Satisficing Trade-off Method Research Articles

Pareto Optimal Solutions for Network Defense Strategy Selection Simulator in Multi-Objective Reinforcement Learning

Using Pareto optimization in Multi-Objective Reinforcement Learning (MORL) leads to better learning results for network defense games. This is particularly useful for network security agents, who must often balance several goals when choosing what action to take in defense of a network. If the defender knows his preferred reward distribution, the advantages of Pareto optimization can be retained by using a scalarization algorithm prior to the implementation of the MORL. In this paper, we simulate a network defense scenario by creating a multi-objective zero-sum game and using Pareto optimization and MORL to determine optimal solutions and compare those solutions to different scalarization approaches. We build a Pareto Defense Strategy Selection Simulator (PDSSS) system for assisting network administrators on decision-making, specifically, on defense strategy selection, and the experiment results show that the Satisficing Trade-Off Method (STOM) scalarization approach performs better than linear scalarization or GUESS method. The results of this paper can aid network security agents attempting to find an optimal defense policy for network security games.

Investor-friendly and robust portfolio selection model integrating forecasts for financial tendency and risk-averse

This paper proposes an investor-friendly random fuzzy portfolio selection model considering both robustness and adjustment of future asset returns derived from investor’s forecasts for financial tendency using a fuzzy inference method. It is important to predict the price or the return of each asset appropriately considering current market trends in portfolio optimization. In this paper, a standard multi-factor model, namely Arbitrage Pricing Theory (APT), is introduced as an asset pricing model. In addition, in order to extend standard APT by integrating important rules of current markets trends derived from technical analysis and fundamental analysis, each factor of APT is assumed to be a random fuzzy variable whose mean is adjusted by the fuzzy reasoning method, particularly product–sum-gravity method. Furthermore, it is also important for the investor to reduce the worst case of the total loss in terms of risk-averse. Therefore, worst-case conditional Value-at-Risk, which is a robust programming approach without assuming some specific random distribution, is considered. Since the proposed model is formulated as a biobjective programming problem both minimizing the value of worst-case conditional Value-at-Risk and maximizing the total expected return, it is equivalently transformed into the deterministic nonlinear programming problem using the satisficing trade-off method, and the efficient algorithm to obtain the optimal portfolio is developed. By solving our proposed model, the investor can obtain the risk-averse optimal portfolio with the large total return complying with current market trends.

Modal Sensor Design based on Satisficing Trade-off Method and Its Application to Multi-modal Vibration Control

Modal Sensor Design based on Satisficing Trade-off Method and Its Application to Multi-modal Vibration Control

多目的最適設計法を利用した気球VLBIにおける副鏡調整機構の故障を考慮した主鏡面構造設計

The balloon-borne VLBI (very long baseline interferometry) is a radio telescope for space observation from the stratosphere in the submillimeter wave band. The primary reflector has an aperture of 3 m in diameter whose degradation of aperture efficiency is required to be less than 17 % under the deformation due to variations of elevation angle and temperature during observation.In order to alleviate the deterioration of the aperture efficiency, the sub-reflector is equipped with a focal position adjustment mechanism. However, the adjustment mechanism may fail during observation, so that the focal position will be fixed at the prescribed position.This study evaluates the effect of the adjustment mechanism failure on the aperture efficiency through multiobjective optimization approach. The design problem has thirteen objective functions that correspond to the nominal observation condition and the other six conditions considering elevation angles and temperatures with normal and failure cases of the adjustment mechanism.The design problem is solved using the satisficing trade-off method (STOM). As STOM can obtain the single Pareto solution corresponding to the user's preference for each objective function by introducing an aspiration level, the trade-off analysis is easily performed.

