航班需求预测理论和方法是航空公司收益管理的技术支柱之一。构建基于加法、乘法、回归的组合预测模型,利用合作博弈中的Shapley值确定各个模型在组合模型中的权重,以此对航班舱位等级座位需求进行预测。算例验证表明该组合预测模型的预测精度高于选定的各单项预测模型,可以作为预测航班需求的有效工具。%Flight demand forecasting theories and methods is one of the pillars of the airline revenue management technology. Build combined forecasting model based on aggression model,addition model,and multiplication model,using a game theory the Shapley value allocation model, to determine reasonable weights for each single model and in order to build demand for flights combined forecasting model to predict the demand for flights. Using historical data from a flight reservation,example verifying that the accuracy of the combined forecasting model is higher than the individual selected prediction model,can be used as an effective tool to forecast demand for flights.
展开▼