شماره ركورد كنفرانس :
4403
عنوان مقاله :
A Recurrent Neural Network Model for solving BCC Model in Data Envelopment Analysis
پديدآورندگان :
Ghomashi A a_gh_l@yahoo.com Department of Mathematics, Kermanshah Branch, Islamic Azad University, Kermanshah, Iran , Abbasi M Department of Mathematics, Kermanshah Branch, Islamic Azad University, Kermanshah, Iran , Shahghobadi S Department of Mathematics, Kermanshah Branch, Islamic Azad University, Kermanshah, Iran
كليدواژه :
Recurrent neural network , Gradient method , Data envelopment analysis , Efficient DMU , Stability , Global convergence
عنوان كنفرانس :
نهمين كنفرانس ملي تحليل پوششي داده ها - توسعه ملي
چكيده فارسي :
In this paper we present a recurrent neural network model for solving BCC Model in Data Envelopment Analysis(DEA).
The proposed neural network model is derived from an unconstrained minimization problem. In theoretical aspect, it is shown that the proposed neural network is stable in the sense of lyapunov and globally convergent. The proposed model has a single-layer structure. Simulation shows that the proposed model is effective to identify efficient DMUs in DEA..
Keywords: Recurrent neural network, Gradient method, Data envelopment analysis, Efficient