Robust multiobjective optimization method using satisficing trade-off method

This study proposes a robust multiobjective optimization approach using the satisficing tradeoff method (STOM). STOM is a multiobjective optimization method that obtains a highly accurate single Pareto solution. Conventionally, a robust design is formulated as a single-objective optimization problem, where the objective function is defined as the weighted sum of the mean and standard deviation of the performance index. In this study, the mean and standard deviation are formulated as individual objective function. The effect of uncertainty can be investigated through Pareto surface. In STOM, the multiobjective optimization problem is transformed into the equivalent single objective problem by introducing an aspiration level. As the obtained single Pareto solution corresponds to the aspiration level that implies the ratio of the designer's desired objective function, the designer can investigate only the desired space in detail through setting the aspiration without obtaining full Pareto surfaces. The validity of using STOM for a robust multiobjective optimization problem is discussed using numerical examples.

2514 多目的最適化手法を用いた二足歩行ロボットの関節動作3次元軌道の生成(設計と最適化VII)

The purpose of this study is to generate the three dimensional gait of a two-legged robot to avoid obstacles. It is expected that two-legged robot can avoid obstacle more smoothly in the same way that animal and human adjust stride naturally to step over obstacles. Stepping points are determined optimally under the condition of maximizing walking speed and minimizing energy for walking. The gait generation problem is reduced to a combinatorial optimization problem solved by using genetic algorithm. Orbits of toes and hip between stepping points are generated by means of parametric modeling. The stable walking patterns are obtained under the condition of the maximizing walking speed and the minimizing energy consumption. The pareto front of the multi-objective optimization for the given robot model is visualized in advance by the MOGA, the optimum walking pattern is determined by using the satisficing trade-off method. The effectiveness of the proposed method is shown by simulation results.

Reliability-based multiobjective optimization using the satisficing trade-off method

Reliability-based multiobjective optimization using the satisficing trade-off method

An improved procedure for updating finite element model based on an interactive multiobjective programming

Finite element model updating is an optimization problem to identify and correct uncertain modeling parameters. In conventional model updating, physically incompatible criteria, which designate differences between analytical and experimental results, are combined into a single-objective function using weighting factors. There are no general rules for selecting the weighting factors since they are not directly related to the dynamic behavior of the updated model. Thus, a necessary approach is to solve the time-consuming optimization problem repeatedly by varying the values of weighting factors until a satisfactory solution is obtained. In this work, an interactive multiobjective optimization technique called satisficing trade-off method is introduced to avoid the difficulty. It is relatively easy to state what kind of solutions is satisfactory considering the correlations of the initial FE model with the experimental results, the importance of individual modal properties, and measurement errors, etc. The satisficing trade-off method uses this information directly in the optimization process and finds a Pareto solution which is nearest to the given information. Moreover, as the method provides the tangent hyperplane which approximates the Pareto surface in the neighborhood of the obtained Pareto solution, the desired updated model can be found in a few iterations.

2405 有機EL素子における射出光特性向上のための多目的最適設計に関する検討

Optical characteristics of design specifications required for organic light emitting diode are emitted light intensity, spectrum, color rendering and so on. The multi-objective design of the OLED thin film structure is studied for simultaneously satisfying these specifications using a weighted sum method and a satisficing trade-off method. It is hardly to balance between objectives with trade-off relationship in the weight sum method. On the other hand, it is possible to easily perform OLED thin-film structure design considering multiple emission properties by appropriately setting the aspiration level in the satisficing trade-off method.

Interactive Multi-objective Route Planning for Sightseeing on Time-expanded Networks under Various Conditions

This paper proposes a versatile and interactive route planning problem for sightseeing under various conditions based on constraints of required traveling times and total satisfaction of sightseeing activities, which are time-dependent. In general, traveling times among sightseeing places and the satisfactions of activities also change dependent on various conditions such as weather, climate and season. It is important to represent a satisfying route to ensure the large satisfaction even if any assumable condition happens. In order to incorporate the above-mentioned situations, a multi-objective route planning problem is formulated as a network optimization problems based on Time-Expanded Network to represent time-dependent parameters in a static network. Furthermore, an interactive approach based on the Satisficing Trade-Off Method is introduced to transform the multi-objective into the single. In addition, the proposed problem is equivalently transformed into an extended model of network optimization problems by introducing parameters, and the interactive algorithm is developed.

