multiple regression analysis model

Today, travel demand modeling techniques and use of new knowledge, an important part of transportation engineering and traffic planning in urban transportation within the accounts is, So that in recent years using modern scientific techniques and promising results have come at too. The key question now is the paper where the models of travel demand characteristics have what has been, and to obtain the actual amount of travel demand forecasting and its effect on traffic flow and reduce the necessary adjustment has? To answer this question of the scientific method combined Geographic Information System (GIS), remote sensing (RS) and multivariate linear regression analysis for spatial modeling of demand in the area 6 of Tehran has been used. The final results indicate that the variable (C3), i.e. 14.23 percent of the population of the region with the highest degree of correlation coefficient with the dependent variable in the model has been, fret the commercial user variables (C16) with 11.9-percent ratio and the number of employees working in the region 3.10 - percent respectively second and third parameters of the model are, Order through the multivariate linear regression equations obtained. The variable number of units Business region (C11) with a coefficient of -0.008930 per cent minimum impact on travel demand modeling has shown that spatial. So for testing, evaluation and final results of modeling spatial Validity travel demand, the criteria for test, The criteria for testing, fitness model results and comparison with results of existing models of multivariate regression analysis technique is derived. Final results were used criteria value (R2 = 0.73415936), ((MARE = 78.628), (RMSE = 1.43) percent is derived. According to the results, model where travel demand capabilities required for calculating the actual amount of travel demand for transportation within the city and its effect on traffic flow and decrease predicted in the study area has enjoyed.