• 제목/요약/키워드: Loss Function Types

검색결과 131건 처리시간 0.026초

손실함수의 형태를 고려한 생산자 규격한계의 결정 (Decision of Producer's Specification Limits Considering Types of Loss Function)

  • 김동혁;정영배
    • 산업경영시스템학회지
    • /
    • 제41권3호
    • /
    • pp.145-153
    • /
    • 2018
  • Taguchi regarded the concept of quality as 'total loss to society due to fluctuations in quality characteristics from the time of supplied to the customer.' The loss function is a representative tool that can quantitatively convert the loss that occurs due to the deviation of the quality characteristic value from the target value. This has been utilized in various studies with the advantage that it can change the social loss caused by fluctuation of quality characteristics to economic cost. The loss function has also been used extensively in the study of producer specification limits. However, in previous studies, only the second order loss function of Taguchi is used. Therefore, various types of losses that can occur in the process can't be considered. In this study, we divide the types of losses that can occur in the process considering the first and second loss functions and the Spiring's reflected normal loss function, and perform total inspection before delivering the customer to determine the optimal producer specification limit that minimizes the total cost. Also, we will divide the quality policy for the products beyond the specification limits into two. In addition, we will show the illustration of expected loss cost change of each model according to the change of major condition such as customer specifications and maximum loss cost.

Selection of Optimal Values in Spatial Estimation of Environmental Variables using Geostatistical Simulation and Loss Functions

  • Park, No-Wook
    • 한국지구과학회지
    • /
    • 제31권5호
    • /
    • pp.437-447
    • /
    • 2010
  • Spatial estimation of environmental variables has been regarded as an important preliminary procedure for decision-making. A minimum variance criterion, which has often been adopted in traditional kriging algorithms, does not always guarantee the optimal estimates for subsequent decision-making processes. In this paper, a geostatistical framework is illustrated that consists of uncertainty modeling via stochastic simulation and risk modeling based on loss functions for the selection of optimal estimates. Loss functions that quantify the impact of choosing any estimate different from the unknown true value are linked to geostatistical simulation. A hybrid loss function is especially presented to account for the different impact of over- and underestimation of different land-use types. The loss function-specific estimates that minimize the expected loss are chosen as optimal estimates. The applicability of the geostatistical framework is demonstrated and discussed through a case study of copper mapping.

다양한 손실 함수를 이용한 음성 향상 성능 비교 평가 (Performance comparison evaluation of speech enhancement using various loss functions)

  • 황서림;변준;박영철
    • 한국음향학회지
    • /
    • 제40권2호
    • /
    • pp.176-182
    • /
    • 2021
  • 본 논문은 다양한 손실 함수에 따른 Deep Nerual Network(DNN) 기반 음성 향상 모델의 성능을 비교 평가한다. 베이스라인 모델로는 음성의 위상 정보를 고려할 수 있는 복소 네트워크를 사용하였다. 손실 함수는 두 가지 유형의 기본 손실 함수, Mean Squared Error(MSE)와 Scale-Invariant Source-to-Noise Ratio(SI-SNR)를 사용하였으며 두 가지 유형의 지각 기반 손실 함수 Perceptual Metric for Speech Quality Evaluation(PMSQE)과 Log Mel Spectra(LMS)를 사용한다. 성능은 각 손실 함수의 다양한 조합을 사용하여 얻은 출력을 객관적인 평가와 청취 테스트를 통해 측정하였다. 실험 결과, 지각기반 손실 함수를 MSE 또는 SI-SNR과 결합하였을 때 전반적으로 성능이 향상되며, 지각기반 손실함수를 사용하면 객관적 지표에서 약세를 보이는 경우라도 청취 테스트에서 우수한 성능을 보임을 확인하였다.