Compromise point incorporating trade-off ratio in multi-objective optimization

The aims of multi-objective optimization are (1) to find pareto-optimal solutions and (2) to analyze the trade-off between conflicting objectives. This paper proposes an interactive method for solving multi-objective optimization problems using the satisficing trade-off method (STOM). In particular, we introduce a trade-off matrix to quantitatively analyze the trade-off between conflicting objectives. Each element of the trade-off matrix consists of a projection matrix of active constraints and gradients of objective functions. In addition, the compromise point and the compromise solution incorporating the trade-off ratio that the decision-maker desires are defined in this paper. The technique to obtain the compromise point is proposed in this paper. Through numerical examples, the validity proposed method is examined.

A Buffer Material Optimal Design in the Radioactive Wastes Geological Disposal using the Satisficing Trade-off Method and the Self-organizing Map

A Buffer Material Optimal Design in the Radioactive Wastes Geological Disposal using the Satisficing Trade-off Method and the Self-organizing Map

Fundamental Study on Ergonomic Design Method Based on Comprehensive Evaluation of Muscle Load Using Surface Electromyography

This paper proposes a formulation method on comprehensive evaluation of plural muscle load by applying the methodology of multi-objective optimization. Muscle loads are formulated as objective functions that should be minimized, and comprehensive evaluation of muscle loads is formulated as a multi-objective optimization problem. Design of handrail position for sit-to-stand movement was used as the case study of the proposed method. The height and distance of handrail are design variable, and surface electromyography of anterior deltoid, triceps brachii, rectus femoris, and tibialis anterior were measured when participants stood from a chair with handrails. The ratio of the maximum voluntary contraction (MVC) was used as the indicator of muscle loads. Result from the measurement, it is found that the muscle loads of triceps brachii and tibialis anterior have trade-off relationship on the assumption that %MVC indicates the muscle loads. The higher and the closer the position of handrail is, the higher the muscle load of triceps brachii is, conversely, the lower the muscle load of tibialis anterior is. The two muscle loads were approximated as quadratic functions, and formulated as a multi-objective optimization problem. The optimization problem was solved by the satisficing trade-off method, and the Pareto optimal solution for the participants was obtained at which the height is 234 mm and the distance is 585 mm. The comprehensive evaluation of the two muscle loads probably expresses the sense of the participants more precisely than the use of either of the two muscles.

2206 RBFネットワークによる多目的逐次近似最適化(OS7-1 近似最適化I,OS7 近似最適化)

In this paper, a sequential approximate multi-objective optimization procedure by the Radial Basis Function (RBF) network with the Satisficing Trade-Off Method (STOM) is proposed. The sampling strategy is an important issue in the sequential approximate optimization. In this paper, the density function and the pareto fitness function are proposed. The objective of the density function is to find the sparse region in the design variable space. New samplings point are obtained by optimizing the density function. The objective of the pareto fitness function is to find the approximate set of pareto optimal solutions from the given data. New sampling point is obtained by optimizing the pareto fitness function. Both functions are constructed by the RBF network. By using both functions, the approximate set of pareto optimal solutions can be found effectively even when the set of pareto optimal solutions are separeted. Through simple numerical examples, the validity of proposed sampling strategy is examined.

RBFネットワークによる多目的逐次近似最適化(機械力学,計測,自動制御)

In this paper, a sequential approximate multi-objective optimization procedure by the Radial Basis Function (RBF) network with the Satisficing Trade-Off Method (STOM) is proposed. The sampling strategy is an important issue in the sequential approximate optimization. In this paper, the density function and the pareto fitness function are proposed. The objective of the density function is to find the sparse region in the design variable space. New samplings point are obtained by optimizing the density function. The objective of the pareto fitness function is to find the approximate set of pareto optimal solutions from the given data. New sampling point is obtained by optimizing the pareto fitness function. Both functions are constructed by the RBF network. By using both functions, the approximate set of pareto optimal solutions can be found effectively even when the set of pareto optimal solutions are separeted. Through simple numerical examples, the validity of proposed sampling strategy is examined.

