Abstract: Car ownership depends on the land use patterns in a city． Land use characteristics such as employment，housing， leisure，education and shopping opportunities in each zone affect the choice of destination and mode in the short and medium term． Socioeconomic factors including age，gender，income，and family size also affect the choice of mode and destination in the short and medium term，while the choice of housing location is affected in the long term． This research models car ownership by the estimating multinomial logit model，and Riyadh，Saudi Arabia and Melbourne，Australia are the two cities used as case studies． Aggregate level land use data as well as disaggregate level individual，household and journey-to-work data are used to determine the effects of land use and socioeconomic factors on household decisions to own zero，one，two or more cars． The model results determine that urban sprawl is the main factor that affects household decisions to own more than two cars in Riyadh and one car in Melbourne．

Key words: land use and socioeconomic, car ownership, multinomial logit model

摘要： 汽车保有量取决于城市的土地开发模式。在一定时期内,一个地区的就业、住房、休闲、教育和购物条件以及 家庭成员的年龄、性别、收入和家庭规模都会影响到人们对出行目的地和出行方式的选择,而家庭住房位置的影响 则是长期的。应用沙特阿拉伯利雅得和澳大利亚墨尔本两座城市数据,分析了城市总体用地水平和人均水平、家庭 人口及其距离工作地点的路程等对家庭汽车拥有量的影响,建立了家庭汽车拥有量多项式逻辑模型。模型表明,城 市扩张是影响家庭汽车拥有量的主要因素,利雅得市家庭汽车拥有量在2辆以上,而墨尔本市不少于1辆。

关键词: 土地利用和社会经济, 汽车拥有量, 多项式逻辑模型