基于DEA的油田开发单元生产效率评价

油田开发单元生产效率评价是油田开发管理决策的一项重要内容。建立了油田开发单元生产效率评价的CCR模型、BCC模型、投影模型和超效率评价模型等DEA模型。选取采油井总开井数、含水上升率、注水开井数、新井投产井数、老井措施年累总井次、老井自然递减率和吨油操作成本等7个指标为输入指标,老井年产油、新井年产油、老井措施年增油和地质采油速度等4个指标为输出指标,运用DEA模型对某油田下辖的9个油田开发单元生产效率进行实证评价,得到各开发单元的技术效率、纯技术效率、规模效率、DEA投影、输入剩余和输出亏空。实例结果表明,CCR模型和BCC模型可以实现对油田开发单元在技术或规模方面的投入产出配置有效性的评价,帮助它们寻求技术或规模方面的改进建议;DEA投影模型为非DEA有效的油田开发单元实现投入产出最佳配置提供了改进方向和改进力度;DEA超效率模型则可实现不同开发单元在投入产出配置效率方面的评价排序。

Production Efficiency Evaluation of Oilfield Development Units Based on DEA

Production efficiency evaluation of oilfield development units is an important content of oilfield de-velopment decision-making.CCR model,BCC model,projection model,super-efficiency model and other DEA models for production efficiency evaluation of oilfield development units were established.It selects seven in-dexes i.e. total number of producing oil wells,increasing rate of water cut,number of producing water injection wells,number of new wells,the accumulated well times of old well measures,natural decline rate of old wells,and tonnage oil operation cost as input indexes,and four indexes i.e. yearly rate oil production of old wells and new wells,yearly oil increase of old well measures and geological production rate as output index-es.DEA models are applied in evaluating the production efficiencies of nine oilfield development units in one oilfield,and their technological efficiencies,pure technological efficiencies,scale efficiencies,DEA projections,input surpluses and output deficits are known.The case shows that CCR model and BCC model can realize the evaluation of effectiveness of input-output configuration in the aspect of technology or scale for the oilfield development units, and help them to get the improvement suggestions on technology or scale.DEA projection model can provide the improvement directions and improvement efforts for the non DEA efficient oilfield de-velopment units to realize optimal input-output configuration.DEA super efficiency model can realize the sort of different development units in terms of their input-output configuration efficiencies.