• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.307-332
• /
• 2022

One of the most intensively conducted research areas in business application study is a bankruptcy prediction model, a representative classification problem related to loan lending, investment decision making, and profitability to financial institutions. Many research demonstrated outstanding performance for bankruptcy prediction models using artificial intelligence techniques. However, since most machine learning algorithms are "black-box," AI has been identified as a prominent research topic for providing users with an explanation. Although there are many different approaches for explanations, this study focuses on explaining a bankruptcy prediction model using a counterfactual example. Users can obtain desired output from the model by using a counterfactual-based explanation, which provides an alternative case. This study introduces a counterfactual generation technique based on a genetic algorithm (GA) that leverages both domain knowledge (i.e., causal feasibility) and feature importance from a black-box model along with other critical counterfactual variables, including proximity, distribution, and sparsity. The proposed method was evaluated quantitatively and qualitatively to measure the quality and the validity.

• Korean Chemical Engineering Research
• /
• v.55 no.3
• /
• pp.332-340
• /
• 2017

This study aims to propose a new optimization-based approach for design and analysis of the stand-alone hybrid energy supply system using renewable energy sources (RES). In the energy supply system, we include multiple energy production technologies such as Photovoltaics (PV), Wind turbine, and fossil-fuel-based AC generator along with different types of energy storage and conversion technologies such as battery and inverter. We then select six different regions of Korea to represent various characteristics of different RES potentials and demand profiles. We finally designed and analyzed the optimal RES stand-alone energy supply system in the selected regions using multiobjective optimization (MOOP) technique, which includes two objective functions: the minimum cost and the minimum $CO_2$ emission. In addition, we discussed the feasibility and expecting benefits of the systems by comparing to conventional systems of Korea. As a result, the region of the highest RES potential showed the possibility to remarkably reduce $CO_2$ emissions compared to the conventional system. Besides, the levelized cost of electricity (LCOE) of the RES-based energy system is identified to be slightly higher than conventional energy system: 0.35 and 0.46 $/kWh, respectively. However, the total life-cycle emission of $CO_2$ ($LCE_{CO2}$) can be reduced up to 470 g$CO_2$/kWh from 490 g$CO_2$/kWh of the conventional systems.

• Journal of Korean Society of Forest Science
• /
• v.85 no.2
• /
• pp.172-179
• /
• 1996

This paper described the application of fuzzy multiobjective linear programming to solving a multiple-use problem of forest resources management. At first the concepts of linear programming, fuzzy linear programming and fuzzy multiobjective linear programming were introduced briefly. In order to illustrate a role of fuzzy multiobjective linear programming in the process of multiple-use forest planning, the natural recreation forest in Mt. Yoomyung was selected for this study. A fuzzy multiobjective linear programming model is formulated with data obtained from this Mt. Yoomyumg natural recreation forest to solve the multiple-use management planning problem of forest resources. Finally, the results, which were obtained from the calculation of this model, were discussed. The maximal value of the membership function(${\lambda}$) was 0.29, when the timber production and the forest recreation function were optimized at the same time through the fuzzy multiobjective linear programming. The cutting area in each period was 102.7ha, while total cutting area was 410.8ha for 4 periods. During 4 periods $57,904m^3$ will be harvested from this natural recreation forest and at the same time total visitors were estimated to be about 8.6 millions persons.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.4 no.3
• /
• pp.27-42
• /
• 1984

The complexity and expensiveness of water resources projects have made optimum operation and design by computer-based techniques of increasing interest in recent years. Water resources problems in real world need many decisions under numerous constraints. In addition there are nonlinearities in the state and return function. This mathematical and technical troublesome must be overcome so that the optimum operation polices are determined. Then traditional dynamic optimization method encountered two major-cruxes: variable discretization and appearance of constraints. Even several recent methods which based on the Differential Dynamic Programming(DDP) have some difficulties in handling of constraints. This paper has presented New DDP which is applicable to multi-reservoir control. It is intended that the method suggested here is superior to abailable alternatives. This belief is supported by analysis and experiments(New DDT does not suffer course of dimensionality and requires no discretization and is able to handle easily all constraints nonlinearity).

• Proceedings of the Korea Water Resources Association Conference
• /
• 2012.05a
• /
• pp.640-640
• /
• 2012

