Scale, Indivisibilities and Production Function in Data Envelopment Analysis
Scale, Indivisibilities and Production Function in Data Envelopment Analysis
复制标题
DOI：
10.1016/s0925-5273(02)00412-7
复制
发表时间：
2003-05
期刊：
影响因子：
--
通讯作者：
K. Tone;Biresh K. Sahoo
中科院分区：
文献类型：
--
作者：
K. Tone;Biresh K. Sahoo
本文批判性地重新审视了自亚当·斯密著作以来规模收益与规模经济的概念，将前者与生产单位的概念联系起来，后者与企业的概念联系起来。尽管迄今为止，在探索 DEA 模型背后的经济理论方面已经取得了一些有价值的进展，但这些进展仍处于起步阶段，尚未得到足够的应用。本文试图从经济学的角度来考察DEA效率测算模型及其应用。我们在这里表明，所有多阶段生产过程中不可分割性的存在使得技术结构非凸，因此，标准凸 DEA 生产模型（例如 CCR 和 BCC）由于不可分割性而无法表现出递增收益。然而，FDH 模型中嵌入的非凸技术有助于揭示特定任务过程中产生的过程不可分性，而生产函数的同质表征却无法做到这一点。
This paper critically re-examines the concept of returns to scale vis-à-vis economies of scale since the writings of Adam Smith by relating the former to the concept of production unit and the latter to the concept of firm. Though to date some valuable progress has been made exploring the economic theory underlying DEA models, these developments have remained in their infancy and have not found enough application. An attempt is made here to examine the DEA model of efficiency measurement and its application from an economic viewpoint. We show here that the presence of indivisibilities in all multi-stage production processes makes the technology structure non-convex, and therefore, the standard convex DEA production models (e.g., CCR and BCC) fail to exhibit increasing returns due to indivisibilities. However, the non-convex technology embedded in FDH model helps revealing process indivisibilities arising from task-specific processes whereas a homogeneous characterization of production function fails to do so.