• Journal of Institute of Control, Robotics and Systems
• /
• v.12 no.12
• /
• pp.1169-1172
• /
• 2006

The prices of chemical products are fluctuated by several factors. The chemical companies can't predict and be ready to all of these changes, so they are exposed to the risk of a profit fluctuation. But they can reduce this risk by making a well-diversified product portfolio. This problem can be thought as the optimization of the product portfolio. We assume that the profits come from the 'spread' between a naphtha and a chemical product. We calculate a mean and a variation of each spread and develop an automatic module to calculate the optimal portion of each product. The theory is based on the Markowitz portfolio management. It maximizes the expected return while minimizing the volatility. At last we draw an investment selection curve to compare each alternative and to demonstrate the superiority. And we suggest that an investment selection curve can be a decision-making tool.

• Journal of Korean Institute of Industrial Engineers
• /
• v.39 no.5
• /
• pp.361-366
• /
• 2013

Prediction of process parameters is very important in parameter design. If predictions are fairly accurate, the quality improvement process will be useful to save time and reduce cost. The concept of dual response approach based on response surface methodology has widely been investigated. Dual response approach may take advantages of optimization modeling for finding optimum setting of input factor by separately modeling mean and variance responses. This study proposes an alternative dual response approach based on machine learning techniques instead of statistical analysis tools. A hybrid neural network-genetic algorithm has been proposed for the purpose of parameter design. A neural network is first constructed to model the relationship between responses and input factors. Mean and variance responses correspond to output nodes while input factors are used for input nodes. Using empirical process data, process parameters can be predicted without performing real experimentations. A genetic algorithm is then applied to find the optimum settings of input factors, where the neural network is used to evaluate the mean and variance response. A drug formulation example from pharmaceutical industry has been studied to demonstrate the procedures and applicability of the proposed approach.

• Korean Management Science Review
• /
• v.30 no.3
• /
• pp.55-70
• /
• 2013

In this study we suggested two optimization models to determine conversion weight of convertible bonds. The problem of this study is same as that of Park and Shim [1]. But this study used Value-at-Risk (VaR) for risk measurement instead of CVaR, Conditional-Value-at-Risk. In comparison with conventional Markowitz portfolio models, which use the variance of return, our models used VaR. In 1996, Basel Committee on Banking Supervision recommended VaR for portfolio risk measurement. But there are difficulties in solving optimization models including VaR. Benati and Rizzi [5] proved NP-hardness of general portfolio optimization problems including VaR. We adopted their approach. But we developed efficient algorithms with time complexity O(nlogn) or less for our models. We applied examples of our models to the convertible bond issued by a semiconductor company Hynix.

• Korean Journal of Computational Design and Engineering
• /
• v.15 no.5
• /
• pp.325-332
• /
• 2010

A lens system for mobile phone cameras is comprised of various lenses and designed so as to satisfy design requirements for responses such as a modular transfer function (MTF). However, it is difficult to manufacture and assemble camera modules to maintain the same performance compared with the designed camera modules, because of uncertainty. We should always design a lens system by considering uncertainty that can be caused by errors in the manufacturing and assembly process of mobile phone cameras. The robust optimization offers tools of making robust decisions with the consideration of design parameters, uncontrollable parameters, and the variance of the system. Using an efficient reliability analysis method and an optimization algorithm, we obtained robust optimization results that maximize the mean of MTF and minimize the standard deviation and proposed a new robust design process for a lens system.

• Nuclear Engineering and Technology
• /
• v.38 no.3
• /
• pp.225-230
• /
• 2006

With the advent of inexpensive computing power over the past two decades, applications of Monte Carlo radiation transport techniques have proliferated dramatically. At Los Alamos, the Monte Carlo codes MCNP5 and MCNPX are used routinely on personal computer platforms for radiation shielding analysis and dosimetry calculations. These codes feature a rich palette of variance reduction (VR) techniques. The motivation of VR is to exchange user efficiency for computational efficiency. It has been said that a few hours of user time often reduces computational time by several orders of magnitude. Unfortunately, user time can stretch into the many hours as most VR techniques require significant user experience and intervention for proper optimization. It is the purpose of this paper to outline VR strategies, tested in practice, optimized for several common radiation shielding tasks, with the hope of reducing user setup time for similar problems. A strategy is defined in this context to mean a collection of MCNP radiation transport physics options and VR techniques that work synergistically to optimize a particular shielding task. Examples are offered in the areas of source definition, skyshine, streaming, and transmission.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.4
• /
• pp.560-566
• /
• 2008

