VALIDASI PREDIKSI BANGKITAN PERJALANAN DARI METODE PENGINDERAAN JAUH QUICKBIRD DI KOTA MAKASSAR

As a new breakthrough, along with the development of geoinformatics in Indonesia, Quickbird as a satellite with the ability to produce a high spatial resolution that can display image of the building such as houses, office building and other buildings accurately. It allows recording of land use components such as: dwelling units, office buildings, green spaces, and other types of built environments and it can replace the conventional method of trip generation data collection. The purpose of this research is to make a model that can predict not only trip generation by land use data base and the building area but also the approximate correction factor. The variable of this research which is suitably used with Quickbird image is houses/building area. Research site selected base on the result of remote sensing processed by a student doctoral of geography UGM in Makassar with six district as a zone boundaries. The research method used empirical method and analytical tools by Microsoft Excel dan SPSS 17 (Statistical Product and Service Solutions) softwares. Regression analysis method is used for modelling, with the first consideration of regression form which can lead into linear or non-linear result. The analysis model obtained a linear regression form with the formula of Y= 1,194 + 0,062.X and R Square 0,933. As for the results of the t-test, there are no difference between the model estimated and observations found. The trip estimation model then formed origin-destination matrix, which was previously performed backcasting projection to the year 2007, to be calibrated afterwards with a matrix of origin-destination of year 2007. Through the calibration process a correction factor (FCr) for PJ model would be generated, with the formula of FCr = 1826 + 0.831X, so the model of Makassar is Fcr = 1829 + 0,831.YPJ