가족성 난청 (Familial Hearine Loss)

  • 전광수;장선문;박찬일
    • 대한기관식도과학회:학술대회논문집
    • /
    • 대한기관식도과학회 1978년도 제12차 학술대회연제 순서 및 초록
    • /
    • pp.7.2-7
    • /
    • 1978
  • 유전성 난청은 거의 60종이 보고되었는데 대부분 동일 유전자에 의하여 다른 선친성 기형 또는 질병을 동반하고 있으나 단지 12여종에서 다른 기형을 동반하지 않고 난청만을 보이게 된다. 저자들은 유전형태가 비슷한 가족성 난청을 가진 두 가계를 경험하였기에 보고하는 바이다. 이들은 상염색체성 우성 유전의 형태를 보였고 소년기때 난청이 발생하였으며 점진적이고 양측이 동일한 중등도 내지 고도의 감각신경성 난청을 보였다. 청력도상 수평형을 보였고 전정기능 검사는 정상이었으며 동반된 기형을 발견할 수 없었다.

  • PDF

Allelic Distribution of OsPRR37, a Major Heading Date Gene in Korean Rice Cultivars

  • Hayeong Lee;Yurim Kim;Jiheon Han;Sieun Choi;Yeeun Jun;Hosun Chun;Soonhwa Kwak;Youngjun Mo
    • 한국작물학회:학술대회논문집
    • /
    • 한국작물학회 2022년도 추계학술대회
    • /
    • pp.225-225
    • /
    • 2022
  • Rice is a major food crop consumed by approximately half of the world's population. Heading date is one of the major agronomic traits and has a wide impact on the productivity and quality of rice. Recently, shortening the growth period of rice through modulating heading date has been emphasized as one of the most effective strategies for reducing methane emissions from paddy fields. OsPRR37, a major heading date gene in rice, represses flowering under both short-day and long-day conditions. Plants carrying the loss-of-function alleles of OsPRR37 have been reported to flower approximately seven days and 20 days earlier than those carrying the functional alleles in short day and long day conditions, respectively. In this study, we investigated the nucleotide sequence variation existing in the exonic regions of OsPRR37 and catalogued the allelic distribution in 208 Korean rice cultivars. We used four sets of primers for amplifying and sequencing the eight exons of OsPRR37. As a result, two types of loss-of-function alleles and four types of functional alleles were found in 208 Korean rice cultivars. Interestingly, only three cultivars (Jinbuolbyeo, Jinseolchal, and Mimyeon) carried loss-of-function alleles while 205 carried functional alleles, indicating that OsPRR37 loss-of-function alleles have been used very rarely in Korean rice breeding programs. To generate useful information for the development of early-maturing rice cultivars, our future work will focus on analyzing the effect of different OsPRR37 alleles on heading date and other major agronomic traits.

  • PDF

EDF 기대손실에 기초한 공정능력지수의 붓스트랩 신뢰구간 (Bootstrap Confidence Intervals of the Process Capability Index Based on the EDF Expected Loss)

  • 임태진;송현석
    • 품질경영학회지
    • /
    • 제31권4호
    • /
    • pp.164-175
    • /
    • 2003
  • This paper investigates bootstrap confidence intervals of the process capability index(PCI) based on the expected loss derived from the empirical distribution function(EDF). The PCI based on the expected loss is too complex to derive its confidence interval analytically, so the bootstrap method is a good alternative. We propose three types of the bootstrap confidence interval; the standard bootstrap(SB), the percentile bootstrap(PB), and the acceleration biased­corrected percentile bootstrap(ABC). We also perform a comprehensive simulation study under various process distributions, in order to compare the accuracy of the coverage probability of the bootstrap confidence intervals. In most cases, the coverage probabilities of the bootstrap confidence intervals from the EDF PCI turned out to be more accurate than those from the PCI based on the normal distribution. It is expected that the bootstrap confidence intervals from the EDF PCI can be utilized in real processes where the true distribution family may not be known.

Bayesian Estimation of Uniformly Stochastically Ordered Distributions with Square Loss

  • Oh, Myong-Sik
    • Communications for Statistical Applications and Methods
    • /
    • 제18권3호
    • /
    • pp.295-300
    • /
    • 2011
  • The Bayesian nonparametric estimation of two uniformly stochastically ordered distributions is studied. We propose a restricted Dirichlet Process. Among many types of restriction we consider only uniformly stochastic ordering in this paper since the computation of integrals is relatively easy. An explicit expression of the posterior distribution is given. When square loss function is used the posterior distribution can be obtained by easy integration using some computer program such as Mathematica.

