Simulation-Based Evolution Algorithm for Automated Material Handling System in a Semiconductor Fabrication Plant

To perform a high fidelity simulation study on a 300 mm wafer fabrication plant, modeling of the manufacturing process (MP) alone is not sufficient. Inclusion of the automated material handling system (AMHS) model is necessary due to the high degree of factory automation. There isn't, however, a single tool that is capable of modeling both the AMHS and MP with sufficient accuracy and granularity. A commercial simulation package such as AutoMod is usually used to model the AMHS while AutoSched AP is usually used to model the MP. These packages can be integrated using the supplied interoperation module but flexibility in optimizing the execution performance for different simulation models is lacking. In this paper, we present an approach to interoperation based on the High Level Architecture standard. We note that the typical characteristics of disparity in the models' time granularity and frequent model interactions are the obstacle to good execution performance.

In this paper, a simulation-based optimization approach, named NHOCBA, for a typical resource allocation problem is presented. The hybrid algorithm based on neighborhood algorithm is applied to explore toward optimal direction in design space. For increasing efficiency, an optimal computing budget allocation (OCBA) is adopted to compute the optimal number of replications and to provide reliable evaluation of variance. In addition, we deal with the resource allocation problem which exist multiple global optima in design space. Therefore, a trim procedure which prevents to allocate extra numbers of replication to local optima has been proposed to enhance efficiency. Then, we use confidence interval at the end of algorithm procedure to find an optimal set instead of an optimal solution from design space. Finally, we compare the NHOCBA with different algorithms by experimentation study which shows that NHOCBA approach can perform better than the other algorithms under certain conditions.

Evolutionary algorithm assisted by surrogate model in the framework of ordinal optimization and optimal computing budget allocation

A new concept of the dynamic connecting transport automated material handling system (AMHS) is proposed. This paper analyses the performance of the number of vehicles in a dynamic connecting transport AMHS in a simplified 300 mm wafer fab. Discrete-event simulation models are developed in e-M Plant™ to study the system performance. To avoid congestion or idle time in the intrabay system, the control of the upper limit or the lower limit on the number of vehicles can be the feasible solution. Thus, four strategies are identified and their performances are also investigated. In addition, several different operation environments such as the flow rate and the product-mix are used to investigate the performance of the dynamic connecting transport AMHS.

Embedding evolutionary strategy in ordinal optimization for hard optimization problems

Simulation models of emergency departments (EDs) are often built based on incomplete data, for example, missing arrival times or service-time durations. The difficulty in collecting reliable and complete data can subsequently lead to invalid simulation results. To tackle this problem, we propose a simulation and optimisation method to characterise the unavailable durations of service times. Since many services in an ED are sequential and dependent on each other, this paper considers these multiple process steps cooperatively. We first use lognormal distributions to characterise the key service durations. Then we propose a new meta-heuristic approach, which combines an Improved Adaptive Genetic Algorithm (AGA) and Simulated Annealing (SA), IAGASA, to search for the optimal set of service-time distribution parameters. To address the difficulties of applying IAGASA when noise is involved in the performance measures and improve the simulation efficiency, we jointly apply IAGASA and Optimal Computing Budget Allocation (OCBA) technology. OCBA minimises the total simulation cost for achieving a desired level of probability of correctly selecting the best set of distribution parameters, which improves the search efficiency significantly. The experimental results indicate that our proposed method can find accurate estimates of service-time distribution parameters within a relatively short time.

Choosing suitable buffer setups for network-flow production lines of automated manufacturing systems to augment throughputs is a pragmatic issue. In this work, an approach merging artificial immune system (AIS) and ordinal optimization (OO) is developed to determine an optimal buffer resource allocation of a network-flow production line for maximizing the throughput. The proposed approach consists of two levels. The first level is using AIS assisted by a rough estimation to select a candidate subset of solutions. The second level is to identify an excellent solution among the solutions obtained from the first level using optimal computing budget allocation (OCBA) technique. We have tested the proposed approach on a three-stage ten-node network-flow production line and compared the test results with those obtained by three swarm intelligence methods with accurate estimation. Test results demonstrate that the proposed approach can obtain an excellent solution within a reasonable computing time and outperforms the three swarm intelligence methods.

This paper presents a simulation optimization with genetic algorithm and optimal computing budget allocation for semiconductor back-end assembly scheduling problem to achieve minimal average order flow time. In particular, this research explores characteristics of hybrid flow shop scheduling problem in complicated identical and unrelated parallel machines, orders with specific product and demand scheduled with different release time, order split for parallel and merges for batch processing during manufacturing under product-machine dedication with stochastic processing times and sequence-dependent setup times. As this is a real-life stochastic event system and discrete simulation is usually the only resort for performance. Coupling genetic algorithms in simulation optimization as the solution space is large. Optimal computing budget allocation was used to reduce simulation budget usage while stochastic situation. Numerical result shows it is effective to improve performance better than practical heuristics and efficient to generate few replications which conquer the barriers utilize the advantages of simulation-based approach. This work can be as solution architecture for providing superior scheduling decision on allocation of orders and subsequent jobs to machines in a complex hybrid flow shop.

