Journal of Optimization in Industrial Engineering
Volume:4 Issue: 9, Summer and Autumn 2011

The multivariate exponentially weighted moving average (MEWMA) control chart is one of the best statistical control chart that are usually used to detect simultaneous small deviations on the mean of more than one cross-correlated quality characteristics. The economic design of MEWMA control charts involves solving a combinatorial optimization model that is composed of a nonlinear cost function and traditional linear constraints. The cost function in this model is a complex nonlinear function that formulates the cost of implementing the MEWMA chart economically. An economically designed MEWMA chart to possess desired statistical properties requires some additional statistical constraints to be an economic-statistical model. In this paper, the efficiency of some major evolutionary algorithms that are employed in economic and economic-stati stical design of a MEWMA control chart are discussed comparatively and the results are presented. Theinvestigated evolutionary algorithms are simulated annealing (SA), differential evolution (DE), genetic algorithm (GA), and particle swarm optimization (PSO), which are the most well known algorithms to solve complex combinatorial optimization problems. The major metrics to evaluate the algorithms are (i) the quality of the best solution obtained, (ii) the trends of responses in approaching the optimum value, (iii) the average objective-function-value in all trials, and (iv) the computer processing time to achieve the optimum value. The result of the investigation for the economic design shows that while GA is the most powerful algorithm, PSO is the second to the best, and then DE and SA come to the picture. For economic-statistical design, while PSO is the best and GA is the second to the best, DE and SA have similar performances.

The classical method of process capability analysis necessarily assumes that collected data are independent; nonetheless, some processes such as biological and chemical processes are autocorrelated and violate the independency assumption. Many processes exhibit a certain degree of correlation and can be treated by autoregressive models among which the autoregressive model of order one (AR (1)) is the most commonly used one. In this paper, we discuss the effect of autocorrelation on the process capability analysis when a set of observations are produced by an autoregressive model of order one. We employ the multivariate regression model to modify the process capability estimated from the classical method where AR (1) parameters are utilized as regression explanatory variables. Finally, the performance of the method developed in this paper is investigated using a Monte Carlo simulation.

In this study, we focused on Tehran stock exchange market analysis based on applying moving average rules. The Tehran stock exchange in the Middle East has evolved into an exciting and growing marketplace where individual and institutional investor trade securities of over 420 companies. In an attempt to examine the ability to earn excess return by exploiting moving average rules, the average annual return on exponential moving average and simple moving average strategies were compared with annual return generated by naive buy and hold strategy. The finding based on the paired t-confidence interval hypothetical test procedures indicates that the moving average rule has more capability in predicting Tehran market and employment of the proposed technique generates excess returns for investors. Based on the findings, it is concluded that the Tehran capital market has great opportunities to apply such technique for yield enhancement and portfolio diversification.

Assigning facilities to locations is one of the important problems, which significantly is influence in transportation cost reduction. In this study, we solve quadratic assignment problem (QAP), using a meta-heuristic algorithm with deterministic tasks and equality in facilities and location number. It should be noted that any facility must be assign to only one location. In this paper, first of all, we have been described exact methods and heuristics, which are able to solve QAP; then we have been applied a meta-heuristic algorithm for it. QAP is a difficult problem and is in NP-hard class, so we have been used honey bee mating optimization (HBMO) algorithm to solve it.This method is new and have been applied and improved NP-hard problems. It’s a hybrid algorithm from Honey-Bee Mating system, simulated annealing and genetic algorithm.

In Today’s quality- based competitive world, known as knowledge age, customer attraction is of ultimate importance. In respect to the slogan “customer is always right”, customer relation management is the core of an organizational strategy playing an important role in four aspects of customer identification, customer attraction, customer retaining, and customer satisfaction. Commercial organizations have perceived increased value of customers through analysis of customers’ life cycle. Data storing and data mining tools along with other customer relation management methods have provided new opportunities for the business. This paper tries to help organizations determine the criteria needed for the identification of potential customers in the competitive environment of their business by employing data mining in practice. It also provides a mechanism for the identification of potential customers liable to becoming real customers. Using Decision Tree tool, the main criteria are identified and their importance are determine din this paper and then assuming that each main criterion consists of several sub-criteria, their importance in turning potential customers into real ones is in turn determined. By utilizing the identified criteria and sub- criteria, organizations are able to drive selling processes in each attendance in a direction which results in attendants’ (future customers) purchase.

Multi attribute decision making methods are considered as one of the most useful methods for solving ranking problems. In some decision making problems, while the alternatives for corresponding criteria are compared in a pairwise comparison manner, if the criteria are inherently fuzzy, debates will arise in ranking alternatives due to the closeness of the values of the criteria. In this research, the fuzzy PROMETHEE II is proposed as a solution in such conditions. First, using the ANP method, the criteria are weighted. Then, the ranking process is accomplished both by the fuzzy PROMETHEE II and the fuzzy TOPSIS methods. Finally, calculating Spearman correlation coefficient, the results of these two approaches are compared. Then, the most important risks are selected via the fuzzy binary goal programming and ranked again through the fuzzy PROMETHEE II and fuzzy TOPSIS methods finally, in the last step, these ranking two are compared. As a case study, highway construction risks are ranked through this method.

This paper develops a single-vendor single-retailer supply chain for multi-product. The proposed model is based on Vendor Managed Inventory (VMI) approach and vendor uses the retailer''s data for better decision making. Number of orders and available capital are the constraints of the model. In this system, shortages are backordered; therefore, the vendor’s warehouse capacity is another limitation of the problem. After the model formulation, an Integer Nonlinear Programming problem will be provided; hence, a genetic algorithm has been used to solve the model. Consequently, order quantities, number of shipments received by a retailer and maximum backorder levels for products have been determined with regard to cost consideration. Finally, a numerical example is presented to describe the sufficiency of the proposed strategy with respect to parameter-tuned by response surface methodology (RSM).

Recently, the economic crisis has resulted in instability in stock exchange market and this has caused high volatilities in stock value of exchanged firms. Under these conditions, considering uncertainty for a favorite investment is more serious than before. Multi-objective Portfolio selection (Return, Liquidity, Risk and Initial cost of Investment objectives) using MINMAX fuzzy goal programming for a Fuzzy Allocated Portfolio is considered in this research and all the main sectors of investment are assumed under uncertainty. A numerical example on stock exchange is presented to demonstrate the validity and strengths of the proposed approach.