摘要
通过对公交运力配置影响因素的分析,将公交运力优化配置问题定性为多目标优化问题。首先通过分析公交客流量时空分布的复杂性和规律性,结合优秀算法对客流量的预测进行建模,提出了基于Elman神经网络的公交客流量预测流程。并用实际客流量数据对未来的客流量进行了实例预测。在完成了客流量的预测后,在现有配置模型的基础上,引进基尼系数,作为新的评价指标,通过量化表示,将资源配置均衡度作为新的目标函数加入原模型,加以改进,从而建立基于基尼系数的多目标优化模型。并对新旧模型进行了对比,对各站点的客流量分担率进行了优化。
According to the analysis of influencing factors of public transport capacity, this paper characterized optimal allocation of public transport capacity as multi-objective optimization problem. First, by analyzing the complexity and regularity of the spatial and temporal distribution of bus passenger traffic, it models the forecast of passenger traffic by using the excellent algorithm, proposed the bus passenger traffic forecasting model based on the Elman neural network. And the instances forecasted future passenger traffic using acturally passenger traffic data. Then, based on the original model, it introduced the Gini coefficient as new evaluation indicators. After quantization, the degree of resource allocation equilibrium is viewed as a new objective function by putting in the original model to be improved. It established the multi-objective optimization mode based on Gini coefficient. And it compared the original model and the new model, optimized site traffic sharing rate.
出处
《信息技术》
2013年第2期117-120,共4页
Information Technology
