Journal of Advances in Industrial Engineering
Volume:48 Issue: 1, 2014

Optimization through simulation is considered an efficient tool in dealing with optimization problems. It’s a tool that is capable to cover real world problems to a much more complete way than the other optimization tools. The problem of this tool is the necessity of heavy processing computations. This problem is caused not only since the simulation process is time consuming, but the objective function value at a point of thesolution space is required to simulate multiple performances. Meta-models are approaches to deal with this dilemma and have other problems that limit their use. Meta-model remodel the simulation model; it reduces the complexity of the model and yet the model still further away from the real world In this paper, the particle swarm optimization is modified so that can addition to remove the meta-models, reduce time to achieve optimal response.

In this paper, we study the network location problem for two congestible facilities, considering impatient customers. We suppose that when a customer arrives his/her corresponding facility and faces a queue with the size more than a predetermined threshold, the customer decides not to join the queue but to go to the closest facility to current facility. In the second facility, if the queue size is also more than the threshold, the customer gives up the service. Since the impatience is the case for many real world situations, its consideration helps to construct more realistic mathematical models. The aim of the proposed mathematical model is to select two locations among a prespecified number of candidate locations and then to assign all customers in the network in such a way that the number of the lost customers is minimized. Based on a new kind of hypercube queue models, the queue systems of the servicing facilities is analyzed and modeled as a mathematical programming model. In order to evaluate the performance of the proposed model, a numerical example is presented and solved using GAMS software.

In today's competitive environment, customers are the most important asset to any company. Therefore, in order to retain customers, it is essential to understandtheir behaviour for developing effective strategies. One of the most common methods for customer analysis is market segmentation that helps companies to develop marketing technique by dividing market into several smaller homogeneous groups. Optimizedmarket segmentation based on relevant attributes help allocating company’s resource. In this paper, we proposed a new two-dimensional framework, in which customers was measured according to debt and revenue status(The combination of both values ​​and anti-values), based on RFM model for segmenting customers in fixed-line telecom. Finally, strategies are presented for any customer group based on segmentation using data mining techniques.

This paper proposes a new mathematical model for designing and planning postal parcels carriage and postal networks. The objective of postal network designing is to determine the location, size and connection between the postal stations in a city, province or country. In this paper, a kind of multiple locations-allocations hub problem with limited capacity in collecting, distributing and broadcasting by the hub in a network with the desired interval structure is studied. A multi-commodity network transportation with demand, transportation cost, possible capacity and three kind of hubs is considered in a long-term time horizon. Since the mathematical modeling is a non-linear mixed integer programming problem, the aim is to decide on hub kinds and on the points that should be assigned to the established hubs. Numerical experiments demonstrate the significance and applicability of the proposed model for actual decision-making problems.

In today’s business competitive world, decision makers of companies try to employ standard, efficient, theoretical and operational proven methods as a competitive advantage for making their critical strategic business decisions in order to survive in their industry. In this paper, a hybrid model based on Fuzzy Analytic Hierarchy Process (FAHP) and Artificial Neural Network (ANN) is presented. This model can be used as a Decision Support System (DSS) to capture and represent the decision makers’ preferences, without their direct or interference, in order to represent them the ability to select the best product among candidates. The priority of products is determined by the factors elicited from interviews and surveys. Firstly, the weight of each factor is determined by the means of FAHP technic. Then, the priority of products will be determined by weighing each of them with respect to the factors. But, the output of FAHP is just a possible answer. The topology of the neural network model is developed to train the model and give the decision makers the best possible answer. In fact, neural network is used to learn the relation among criteria and alternatives and rank the alternatives. A method based on AHP is developed and used as the MCDM method, and Multi level perceptron is used as our selected kind of artificial neural network. A case study is discussed and developed through the paper. The comparison of the real investment data of the studying case and the results of the model proves the effectiveness of the proposed model.

Today, due to rising passenger expectations and other resource constraints, the airport managers and staff do to develop appropriate programs in order to optimize the allocation of the resources. Gates and counters are important resources in airport. In this research problem of gate and counter for arrival flights is considered simultaneously and mathematical model is prepared. The objectives of model includes minimizing passenger walking distance, minimizing the number of flights is allocated parking and alternate assign of gates (as a soft constraint). Zero- one Mathematical model with multi objective function (MODM) has been solved with precise method (GAMS software). Assigned similar flights to a terminal and balance in terminals are constraints of the model that there are low studies about it.

The term business intelligence (BI) can refer to various computerized methods and processes of turning data into information and then into knowledge, which is eventually used to enhance organizational Decision making. In the last decade, Business intelligence has evolved as one of the critical applications in organizations to provide useful insight, support decision-making, and drive organizational performance. Grey system theory was developed by Deng based upon the concept that information is sometimes incomplete or unknown. The intent is the same as with factor analysis, cluster analysis, and discriminant analysis, except that those methods often dont work well when sample size is small and sample distribution is unknown. With grey related analysis, interval numbers are standardized through norms, which allow transformation of index values through product operations. This article is a descriptive survey. We have tried to find a suitable solution for selecting business intelligence systems. In the proposed solution Grey theory is used to assigned adequate relative importance to the criteria then an optimum alternative between four brands was selected.

In this research two models for pricing and inventory decisions of complementary products with and without deterioration assumption are developed. In both of them, we assumed that the demand function of each product is dependent on own price and the price of another one. The aim of this paper is to optimize the total profit of integrated system such that the optimal values for the price of each product, order quantity and period length are determined. Using two theorems concavity of both profit functions is proved and in order to show the applicability of the proposed models two numerical examples are provided. Then sensitivity analysis on complementary degree for both models is done.

In this paper, a new hybrid model has been proposed for the loans allocation of banking investment projects. In this model, firstly uncertainty in the cash flows parameters of projects is simulated using probability distributions. The statistical analysis about the indexes of engineering economic is used for the elimination of non-economical projects. Secondly, remained projects are ranked based on financial and non-financial indexes and AHP-TOPSIS approaches. Finally, a 0-1 mathematical programming is used for loans allocation to projects based on ranking of projects (as projects utilities) and the constraints of banking investment. To consider the performance of the proposed model, the investment projects of bank Toseeye Taavon in a specified time interval are evaluated using this model. The results of allocations are shown the validity and efficiency of proposed model based on the opinions of banking experts.

Hub location problems are among the extensions of classical location problems that have numerous applications in designing transportation, postal, and telecommunication networks. Despite the evident attention of researchers to these problems, there are few studies on designing hub networks under uncertainty and disruption. In this study, the capacitated single allocation hub location problem under demand uncertainty and disruption is investigated. First, a mathematical model based on budget of uncertainty concept is proposed, then a hybrid solution method based on genetic algorithm and variable neighborhood search is proposed. Computational experiments demonstrate the accuracy of the proposed solution method in solving the testinstances. In addition, the results obtained from conducting Monte-Carlo simulation experiments to analyze the effects of uncertainty on the performance of different hub network configurations show the necessity of considering demand uncertainty and disruption while designing hub networks.

Risk management is one of the nine project management processes, which provides positive results and minimizes the probability and consequences of adverse effects in the project. In this article a zero-one linear programming model with multiple objectives is used to characterize the potential risks. The proposed model has two main objectives. The first objective is the minimization of the expected cost of different risks and the Second one is to minimize expected time resulting from potential risks. The model has also two limitations, the first ensures that time effect of occurred risks, does not excess the total effect of the time value of expected risk and the second one controls the cost of occurred risk to be less than or equal to the expected cost values. The LP Metric method is proposed to solve the model and obtain the Pareto optimal solutions for a case study in a dam in Asalouyeh.