JOURNAL OF OPTIMIZATION IN INDUSTRIAL ENGINEERING (JOURNAL OF INDUSTRIAL ENGINEERING)
Year:2013 | Volume:6 | Issue:13|Papers Count:7

This paper presents a new mathematical model for a hybrid flow shop scheduling problem with multiprocessor tasks in which sequence dependent set up times and preemption are considered. The objective is to minimize the weighted sum of makespan and maximum tardiness. Three meta-heuristic methods based on genetic algorithm (GA), imperialist competitive algorithm (ICA) and a hybrid approach of GA and ICA are proposed to solve the generated problems. The performances of algorithms are evaluated by computational time and Relative Percentage Deviation (RPD) factors. The results indicate that ICA solves the problems faster than other algorithms and the hybrid algorithm produced best solution based on RPD.

Many real-world production systems produce large-sized commodities. Due to the size of the production unit, it is typical to see that more than one worker is working on the same work-piece. This type of assembly line in which multiple workers operate on the same work-piece simultaneously is called multi-manned assembly line (MAL). In the classical multi-manned assembly line balancing problem (MALBP) the objective is to minimize the manpower needed to manufacture one product unit. Apart from the manpower, other cost drivers like wage rates or machinery are neglected in this classical view of the problem.However due to the high competition in the current production environment, reducing the production costs and increasing utilization of available resources are very important issues for manufacturing managers. In this paper a cost-oriented approach is used to model the MALBP with the aim of minimizing total cost per production unit. A mathematical model is developed to solve the problem. Since the proposed model is NP-hard, several heuristic algorithms and a genetic algorithm (GA) are presented to efficiently solve the problem.Parameters and operators of the GA are selected using the design of experiments (DOE) method. Several examples are solved to illustrate the proposed model and the algorithms.

A shortest path problem is a practical issue in networks for real-world situations. This paper addresses the fuzzy shortest path (FSP) problem to obtain the best fuzzy path among fuzzy paths sets. For this purpose, a new efficient algorithm is introduced based on a new definition of ideal fuzzy sets (IFSs) in order to determine the fuzzy shortest path. Moreover, this algorithm is developed for a fuzzy network problem including three criteria, namely time, cost and quality risk. Several numerical examples are provided and experimental results are then compared against the fuzzy minimum algorithm with reference to the multi-labeling algorithm based on the similarity degree in order to demonstrate the suitability of the proposed algorithm. The computational results and statistical analyses indicate that the proposed algorithm performs well compared to the fuzzy minimum algorithm.

Six-Sigma has some measures which measure performance characteristics related to a process. In most of the traditional methods, exact estimation is used to assess these measures and to utilize them in practice. In this paper, to estimate some of these measures, including Defects per Million Opportunities (DPMO), Defects per Opportunity (DPO), Defects per unit (DPU) and Yield, a new algorithm based on Buckley's estimation approach is introduced. The algorithm uses a family of confidence intervals to estimate the mentioned measures. The final results of introduced algorithm for different measures are triangular shaped fuzzy numbers. Finally, since DPMO, as one of the most useful measures in Six-Sigma, should be consistent with costumer need, this paper introduces a new fuzzy method to check this consistency. The method compares estimated DPMO with fuzzy customer need. Numerical examples are given to show the performance of the method. All rights reserved.

Injection molding is one of the most important and common plastic formation methods. Combination of modeling tools and optimization algorithms can be used in order to determine optimum process conditions for the injection molding of a special part. Because of the complication of the injection molding process and multiplicity of parameters and their interactive effects on one another, analytical modeling of the process is either impossible or difficult. Therefore Artificial Neural Network (ANN) is used for modeling the process.Process conditions data is needed for modeling the process by the neural network. After modeling step, the model is combined with the Genetic Algorithm (GA). Based on the injection molding goals that have been turned into fitness function, the optimized conditions are obtained.

Lot-sizing problems (LSPs) belong to the class of production planning problems in which the availability quantities of the production plan are always considered as a decision variable. This paper aims at developing a new mathematical model for the multi-level capacitated LSP with setup times, safety stock deficit, shortage, and different production manners. Since the proposed linear mixed integer programming model is NP-hard, a new version of simulated annealing algorithm (SA) is developed to solve the model named revised SA algorithm (RSA). Since the performance of the meta-heuristics severely depends on their parameters, Taguchi approach is applied to tune the parameters of both SA and RSA. In order to justify the proposed mathematical model, we utilize an exact approach to compare the results.To demonstrate the efficiency of the proposed RSA, first, some test problems are generated, then, the results are statistically and graphically compared with the traditional SA algorithm.

This paper investigates the problem of selecting and scheduling a set of projects among available projects. Each project consists of several tasks and to perform each one some resource is required. The objective is to maximize total benefit. The paper constructs a mathematical formulation in form of mixed integer linear programming model. Three effective metaheuristics in form of the imperialist competitive algorithm, simulated annealing and genetic algorithm are developed to solve such a hard problem. The proposed algorithms employ advanced operators. The performance of the proposed algorithms is numerically evaluated. The results show the high performance of the imperialist competitive algorithm outperforms the other algorithms.