Measurements of Productivity Differences in Forage Production for a Basis of the Development of Farm Diagnosis Expert System

本研究的目的是开发和说明个体生产力效率的测量和评估方法，该方法将成为农场诊断专家系统的“问题发现”部分和主要组成部分。由于个体农场管理者在制定农场持续增长战略的过程中，必须认识到其农场的比较生产率和差异的原因，因此必须开发生产率评估框架，作为计算机辅助农场诊断系统的重要组成部分。虽然我们无法开发出一个专家系统来寻找单个农场低生产率的原因，但我们成功地获得了测量每个农场比较生产率的方法，并根据统计估计的生产函数确定了导致因素。以饲料生产为例说明了这种方法。根据我们对北海道饲料生产的案例研究，造成这种情况的主要原因之一是养殖规模。所有实现高生产率的农场都出现在规模报酬递增的大型农场中，而大多数农场呈现不变的规模报酬。其次，高度机械化的农场并不总是被认为是高生产率的农场。同样重要的是，诸如肥料之类的可变投入必须彼此适当组合使用。本文的结论是，我们的农场生产率差异测量方法可以用于建立有效的专家系统，旨在发现问题并形成提高个体农场生产率的策略。

This study was designed to develop and liiustrate methods for measuring and evaluating individual productivity efficiency which would be a tool of " problem finding" part and a principal component of farm diagnosis expert system. Since individual farm manager must recognize comparative productivity of his farming and the causes of difference in an effort of setting-up his strategies for continuous growth of the farm, framework for productivity evaluation must be developed as an essential part of compouter aided farm diagnosis system.We could not completed in developing an expert system for finding the causes of low productivity at individual farm level which would be available for practical use though, we successfully obtained methods of measuring comparative productivity of each farm and identified the causing factors based on statistically estimated production functions. Forage production was chosen to illustrate this methods.According to our case studies in forage production in Hokkaide, one of primal causes of this was farming size. All farms realizing high productivity are found among large scale farms which showed increasing return to scale, while most farms showed constant return to scale. Secondly, highly mechanized farms are not alwas identified as high productivity farms. Equally important was that variable inputs such as fertilizer must be used in a proper combination each other.It is our conclusion that our methods of measuring productivity difference among farms are able to incorporate to built effective expert system aiming to find problems and to form strategies for the improvement of productivity at individual farm base.