东南大学学报（自然科学版）2017，Vol.47Issue(5)：1042-1047,6.DOI:10.3969/j.issn.1001-0505.2017.05.032
基于贝叶斯网络的城市常规公交服务满意度分析模型
Assessment of passenger satisfaction with urban bus service quality using Bayesian networks
摘要
Abstract
To discover the underlying relationships among service aspects and assess their influences on passenger overall satisfaction with regular bus service,a Bayesian network (BN) was developed by the greedy thick thinning (GTT) and maximum likehood estimation (MLE) algorithms based on 609 questionnaires from 2013 regular bus service survey in Nanjing,China.The established model could elaborately capture the network relationships among service attributes and passenger overall satisfac tion.The validation result proves that BN is more applicable and efficient.Among the 7 main contributors inferred by the belief propagation algorithm,the negative effect of waiting time is the most significant at-31.4％ and the positive effect of onboard environment is the most significant at 33.7％.The punctuality and the convenience have larger influence ranges,from-26.6％ to 23.1％ and from -20.9％ to 27.3％,respectively.The seat availability,environment at stations,and the provision of airconditioning onboard are the most influential with the absolute ranges up to 55％.关键词
贝叶斯网络/公交服务/乘客满意度/贝叶斯推理Key words
Bayesian networks/public transit service/passenger satisfaction/Bayesian inference分类
交通工程引用本文复制引用
吴静娴,杨敏..基于贝叶斯网络的城市常规公交服务满意度分析模型[J].东南大学学报（自然科学版）,2017,47(5):1042-1047,6.基金项目
国家自然科学基金资助项目(51338003,51378120). （51338003,51378120）
