International Journal of Applied Operational Research

Three levels of sustainable production inventory model for defective and deteriorating items with rework under marketplace selling price

T Sekar

The intention of the article is to present an imperfect production process inventory model for deteriorating items where the demand rate depends on advertising cost and price. In this research, an industrial manager produces products in a determined imperfect replenishment cycles and also in each replenishment cycle, the manufacturer produces items in three imperfect production cycles and one rework setup, (M/P3/R1) inventory system. Rework is one of the important issues in reverse logistics and green supply chain, because it reduces waste, environmental harms and also the overall inventory cost significantly. Here shortages are considered under completely backlogging strategy. The major objective is to establish the optimum advertising cost, optimum cycle time, optimum lot size quantity and optimum selling price with the intend of minimizing the total inventory cost. A numerical example and a sensitivity analysis are established to validate the theoretical results, and managerial insights for industry managers are provided.

Multi-mode resource constrained time-cost trade off problem by including tardiness penalty cost of renewable resources and fuzzy duration

R. Sadeghian

Nowadays, the project scheduling problem is one of the most common issues in project control and planning, especially the time-cost trade-off problem. In this paper, a bi-objective problem is presented to minimize the cost and makespan simultaneously. Besides the common constraints in literature, it is assumed some renewable resources are hired. Each of them has a specific access time and deadline; they cannot be used before the access time but can be used after the deadline due to the cost of delay penalty. The project costs consist of direct costs, indirect costs, and tardiness penalty cost of renewable resources. Because of the uncertainty, the project times are considered as fuzzy numbers. Due to the NP-Hard nature of these problems, the Tabu search algorithm is used to solve them. The results are also compared with the genetic algorithm to check the quality of answers.

Measuring attractiveness by a categorical based evaluation technique (macbeth) for churn and retention of decision of subscribers in the Nigeria telecommunication industry

S. O. Adebiyi

This study investigates the use of Multi-Criteria Decision Making (MCDM) method known as Measuring Attractiveness by a Categorical Based Evaluation Technique (MACBETH) for evaluating the motivations for churn and retention decisions among subscribers in the Nigeria telecommunication industry. The study adopts quantitative and analytical methods with the aid of a questionnaire administered to one hundred and twenty-three subscribers of mobile telecommunication in the University of Lagos, Nigeria. Based on this sample data the MACBETH model was built with the aid of M-MACBETH to assess the determinants of customers’ churn and retention decision among subscribers and to rank the telecommunication service providers based on service attributes. The results reveal that subscribers considered signal quality /strength as the most influencer criterion for churn and retention decision with higher ranks favouring retention and lower ranks in favours customer churn decision in the telecommunication industry. Thus, telecommunication firms should invest more in technology to boost and sustain high quality signal of their service in order to be thriven in the competitive market.

Common set of weights: a double frontier DEA approach

A. Pourhabib Yekta

Data Envelopment Analysis (DEA) is a non-parametric method for efficiency measurement. In the most common DEA models the method selects the most favorable weight set for all units in order to maximize their efficiency scores. The so called optimistic assessment determines the best efficiency score. To make the performance of DMUs more actionable, the evaluation can be addressed from pessimistic perspective. Under the optimistic and pessimistic points of view, the performance of a unit is assessed with two different evaluation methods. As a result, a different set of weights is achieved for each unit. Hence, to have a more realistic results and better discrimination among DMUs, a more applicable method of a common set of weights (CSW) is suggested. The contribution of the paper is three folded. (1) The proposed approach develops the weight restriction approach, taking into account both optimistic and pessimistic points of view, simultaneously. (2) The proposed weight restriction method considering double frontier generates a positive and a dissimilar set of weights. (3) With the achieved common set of weights the efficiency scores are calculated then the units are ranked. To highlight the details of the proposed method, a real world data application consists of real case study confirm that the presented procedure results in a more realistic and the comprehensive assessment. It also shows the superiority of the proposed method considering double frontier.

Efficiency analysis in multi-period system using DEA-R models

B. Daneshian

In measuring the efficiency of a set of units in a time span that covers several periods, the models based on the standard DEA consider the system as a black box and ignore the status of each unit in each period, which causes misleading results. On the other hand, Wei et al. [14] showed that standard DEA models not only underestimate the efficiency score of inefficient DMU, but also identify efficient DMU as inefficient. In order to solve the above deficiencies, this paper develops DEA-R models by applying MOLP techniques in the presence of multi-period data in such a way that the proposed method can evaluate the overall efficiency according to the periodic efficiency of all units. The proposed method is a general method for p-periodic system. To clarify the details of the proposed method, a comparison between the existing models and the proposed multi-period DEA-R model has been made to measure the efficiency of 22 Taiwanese commercial banks in the period of 2009-2011.

Airline passenger’s sentiment analysis for improving the quality of airline services by using a deep learning approach

A. Nourbakhsh

Advances in technology have increased the availability and use of smartphones. Customer experience is one of the major concerns in the aviation industry. Twitter is one of the most popular social media platforms where travelers can share their feedback. Tweets' Classification based on user sentiments, is an important and common task which has addressed in many researches. Data mining, text mining, web mining, classification for analysis, and illustrating Twitter comments are some of the activities carried out in this field. Text mining is one of the prominent fields of data mining that able to extract useful information from travelers' tweets. This study presents a machine learning-based method for tweets analyzing to improve customer experience handling. The deep learning algorithm identifies ambiguous tweets and decides based on the level of ambiguity. The proposed method provides the feature vector for classification by extracting the word vector from the text analysis, constructing the added Message polarity feature with the WordNet dictionary from the trained examples. The results obtained from the deep learning algorithm validation show that the proposed method is able to identify passenger sentiments in two-class analysis with 99.97% accuracy and in a three-class analysis with 88.83% accuracy.