Today, as the power trades between generation companies and power customer are liberalized, the uncertainty level of operated power system is rapidly increased. Therefore, transmission operators as decision makers for transmission expansion are required to establish a deliberate investment plan for effective operations of transmission facilities considering forecasted conditions of power system. This paper proposes the methodology for the optimal solution of transmission expansion in deregulated power system. The paper obtains the expected value of transmission congestion cost for various scenarios by using occurrence probability. In addition, the paper assumes that increasing rates of loads are the probability distribution and indicates the location of expanded transmission line, the time for transmission expansion with the minimum cost for the future by performing the Montecarlo simulation. To minimize the investment risk as the variance of the congestion cost, Mean-Variance Markowitz portfolio theory is applied to the optimization model by the penalty factor of the variance. By the case study, the optimal solution for transmission expansion plan considering the feature of market participants is obtained.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.35 no.10
• /
• pp.1307-1313
• /
• 2011

The mechanical limited slip differential (LSD) in vehicles is being replaced by the electromagnetic LSD because of its fast response and better active control characteristics. The coil housing made of STS 304 is one of the most important parts in the solenoid assembly of the electromagnetic LSD. High geometrical accuracy is a prerequisite for the manufacture of such coil housings, but precision machining is difficult because of the use of STS 304 thin plate and the variance in machining variables. The aim of this study is to optimize the mean and variance of the shape accuracy in the coil housing by finding a robust solution for the machining process conditions. The mean and standard deviation of the jaw contact pressure, cutting speed, and feed rate are considered to be the major parameters for minimizing the geometrical mean and variance. The response surface model based on the second-order Taylor series is combined together to minimize the mean and variance of the shape accuracy of the coil housing.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2000.04b
• /
• pp.168-173
• /
• 2000

Design of experiments is utilized for exploring the design space and for building response surface models in order to facilitate the effective solution of multi-objective optimization problems. Response surface models provide an efficient means to rapidly model the trade-off among many conflicting goals. In robust design, it is important not only to achieve robust design objectives but also to maintain the robustness of design feasibility under the effects of variations, called uncertainties. However, the evaluation of feasibility robustness often needs a computationally intensive process. To reduce the computational burden associated with the probabilistic feasibility evaluation, the first-order Taylor series expansions are used to derive individual mean and variance of constraints. For robust design applications, these constraint response surface models are used efficiently and effectively to calculate variances of constraints due to uncertainties. Robust optimization of automotive seat is used to illustrate the approach.

• Journal of the Korean Society of Industry Convergence
• /
• v.25 no.4_2
• /
• pp.645-653
• /
• 2022

Recently, deep reinforcement learning has been applied to a variety of industries, such as games, robotics, autonomous vehicles, and data cooling systems. An algorithm called reinforcement learning allows for automated asset allocation without the requirement for ongoing monitoring. It is free to choose its own policies. The purpose of this paper is to carry out an empirical analysis of the performance of asset allocation strategies. Among the strategies considered were the conventional Mean- Variance Optimization (MVO) and the Proximal Policy Optimization (PPO). According to the findings, the PPO outperformed both its benchmark index and the MVO. This paper demonstrates how dynamic asset allocation can benefit from the development of a reinforcement learning algorithm.

• Journal of the Semiconductor & Display Technology
• /
• v.8 no.4
• /
• pp.1-5
• /
• 2009

This investigation is applied Taguchi method and the analysis of variance(ANOVA) to the reactive ion etching(RIE) characteristics of $SiO_2$ film coated on a wafer with Experimental Analysis and Optimization of $CF_4/O_2$ Plasma Etching Process mixture. Plans of experiments via nine experimental runs are based on the orthogonal arrays. A $L_9$ orthogonal array was selected with factors and three levels. The three factors included etching time, RF power, gas mixture ratio. The etching rate of the film were measured as a function of those factors. In this study, the etching thickness mean and uniformity of thickness of the RIE are adopted as the quality targets of the RIE etching process. The partial factorial design of the Taguchi method provides an economical and systematic method for determining the applicable process parameters. The RIE are found to be the most significant factors in both the thickness mean and the uniformity of thickness for a RIE etching process.