi Inferensi Bayesian Menggunakan Piranti Lunak EMME/3 (Studi Kasus Kota Surakarta)

摘要

ABSTRACT Wulan Septiyani, 2015, Origin-Destination Matrix Estimation From Traffic Count Using Bayesian Inference Estimation Method Using EMME/3 Software (A Case Study of Surakarta City). Thesis. Civil Engineering Department Faculty of Engineering, Sebelas Maret University Surakarta. Transportation problems such as congestion, air and noise pollution, accidents, delay, and various safety issues can be arise if the need for movement on the rise. To overcome these problems is to create good transportation planning management traffic settings are good, and understanding the patterns of movement. One of the planning is done is to describe the pattern of movement using a origin destination matrix (O-D Matrix). This research aimed to calculate β value and estimate O-D Matrix in 2015 so the inter-zone movement can be known. In addition, to determine the validation level by comparing traffic flow resulted from the model and observation. This research was conducted in Surakarta, by dividing Surakarta into 65 zones with 51 internal zones and 14 external zones. The model used to estimate O-D Matrix using Gravity model with trip generation and attraction constraint. On this model there are parameters β that function as determinants of the magnitude of the spread of values that occur in each zone. Calibration method for obtaining parameters β that are later used in the calculation method of estimation results O-D Matrix Bayesian Inference (with Tanner barrier function) is the Newton-Raphson method of calibration. The calibration is done with a repetition of the process until convergence is reached the limit parameter values using the Matlab software application help. The value of the traffic volume is obtained by way of imposing a new matrix and enter a stream of observations (traffic count) to the road network by the method of loading the User Equilibrium. Test Validation using the coefficient of Determination (R ²). The value of parameter β by using application software Matlab is -0,00034. The result of the calculation by using EMME/3 shows the total number of estimation of movement Surakarta in the year of 2015 is 47549 smp/jam with level of validation (R2) is 0,803. Keywords: O-D Matrix, EMME/3, Matlab, Gravity Model