212 プラント配置設計の多目的最適化

While engineering a power plant layout, objective functions and constraints of the power plant frequently conflict each other in terms of factors such as safety, economic costs, maintainability, and construction term of works. This paper proposes a multi-objective layout optimization system for a power plant through effective interaction between the designer and the computer; this system is developed by applying an interactive Satisficing Trade-Off Method (STOM) for multiple-criteria decision-making. In order to obtain the layout candidates for decision-making, the authors present a suitable genetic algorithm (GA) approach to arrange the plot plan for the layout of buildings and equipment in a site. In this GA approach, the coding method and evolution operation method used for obtaining a diverse optimal layout solution are employed with a low computational load and simple design information. Finally, the validity and usefulness of this proposed system is demonstrated by solving a layout design problem for an actual power plant.

多目的最適設計におけるトレードオフ分析法(機械力学,計測,自動制御)

The aims of the multi-objective design optimization are to find the pareto optimal solutions, and are also to analyze the trade-off between the objectives. The interactive multi-objective design optimization based on the Satisficing Trade-Off Method (STOM) is used in this paper. Especially, the trade-off matrix to analyze the trade-off between objectives quantitatively is introduced. Each element of trade-off matrix consists of the projection matrix of active constraints and the gradient of objective functions. Additionally, it is possible to determine the new aspiration level by using the trade-off matrix. To obtain the new aspiration level, the compromise point and solution are defined by using the trade-off ratio. To obtain the compromise point, the designer input the trade-off ratio between objectives and can obtain the new aspiration level. It is also possible to apply the trade-off matrix to the automatic trade-off method, which is a method to determine the new aspiration level. Through basic numerical examples, the validity is examined.

J1201-4-3 発電プラント機器配置計画の多目的最適化(解析・設計の高度化・最適化IV)

On the design issues of power plant layout engineering, it frequently occurs that objective functions or criteria mutually conflict one another in terms of safety, economical costs, maintainability, etc. This paper proposes a multi-objective layout optimization system for a thermal power plant through effective interaction between the designer and the computer by applying a satisficing trade-off method. In addition, the authors present a suitable genetic algorithm approach to arranging the plot plan for equipments and buildings layout.

2311 多目的最適設計におけるトレードオフ分析法

The aims of the multi-objective design optimization are to find the pareto optimal solutions, and are also to analyze the trade-off between the objectives. The interactive multi-objective design optimization based on the Satisficing Trade-Off Method (STOM) is used in this paper. Especially, the trade-off matrix to analyze the trade-off between objectives quantitatively is introduced. Each element of trade-off matrix consists of the projection matrix of active constraints and the gradient of objective functions.

Intelligent Start-Up Schedule Optimization System for a Thermal Power Plant

This paper proposes an intelligent start-up schedule optimization system for a thermal power plant. This system consists of a dynamic simulation, a neural network, and an interactive multi-objective programming technique. In this study, a new intelligent optimization method using the neural network and a genetic algorithm to realize a satisficing trade-off method, which is an interactive multi-objective programming technique, has been developed and introduced into this system. The features of this system are as follows. (1) The start-up schedule can be optimized based on multi-objective evaluation and (2) an optimal and flexible start-up schedule can be determined with a reasonable computing time and calculation accuracy through human-computer interactions. The system is applied to a simulation for a combined cycle power plant, and to optimize from among multiple objectives, based on varying daily requirements. The application results show that optimal and flexible start-up schedules can be obtained within a reasonable computing time and with acceptable calculation accuracy.