A Hybrid Method for Prediction and Assessment Efficiency of Decision Making Units: Real Case Study: Iranian Poultry Farms
Iman Rahimi (Department of Mechanical and Manufacturing Engineering, Faculty of Engineering,, University of Putra Malaysia, Serdang, Selangor, Malaysia), Reza Behmanesh (Department of Accounting, Islamic Azad University, Khorasgan Branch, Isfahan, Iran) and Rosnah Mohd. Yusuff (Department of Mechanical and Manufacturing Engineering, Faculty of Engineering, University of Putra Malaysia, Serdang, Selangor, Malaysia)
Volume 5, Issue 1. Copyright © 2013. 18 pages.
OnDemand Article PDF Download
Download link provided immediately after order completion
Instant access upon order completion.
The objective of this article is an evaluation and assessment efficiency of the poultry meat farm as a case study with the new method. As it is clear poultry farm industry is one of the most important sub- sectors in comparison to other ones. The purpose of this study is the prediction and assessment efficiency of poultry farms as decision making units (DMUs). Although, several methods have been proposed for solving this problem, the authors strongly need a methodology to discriminate performance powerfully. Their methodology is comprised of data envelopment analysis and some data mining techniques same as artificial neural network (ANN), decision tree (DT), and cluster analysis (CA). As a case study, data for the analysis were collected from 22 poultry companies in Iran. Moreover, due to a small data set and because of the fact that the authors must use large data set for applying data mining techniques, they employed k-fold cross validation method to validate the authors’ model. After assessing efficiency for each DMU and clustering them, followed by applied model and after presenting decision rules, results in precise and accurate optimizing technique.
2.1. Cluster Analysis (CA)
Clustering is a popular data mining technique, which involves the partitioning of a set of objects into a useful set of mutually exclusive clusters so that the similarity between the observations within each cluster (i.e., subset) is high, whereas the similarity between the observations from the different clusters is low (Samoilenko & Osei-Bryson, 2008, 2010).Unlike decision trees which assign a class to an instance (supervised method), clustering procedures are applied when instances are divided into natural groups or clusters (unsupervised method). There are different ways to produce these clusters. The groups may be exclusive i.e. any instance belongs to only one group probabilistic or fuzzy i.e. an instance belongs to each group to a certain probability or degree (membership value) hierarchical i.e. there is a crude division of instances into groups at the top level and each of these groups are refined further up to individual instances (Thomassey & Fiordaliso, 2006).In other literature, overview of two general approaches to clustering was provided: hierarchical clustering, partitional clustering (e.g., k-means, k-median) (Samoilenko & Osei-Bryson, 2008).
- Examples of application of clustering seen in (Banfield & Raftery, 1992; Ben-Dor, Shamir, & Yakhini, 1999; Dhillon, 2001; Fisher, 1997; Hirschberg & Lye, 2001; Lai, Fan, Huang, & Chang, 2009; Okazaki, 2006; Wallace, Keil, & Rai, 2004).
Complete Article List
Search this Journal:
View Complete Journal Contents Listing
Volume 5: 3 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)