基于电力客户的历史数据,采用客户的基本属性、用电行为、缴费行为、客户信用、行业前景信息等多个维度确定模型所需指标体系。通过指标的相关系数矩阵及信息值(information value,IV)筛选出最终进入模型的指标变量,同时采用最优分组的...基于电力客户的历史数据,采用客户的基本属性、用电行为、缴费行为、客户信用、行业前景信息等多个维度确定模型所需指标体系。通过指标的相关系数矩阵及信息值(information value,IV)筛选出最终进入模型的指标变量,同时采用最优分组的方法对变量进行分组,并进行证据权重转化(weight of evidence,WOE)。基于处理后的数据,运用逻辑回归算法构建用电客户电费风险预测模型,并依据得到的模型结果量化输出变量标准评分卡表,从而将客户划分为高风险、中风险和低风险用户,为不同的用户采取差异化的营销措施提供依据。展开更多
An in depth exploration has been made of telecommunication tariff and its decision support system. It is the first attempt to conduct such a study by the integrated use of econometrics, system emulation and system dy...An in depth exploration has been made of telecommunication tariff and its decision support system. It is the first attempt to conduct such a study by the integrated use of econometrics, system emulation and system dynamics. A practically demanded cost model of the telecommunication tariff decision is proposed. The system has been verified with real data.展开更多
In this study, strategic electricity market scenarios are considered in a grid of Scandinavia. This multiple-objective decision environment includes the allocation of a number of renewable forest fuel procurement chai...In this study, strategic electricity market scenarios are considered in a grid of Scandinavia. This multiple-objective decision environment includes the allocation of a number of renewable forest fuel procurement chains to a combined heat and power plant in Finland. The decision environment includes also electricity procurement from Sweden and Russia. The environment is further complicated by sequence-dependent operations of the local procurement chains during different periods. Due to the complex nature of the environment, multiple-objective methods cannot be directly used to solve the electricity production problem in a manner that is techno-economically relevant to the forest energy industry. Therefore, local and time-varying parameters were measured in local wood procurement conditions to improve the solution method. Using these measurements the smart decision-support system automatically adjusted the multiple-objective methodology to better describe the combinatorial complexity of the production sector. The properties of this methodology are discussed and three scenarios of how the system works based on local real-world data and optional feed-in tariff of green electricity are presented. The Finnish electricity market is subject to policy decisions regarding green energy production regulations. These decisions should be made on the basis of local techno-economic analysis presented in this study accounting for the effects of forest operations on the electricity production and import.展开更多
文摘基于电力客户的历史数据,采用客户的基本属性、用电行为、缴费行为、客户信用、行业前景信息等多个维度确定模型所需指标体系。通过指标的相关系数矩阵及信息值(information value,IV)筛选出最终进入模型的指标变量,同时采用最优分组的方法对变量进行分组,并进行证据权重转化(weight of evidence,WOE)。基于处理后的数据,运用逻辑回归算法构建用电客户电费风险预测模型,并依据得到的模型结果量化输出变量标准评分卡表,从而将客户划分为高风险、中风险和低风险用户,为不同的用户采取差异化的营销措施提供依据。
文摘An in depth exploration has been made of telecommunication tariff and its decision support system. It is the first attempt to conduct such a study by the integrated use of econometrics, system emulation and system dynamics. A practically demanded cost model of the telecommunication tariff decision is proposed. The system has been verified with real data.
文摘In this study, strategic electricity market scenarios are considered in a grid of Scandinavia. This multiple-objective decision environment includes the allocation of a number of renewable forest fuel procurement chains to a combined heat and power plant in Finland. The decision environment includes also electricity procurement from Sweden and Russia. The environment is further complicated by sequence-dependent operations of the local procurement chains during different periods. Due to the complex nature of the environment, multiple-objective methods cannot be directly used to solve the electricity production problem in a manner that is techno-economically relevant to the forest energy industry. Therefore, local and time-varying parameters were measured in local wood procurement conditions to improve the solution method. Using these measurements the smart decision-support system automatically adjusted the multiple-objective methodology to better describe the combinatorial complexity of the production sector. The properties of this methodology are discussed and three scenarios of how the system works based on local real-world data and optional feed-in tariff of green electricity are presented. The Finnish electricity market is subject to policy decisions regarding green energy production regulations. These decisions should be made on the basis of local techno-economic analysis presented in this study accounting for the effects of forest operations on the electricity production and import.
