International Journal of Research in Industrial Engineering
Volume:2 Issue: 4, Autumn 2013

Customer churn has become a critical problem for all companies in particular for those that are operating in service-based industries such as telecommunication industry. Data mining techniques have been used for constructing churn prediction models. Past research in churn prediction context have mainly focused on the accuracy aspect of the constructed churn models. However, in addition to the accuracy, comprehensibility aspect should be considered in evaluating a churn prediction model. Being comprehensible, a model can reveal the main reasons for customer churn; thereby mangers can use such information for effective decisions making about marketing actions. In this paper, we demonstrate the application of a genetic-algorithm (GA) method for building accurate and comprehensible churn prediction model. The proposed method, GA-based method uses a wrapper based feature selection approach for choosing the best feature subset. The key advantage of this method, is taking into account the comprehensibility measure (measured as the number of rules extracted from C4.5 decision tree) in evaluating the performance of a candidate model. The GA-based method is compared to the two filter feature selection methods including Chi-squared based and Correlation based feature selection using two telecommunication churn datasets. The results of experiments indicated that the GA-based method performs better than the two filter methods in terms of both accuracy and comprehensibility

Data envelopment analysis (DEA) is a non-parametric analytical methodology widely used in efficiency measurement of decision making units (DMUs). Conventionally, after identifying the efficient frontier, each DMU is compared to this frontier and classified as efficient or inefficient. This thesis introduces the most productive scale size (MPSS), and anti- most productive scale size (AMPSS), and proposes several models to calculate various distances between DMUs and both frontiers. Specifically, the distances considered in this paper include: (1) both the distance to MPSS and the distance to AMPSS, where the former reveals a unit’s potential opportunity to become a best performer while the latter reveals its potential risk to become a worst performer, and (2) both the closest distance and the farthest distance to frontiers, which may proved different valuable benchmarking information for units. Subsequently, based on these distances, eight efficiency indices are introduced to rank DMUs. Due to different distances adopted in these indices, the efficiency of units can be evaluated from diverse perspectives with different indices employed. In addition, all units can be fully ranked by these indices.

In recent years, contractors play a major role in construction projects of buildings, roads, or waterworks under supervision of project owners or employers of these projects. So, that is why contractor selection is an important decision for employers. Contractor selection is a multicriterion decision problem and resolving the problem of evaluation and although ranking the candidate contractors has become a key factor for firms and enterprises. In this article, the proposed a new hybrid AHP and VIKOR methodology is applied to select the best contractor. Because of dealing with uncertain values for the criteria, fuzzy logic is used to solve this problem and the evaluation data for the alternatives have been expressed in linguistic terms. So, both AHP and VIKOR methods are performed under fuzzy environment. The fuzzy AHP is applied to form the structure of the contractor selection problem and also to obtain weights of the evaluation criteria, and fuzzy VIKOR method is used to determine final ranking. A numerical example is proposed to illustrate an application of the proposed method that demonstrates the effectiveness of the proposed model.

Index tracking is one of the most important passive strategies which describes the process of attempting to track the performance of some specified benchmark indexes. Most recent studies determined security returns in conventional models by the precise historical data. However, such precise data are not always available and it is hard to forecast security returns with stochastic values. Therefore, to handle such imprecise uncertainty, considering security returns as variables with imprecise distributions, i.e., fuzzy variables are recommended. In these studies, researchers have studied and experimented with various risk-measure methods for index tracking portfolio selection. Models which were extended based on Markowitz portfolio selection model have used the single period variance of returns as a risk measure. Since forecasting future returns of portfolio is uncertain, we consider these returns as fuzzy variables in this study. We also apply Value-at-Risk as the risk measure whichhas not yet been established as risk measure in index tracking portfolio selection problems. The model is tested, using Tehran Price Index (TEPIX) and computational results are presented at the end.

This paper aims to study the impact of product greenness and part reliability on reverse supply chain network through green supplier selection and disassembly of products. It proposes a bi-objective mathematical modeling for a closed-loop supply chain network considering green supplier selection and disassembly of products to trade-off profit and greenness. To our knowledge, this study is the first paper which considers the greenness of products in part/components level. So, some important issues such as part reliability, part greenness and inventory management of new and recovered parts are included into the model. Part reliability and part greenness are considered as green criteria in green supplier selection stage. Product greenness is defined according to design for disassembly level. Better design for disassembly means more greenness level for the products and better yield of parts at the disassembly stage. Green parts are made of highly recyclable materials. According to the part greenness and part reliability, some scenarios are defined. The greenness level of product is chosen by the model. ε-constraint method is applied to solve the model. A set of Pareto-optimal solution is obtained by using ε-constraint method to show the trade-off between the profit and greenness objectives. The results showed the efficiency of the model.