为解决两阶段生产过程中存在的非期望最终产出问题,将方向性距离函数(DDF)拓展到传统两阶段DEA模型中,构建了两阶段投入产出导向的Malmquist-Luenberger指数模型。结果表明,系统整体生产率变动是各阶段综合作用的结果。与“黑箱”Malmquist-Luenberger指数相比,新建模型能更合理、更贴近实际地评价生产率变化。通过各阶段技术进步指数和技术效率指数的变化分析,可帮助决策者深入探究整体生产率变化的原因,明确改进提升的方向。算例研究证实了该方法的合理优越性。

According to the undesirable outputs generated in the two-stage production process, we extend the directional distance function to the model of traditional two-stage DEA, and propose the two-stage DDF-DEA model. On the base of the new model, the paper constructs two-stage input and output-oriented Malmquist-Luenberger productivity index. The results illustrates that the changes of DMU’s overall total factor productivity (TFP) is the comprehensive effect results of every stage’s TFP changes. Compared with the single process of Malmquist-Luenberger index, the new index model can evaluate period efficiency changes more reasonably, and more close to the actual. By analyzing the changes of each stage’s technological progress index and technical efficiency index, it can help decision makers to further explore the changes reason of total factor productivity, so as to clear the direction of improvement and promotion. The numerical example approves the rationality and superiority of our method.

N945

国家自然科学基金项目（面上项目，重点项目，重大项目）