낙동강 홍수예경보 시스템은 낙동강 유역의 홍수피해 방지를 위해 1986년에 구축되어 낙동강홍수통제소에서 운영되어 온 이래로 여러 차례에 거친 시스템의 개선 및 보완을 통해 현재의 시스템을 갖추게 되었다. 그러나 4대강 사업을 통해 시행된 하도 준설 및 보 설치로 인한 하도 조건의 변경과 기존의 저류함수모형 및 수위-유량 관계식을 이용한 수위예측의 한계로 인해 낙동강 하도에 대한 수리해석모형 구축의 필요성이 대두되었다. 이에 따라 낙동강 홍수통제소에서는 기존의 저류함수모형을 이용한 강우-유출 해석모형과 낙동강 본류 및 주요 지류에 대한 수리해석 모형을 구축하여 연계하는 과업을 수행 중에 있다. 본 연구에서는 하천기본계획의 설계홍수량의 산정시 적용되는 HEC-HMS 모형을 통해 강우-유출해석모형을 구축하고, 낙동강 본류 및 8개 지류에 대해 FLDWAV 모형을 이용해 수리해석 모형을 구축하여 연계하였다. 수자원단위지도의 표준유역과 수위관측소 지점을 기반으로 하여 낙동강 유역을 287개의 소유역으로 분할하였고, 271개의 분할하도 및 10개의 다목적 댐 방류량을 반영하여 강우-유출 모형을 구축하였다. 수치지형도 및 토양도, 토지이용현황도를 통해 유역유출 및 하도유출에 대한 매개변수 산정하였고, 낙동강 본류 및 지류내의 주요 수위관측소를 유량의 검보정 지점으로 설정하였다. 수리학적 모형 구축을 위해 낙동강 본류의 383개의 단면 및 8개 지류의 497개 단면을 반영하였고, 그 이외의 6개 주요 지류는 측방유입으로 처리하였으며 낙동강 본류에 신설된 8개의 다기능보의 운영을 반영하였다. 각각 구축된 강우-유출 모형과 수리학적 모형은 모듈화하여 연계하였으며, 현재 낙동강홍수통제소에서 운영되고 있는 낙동, 왜관, 현풍, 진동, 삼랑진, 구포, 동촌수위관측소를 홍수예보지점으로 선정하여 모형의 검보정을 실시하였다. 구축된 모형은 낙동강홍수통제소의 홍수예보모형의 계산결과와 비교하여 적용성 및 효율성을 입증할 수 있을 것으로 판단되며, 낙동강에서의 실시간 홍수예 경보를 위한 홍수예보모형으로 활용될 수 있을 것으로 사료된다.

• Journal of Fisheries and Marine Sciences Education
• /
• v.27 no.5
• /
• pp.1352-1363
• /
• 2015

This study was carried out to investigate the characteristics of the dependent variables depending on the condition changes of independent variable of the operation and the material during the production of collets added with rice and dried shrimp by using single extruder to utilize as basic data for the manufacture of extrusion collets. A total of 7 independent variables were set up as a raw, 20, 40 and 60 mesh for the powder particle size of rice; 12, 14, 16 and 18% for the moisture content of rice; 2, 4, 6 and 8% for the addition amount of dried shrimp; 90, 95, 100 and $110^{\circ}C$ for the barrel temperature; 210, 280 and 340 rpm for the screw speed; 4, 6, 8 and 10 mm for the discharge port diameter; 30, 40, 50 and 60 kg/h for the input amount of the mixed material. The characteristics of the dependent variables including puffing ratio, moisture content, lightness, uniformity, productivity of collets was to be studied by changing the conditions of the independent variables. As a results of this study, 20 mesh of powder particle size of rice, 14% of moisture content of rice, 4% of addition amount of dried shrimp, $100^{\circ}C$ of barrel temperature, 280 rpm of screw speed, 6 mm of discharge port diameter and 50 kg/h of input amount of mixed material were found to be the most preferable over other independent variables for the production of extrusion collets. In conclusion, it is necessary to set the independent variable in order to produce the high quality collets added with the rice as the main raw material and dried shrimp as the sub-materials.

• Journal of the Society of Naval Architects of Korea
• /
• v.33 no.2
• /
• pp.56-64
• /
• 1996

In this study, a review is made of the previous work(Ref. 1 and 5) for the development of the limit design method of the flat rectangular grillages under the lateral pressure. And the effect of the in-plane loads on the collapse theory is considered. The main part of the work is devoted in developing the standard design method of grillages under the criteria of minimum weight and minimum cost. In the final part, it was shown that Pareto solution methods can be easily applied to structural optimization with the multiple objectives, and the designer can have an appropriate choice from those Pareto optimal solutions.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.14 no.6
• /
• pp.1391-1407
• /
• 1990

In this paper, optimization procedures are presented considering the static and dynamic constraints for laminated composite plate and hybrid laminated composite plate subject to concentrated load on center of the plates. Design variables for this problem are ply angle or ply thickness. Deflection, natural frequency and specific damping capacity are considered as constraints. Using a recursive linear programming method, the nonlinear optimization problems are solved. By introducing the design scaling factor, the number of iterations is reduced significantly. Composite plates could be designed optimally combined with FEM analysis under various conditions. In the optimization procedure, verification for both analysis and design of the laminated composite plates are compared with the results of the others. Various design results are presented for the laminated composite plates and hybrid laminated composite plates.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.28 no.7
• /
• pp.1038-1046
• /
• 2004

In order to acquire high-speed and high-precision performances in servomechanisms, an integrated design method have been proposed. Based on strict mathematical modeling and analysis of system performance according to design and operating parameters, a nonlinear constrained optimization problem including the relevant subsystem parameters of the servomechanism is formulated. Optimum design results of mechanical and electrical parameters are obtained according to the design parameters specified by designers through the integrated design processes. Motors are optimally selected from the servo motor database. Both the geometric errors referring to Abbe offset and the contour errors are minimized while required constraints such as stability conditions and saturated conditions are satisfied. This design methodology both offers the improved possibility to evaluate and optimize the dynamic motion performance of the servomechanism and improves the quality of the design process to achieve the required performance for high-speed/precision servomechanisms.

• Nuclear Engineering and Technology
• /
• v.27 no.6
• /
• pp.845-852
• /
• 1995

We investigate the concurrent solution of differential equations by the waveform relaxation (WR) method, an iterative method for analyzing linear and nonlinear dynamical systems in the time do-main. The method, at each iteration, decomposes the dynamical system into several subsystems, each of which is analyzed for the entire given time interval. The method, when efficiently implemented, results in algorithms with a highly parallelizable concurrent fraction. In this paper, the waveform relaxation method is introduced and applied to two types of reactor dynamics problems. It is concluded that the U method can be applied to reactor dynamics equations, but that its parallel performance on the KMRR dynamics is only modest.