Trip generation modelling forms the basic and first step in the four-step urban transportation planning process. Limited work had been conducted in this field in Palestine. The development of trip generation models requires extensive surveys, in depth analysis, and considerable time, cost, and effort, especially for a developing country in transition with limited resources. Therefore, this study deals with the assessment of the potential of spatial transferability of estimated trip generation models between cities.
The primary objective of this study is examining the transferability of the already estimated trip generation models for Jericho City to Salfit City, both in the West Bank of Palestine. To do so, the paper first considers estimation of trip generation models for Salfit City based on data collected through questionnaires investigating the trip generation pattern and the socio-economic characteristics of a randomly selected sample of households in the city. The general model of the total number of trips generated by a household is estimated, and then trip generation models based on trip purpose and trip making period were estimated. Five trip purposes were considered; work, educational, shopping, social, and recreational. The trip generation models classified based on five trip making periods during a day were also considered.
The transferability of the Jericho City respective general trip general model and those classified by trip purpose and trip making period was investigated using proper statistical testing. The results were compared with the trip generation models estimated for Salfit City. Two of the commonly used approaches, the native transfer and the updating model coefficients approaches, were utilized for testing model transferability. The outcome of the transferability tests show that the general model for the total number of daily generated trips by a household can be transferred as it exhibited acceptable relative transfer error, an indicator usually used in testing the soundness of transferability of models. Similarly, the two important models of the work and educational trips purposes, in addition to key morning trip-making period, can be transferable with good confidence. The positive results on the transferability of these models were obtained utilizing the two indicated approaches used in testing model transferability.
The research shows that when there is similarity in the cities socio-economic explanatory variables, which exist in the estimated model to be tested for transferability, transfer effectiveness will improve. Eventually, policies on developing urban transportation plans considering transferable general trip generation model and key specific trip generation sub-models can be adopted, which will save time, cost, and effort.