This study explores an economic production quantity (EPQ) model designed with the assumptions of discrete delivery orders and storage capacity constraints for a multi-item production inventory system. The main purpose of this study is to determine the optimal order quantity, the optimal number of deliveries and the optimal delivery quantity. First, the developed model as part of this study was analyzed using Genetic Algorithm (GA). Numerical analysis results were compared with those of previous studies and it was found that it is possible to have better results with an increasing number of iteration. The same model was then analyzed using Particle Swarm Optimization (PSO) algorithm. A comparison of the optimization methods showed that PSO gives better results over the GA under the same number of iterations and using the same population. The effects of important model parameters such as number of iterations, population, crossover, mutation rate on the optimal solution were analyzed. The results showed that PSO performs better than the GA with respect to the total cost and the total runtime as the solution of the problem in question.

Genetic algorithm (GA) has been extensively used in recent decades to solve many optimization problems in various fields of science and engineering. In most cases, the number of iterations is the only criterion which is used to stop the GA. In practice, this criterion will lead to prolong execution times to ensure proper solution. A novel approach is presented in this article as the approximate number of decisive iterations (ANDI ) which can be used to successfully terminate the GA optimization method with minimum execution time. Two simple correlations are presented which relate the new parameter (ANDI ) with approximate degrees of freedom (Adf ) of the merit function at hand. For complex merit functions, a linear smoother (such as Regularization network) can be used to estimate the required Adf. Four illustrative case studies are used to successfully validate the proposed approach by effectively finding the optimum point by using to the presented correlation. The linear correlation is more preferable because it is much simpler to use and the horizontal axis represents the approximate (not exact) degrees of freedom. It was also clearly shown that the Regularization Networks can successfully filter out the noise and mimic the true hyper-surface underlying a bunch of noisy data set.

The crowd-creation space is a manifestation of the development of innovation theory to a certain stage. With the creation of the crowd-creation space, the problem of optimizing the resource allocation of the crowd-creation space has become a research hotspot. The emergence of cloud computing provides a new idea for solving the problem of resource allocation. Common cloud computing resource allocation algorithms include genetic algorithms, simulated annealing algorithms, and ant colony algorithms. These algorithms have their obvious shortcomings, which are not conducive to solving the problem of optimal resource allocation for crowd-creation space computing. Based on this, this paper proposes an In the cloud computing environment, the algorithm for optimizing resource allocation for crowd-creation space computing adopts a combination of genetic algorithm and ant colony algorithm and optimizes it by citing some mechanisms of simulated annealing algorithm. The algorithm in this paper is an improved genetic ant colony algorithm (HGAACO). In this paper, the feasibility of the algorithm is verified through experiments. The experimental results show that with 20 tasks, the ant colony algorithm task allocation time is 93ms, the genetic ant colony algorithm time is 90ms, and the improved algorithm task allocation time proposed in this paper is 74ms, obviously superior. The algorithm proposed in this paper has a certain reference value for solving the creative space computing optimization resource allocation.

Lot output time prediction is a critical task to a wafer fab (fabrication plant). Traditional studies are focused on prediction accuracy and efficiency. Another performance measure that is as important but has been ignored in traditional studies is the achievability of an output time forecast, which is defined as the possibility that the fabrication on a wafer lot can be finished in time before the output time forecast. Theoretically, if a probability distribution can be obtained for the output time forecast, then the achievability can be evaluated with the cumulative probability of the probability distribution before the given date. However, there are many managerial actions that are more influential to the achievability. For this reason, a fuzzy set approach is proposed for evaluating the achievability of the output time forecast. The fuzzy set approach is composed of two parts: a fuzzy back propagation network (FBPN) and a set of fuzzy inference rules (FIRs). An example is used to demonstrate the applicability of the proposed methodology.

In this study, we consider ranking and selection problems where the simulation model is subject to input uncertainty. Under the input uncertainty, we compare system designs based on their worst-case performance, and seek to maximize the probability of selecting the design with the best performance under the worst-case scenario. By approximating the probability of correct selection (PCS), we develop an asymptotically (as the simulation budget goes to infinity) optimal solution of the resulting problem. An efficient selection procedure is designed within the optimal computing budget allocation (OCBA) framework. Numerical tests show the high efficiency of the proposed method.

In this paper the image compression problem is analyzed using genetic clustering algorithms based on the pixels of the image. The main problem to solve is to find an algorithm that performs the clustering efficiently. The possibility of solving clustering problems with genetic algorithms provides optimal solution. The present work makes use of genetic clustering algorithms to obtain an ordered representation of the image and then applies the DWT (Discrete Wavelet Transformation) to compress the image. The image quality is measured using PSNR measure. The genetic algorithm (GA)-based compression image has more PSNR value compared to the image without GA. It shows that the quality of the image is good for GA based compression. Also it reduces the problem of memory utilization efficiently due to compression and at the same time quality of the image is maintained due to GA. Keywords: Genetic algorithms, clustering, image compression, discrete wavelet transformation.

This paper presents a fuzzy-neural-network-based fluctuation smoothing rule to further improve the performance of scheduling jobs with various priorities in a wafer fabrication plant. The fuzzy system is modified from the well-known fluctuation smoothing policy for a mean cycle time (FSMCT) rule with three innovative treatments. First, the remaining cycle time of a job is estimated by applying an existing fuzzy-neural-network-based approach to improve the estimation accuracy. Second, the components of the FSMCT rule are normalized to balance their importance. Finally, the division operator is applied instead of the traditional subtraction operator in order to magnify the difference in the slack and to enhance the responsiveness of the FSMCT rule. To evaluate the effectiveness of the proposed methodology, production simulation is applied to generate some test data. According to the experimental results, the proposed methodology outperforms six existing approaches in the reduction of the average cycle times. In addition, the new rule is shown to be a Pareto optimal solution for scheduling jobs in a semiconductor manufacturing plant.