Estimating parking generation rate for Karbala holy city using multi-variables approach

Abstract

Car parking planning, design, and management processes are very important to all cities and places to ensure efficient traffic system. Estimating the demand of car parking represents the significant start point for the success of these processes. Generally, there are many local and international estimating criteria, but such criteria need continuing update due to many reasons related to socioeconomic factors, lifestyle changes, development in technology, etc.  Moreover, the majority of these criteria depend on single parameter for the estimation of parking demand; such as bed or employee for hospital, gross floor area or employee for office, and so on.  The main aim of this research is to estimate the park generation rate for specific land uses depend on multivariable to increase the accuracy and limiting the effect of variation in parameters. Statistical analysis was conducted to create predicting models for each land use. The collected data was nominated for Karbala holy city, where different parameters are scaled for different city sectors. Groups of statistical models (i.e., simple, multi linear and nonlinear statistical models, and Weighted Linear Regression (WLR)) were used to create best representative relationship between the number of demands for car parking and multivariable parameters or factors affecting these lands used demands. Resulted statistical models were tested for best fit using statistical indices for model verification.  Results disclose the significant of multivariable model compare with simple models. Also, WLR model shows it validity compare with multi-regression model for almost land use models. Consequently, for more accurate estimation the multi variable models are initiated with continuous need for updating.