$MAP1, MAP2/G/1 FINITE QUEUES WITH SERVICE SCHEDULING FUNCTION DEPENDENT UPON QUEUE LENGTHS

  • Choi, Doo-Il;Lee, Sang-Min
    • 대한수학회보
    • /
    • 제46권4호
    • /
    • pp.673-689
    • /
    • 2009
  • We analyze $MAP_1,\;MAP_2$/G/1 finite queues with service scheduling function dependent upon queue lengths. The customers are classified into two types. The arrivals of customers are assumed to be the Markovian Arrival Processes (MAPs). The service order of customers in each buffer is determined by a service scheduling function dependent upon queue lengths. Methods of embedded Markov chain and supplementary variable give us information for queue length of two buffers. Finally, the performance measures such as loss probability and mean waiting time are derived. Some numerical examples also are given with applications in telecommunication networks.

격자필터에 의한 배기계의 소음특성 해석 (Analysis of Acoustical Performance of Exhaust System by Lattice Filter)

  • 황원걸
    • 한국정밀공학회지
    • /
    • 제12권5호
    • /
    • pp.126-135
    • /
    • 1995
  • An exhaust muffler is one of the simple and effective means to meet the demand for a quiet vehicle, and it deserves a close attention to effectively reduce the engine noise. The transfer function technique is one of the tools that have been used to analyze the noise characteristics of the exhaust muffler. In this paper we obtained a transfer function using the forward-going and backward-going components of sound pressure in the exhaust muffler, which is compatible wiht lattice filter algorithm. This form of transfer function is obtained for the basic elements of a muffler, such as uniform tube, open termin- ation, closed termination, anechoic termination, expansion, contraction, extended-tube resonator, hole, Helmholtz resonator, and concentric hole-cavity resonator. The results are combined to produce the transfer function of various types of mufflers. With this transfer function we calculate the transmission and insertion losses of mufflers, and examined the effects of various design parameters. Comparisons were made between the calculation and experimental results, which showed a good agreement, and we conclude that the transfer function of lattice form can be used to analyze the noise characteristics of the exhaust mufflers.

  • PDF

전기유변유체의 점탄성 특성에 관한연구 (Viscoelastic properties of electrorheological fluids)

  • 최윤대;김상국
    • 한국재료학회지
    • /
    • 제2권3호
    • /
    • pp.220-227
    • /
    • 1992
  • 전기유변유체(electrorheological fluid : ERF)는 크기가 $10^{-6}m$정도를 가진 분극성이 강한 입자 분말을 유전유체속에 혼합하여 만든 콜로이드용액으로서 외부전기 장에 의하여 액체상태로부터 겔(gel)상태로 상태변환을 한다. 상태변환과정은 가역과정이며 $10^{-3}$초 정도의 빠른 시간내에 이루어진다. 본 논문에서는 주기적응력이나 주기적 변형이 가해지는 강제진동실험을 통하여 ERF의 점탄성 특성을 조사하였다. 점탄성은 복소수로서 실수부인 동적전단탄성율(storage shear modulus, G')과 허수부인 동적손실율(loss modulus, G")의 합으로 표현된다. 동적전단탄성율은 강제진동의 진폭과 주파수의 함수로 외부전기장에 의하여 크게 변화되었으나 손실계수(loss factor, ${\eta}=G"/G'$)는 진동의 진폭, 진동수 및 외부전기 장에 거의 독립적이었다. 실험(1)은 강제진동의 진폭을 고정하고 진동수를 변화하는 방법(frequency sweep)과 실험(2)는 진동 진폭을 변화시키는 방법(amplitude sweep)으로 실행했다. 본 실험에서는 cornstarch입자를 corn oil에 혼합한 것과 zeolite입자를 silicone oil에 흔합한 콜로이드 용액이 ERF로서 사용되었으며 인가된 전기장은 약$(1~3){\times}10^6V/m$의 직류전기장이 가하여졌다.기장이 가하여졌다.

  • PDF