AMIRKABIR
Year:2004 | Volume:15 | Issue:58-D|Papers Count:12

Although cellular manufacturing system provides great benefits, the design of cellular manufacturing systems is complex for real-life problems. Hence, existing design methods employ simplifying assumptions that often deteriorate the validity of the models for obtaining solutions. One of these assumptions is that product mix and demand do not change over the planning horizon. This paper develops a mathematical model and a solution approach for designing cellular manufacturing systems. - This model addresses these shortcomings by assuming dynamic and stochastic production requirements and employing routing flexibility. A simulated annealing based heuristic is proposed to obtain good solutions within reasonable amount of computational time. For evaluating the proposed solution approach, eight design problems are generated at random and solved by using the heuristic and Lingo 8.0 Software as a lower bound. The compared computational results of two approaches (heuristic and lower bound) show that the heuristic perform well under various circumstances, but routing flexibility has a major impact on the performance of the heuristic.

In this paper, we consider a permutation flow shop scheduling problems with n jobs and m machines. The jobs have arbitrary processing times and due dates. Our objective is minimizing the total earliness and tardiness of all jobs. We develop a genetic algorithm to solve this problem. The performance of the algorithm is evaluated through computational experiments.

In this paper we consider the problems related to minimizing number of tardy jobs criteria on a single batch processing machine. A batch processor can process several jobs simultaneously as a batch. We developed a polynomial algorithm to find optimal solution for three special cases: fixed sequence of jobs, identical due dates and identical processing times of jobs. Finally, the results and further research fields are discussed.

In this paper, we present an optimization framework to derive optimal production over a fixed planning horizon for items with stock dependent demand and under inflationary conditions. We consider a perishable item that follows two parameters Weibull distribution. A numerical example is given to illustrate the solution procedure and sensitivity analysis of the optimal solution with respect to the parameters of the system is also carried out.

Almost every method that has been presented to determine residual stress has some limitation and complexities. The aim of this work is to present a new, yet simple method so called" strain indentation" for measuring the residual stresses particularly in thin layers .In this method in addition to the precision measurements, components of residual stress at different directions may be determined. AISI 4140 steel specimens were plasma nitrided at 350°C, 450°C and 550°Cfor 5 hours in the mixture of 75% nitrogen - 25% hydrogen gas. The components of residual stress in the radials axial and hoop directions in the nitrided layer were determined considering the elastic strain recovery after removal of residual stress inducer (i.e. the nitrided layer). Fatigue strength of the nitrided specimens was obtained by plotting the S-N curves and fractographic studies carried out on the fracture surface of the specimens.The effect of residual stress on the stress pattern was simulated. The calculated residual stress components were in the range of 40-210 Mpa and the radial components of residual stress were more than the other two directions. Maximum fatigue strength improvement of up to 110% was observed in the plasma nitrided specimens at 550°C and also 40% improvement in fatigue strength was detected by increasing the nitriding temperature from 350°C to 550 C. This was due to 100% increase in residual stress. Fatigue crack growth velocity in the hoop direction was more than that of radial direction. This seems to be due to higher radial residual stress component compared with the hoop stress component in the sub layer.

Considering the increase of microwave radiation application in different industries, in this research, microwave is used in bleaching of desized and boiled off Cotton fabric. Boiling off and bleaching are important before dyeing and printing. Results of this research showed that microwave can bleach the fabric in 40 seconds. Conventional Processes such as pad-steam and exhaustion need about 30 minutes respectively for acquiring the same degree of whiteness (70%). Microwave did not show negative effects on the strength of the samples. Simultaneous boiling off and bleaching of cotton fabrics in the absence of microwave absorbers does not lead to whitenesses of more than 50%. Urea as microwave absorber increases the degree of whiteness to some extent, but the highest degree obtained is 10% less than that for convertional processes.

This paper reports on the development of a new multiple criteria decision making (MCDM) methodlogy to assess textile products quality using experimental data gathered from standard organization of Iran. A quality attribute is characterised by a fuzzified taguchi quality function and integrated into a hierorchical structure to enable a true assessment of the complex concept of product quality. Real data are used to validata the methodology.

Risk analysis of liquefaction potential was performed for an area in southwest Tehran. SPT blow counts were used besides Iwasaki method to determine the liquefaction potential in the area for different earthquake magnitudes. The seismic hazard curve of Tehran was also evaluated for different earthquake magnitudes, time periods and focal distances. Eliminating the peak horizontal acceleration (PHA), between the two sets of curves, the probability of exceedance was determined for different liquefaction potential values. According to the results, high probability values were determined for liquefaction occurance in some regions. Finally the critical ground water levels to prevent liquefaction occurance was studied in the studied region.  

This paper describes how to predict moment-rotation behaviour, ultimate resistance moment, ultimate rotation and performance of joints using artificial neural networks with back-propagation algorithm. To produce the input data for neural networks, 432 exterior reinforced concrete beam- column joints were non-linearly analyzed. The variables of design in elude the dimensions of beams and columns, longitudinal reinforcement of beams and columns, different arrangement for transverse reinforcement, and different amounts for transverse reinforcement in beams and columns. Two neural network systems have been trained one for predicting the rotation of joints, and the other for ultimate rotation, ultimate resistant moment, and performance of joints. At the end, a method proposed for prediction of moment-rotation curves of reinforced concrete connections.

Determination of Optimum contentsop godarsorkh, sirjan, and godarsefid pozzolans to be replaced by 3 types of cement was carried out by two pozzolanic activity methods. Concrete mixtures with different amounts of pozzolans were made and their compressive strengths were measured. Mixtures showing higher strengths were selected as optimum mixtures for further investigation on their mechanical properties. Durability of concrete mixtures with and without pozzolans was assessed and compared with some other pozzolans. Test results show that concrete mixtures containing pozzolans have better performance than, normal concrete mixtures. Concrete mixtures containing sirjan pozzolan showed the highest, compressive strengths followed by godarsefid and godarsorkh pozzolans.

There are many contributions in connection theory of Finsler geometry, related to their applications in electrical engineering, control theory, relativity, biology, ecology, crystals, Metallurgy, Astronomy and recently Growth of cancer cells and Geology.In this paper, first a scope of applications of finsler geometry in science and engineering. In given then followed by its methoddogy study. In the last part of this work which is an original contribution in connection theory of Finsler connection, we introduce a new Finsler connection which contains some of the well-known connections as a special case.

This paper presents a meta-heuristic algorithm for solving bi-objective fuzzy job shop scheduling problems. These objectives are to minimize the makes pan and minimize the early and late penalty. Processing time and due date are considered as fuzzy triangular numbers. This paper also introduces a novel use of data mining algorithm for solving of combinatorial optimization problems. The proposed algorithm combines genetic algorithms and an attribute-oriented induction algorithm, which is much quicker than previous methods providing the optimal solution. By, considering the structure of proposed algorithm, the whole feasible solutions of a special job shop-scheduling problem are considered as a database. Attribute-oriented induction algorithm must find similar relationships and latent patterns among this database. This algorithm helps genetic algorithms to find the optimal or near-optimal solution much quicker than previous methods. Due to genetic inheritance in solutions obtained by genetic algorithms, the next generation has more similar relationships between their characteristics, operations and sequences. In this paper, three test problems are conducted in order to compare with other solutions reported in the literature and to show the efficiency of the proposed algorithm.