Abstract A desirability function approach has been widely used in multi‐response optimization due to its simplicity. Most of the existing function‐based methods assume that variability response variables is stable; thus, they focus mainly on mean multiple responses. However, this stable assumption often does not apply practical situations; quality product or process can be severely degraded high In regard, we propose a new method simultaneously optimize both and particular, proposed uses...

Dual response surface optimization simultaneously considers the mean and standard deviation of a response. The minimization squared error (MSE) is simple, yet effective, approach in dual optimization. bias variance components MSE need to be weighted properly if they are not same importance given problem situation. To date, relative weights have been equally set or determined only by data. However, should accordance with tradeoffs on various factors quality costs. In this paper, we propose...

In dual-response-surface optimization, the mean and standard deviation responses are often in conflict. To obtain a satisfactory compromise, Decision Maker (DM)'s preference information on trade-offs between should be incorporated into problem. most existing works, DM expresses subjective judgment through parameter before problem-solving process, after which single solution is obtained. This study proposes posterior articulation approach to optimization. The initially finds set of...

Abstract Dual response surface optimization considers the mean and variation simultaneously. The minimization of mean‐squared error (MSE) is an effective approach in dual optimization. Weighted MSE (WMSE) formed by imposing relative weights, (λ, 1−λ), on squared bias variance components MSE. To date, a few methods have been proposed for determining λ. resulting λ from these either single value or interval. This paper aims at developing systematic method to choose when interval given....

A common problem encountered in product or process design is the selection of optimal parameters that involves simultaneous consideration multi-response characteristics, called a surface (MRS) problem. There are several approaches proposed for MRS optimization (MRO), including priority-based approach, desirability function and loss approach. The existing MRO require all preference information decision maker be articulated prior to solving However, it difficult articulate advance. This paper...

To solve multiple response optimization problems that often involve incommensurate and conflicting responses, a robust interactive desirability function approach is proposed in this article. The consists of parameter initialization phase calculation decision-making phases. It considers decision maker's preference information regarding tradeoffs among responses the uncertainties associated with predicted surface models. method first to consider model uncertainty using an approach. allows...

The dual response surface optimization (DRSO) approach attempts to simultaneously optimize the mean and standard deviation of a variable. In DRSO, are often in conflict thus, it is extremely critical compromise two conflicting functions. Recently, posterior preference articulation for DRSO (P-DRSO) was proposed. P-DRSO generates set nondominated solutions allows decision-maker (DM) select most preferred solution. has an advantage that DM can better understand trade-offs between deviation....

Multiple response surface optimization (MRSO) aims to determine a setting of input variables that simultaneously optimizes multiple responses. The process capability function (PCF) approach is an attractive alternative MRSO because it considers both the mean and variance responses systematically through index. In MRSO, are often in conflict. Thus, preference information decision maker (DM) on tradeoffs among should be considered. However, existing PCF methods based unrealistic assumption all...

Multiple-response surface optimization (MRSO) aims to find a setting of the input variables that simultaneously optimizes multiple responses. The process capability approach incorporates decision maker’s (DM’s) preference information in problem-solving while considering dispersion effects responses systematically through framework index. However, existing methods are based on an unrealistic assumption all DM is given advance. Although few incorporate DM’s solving problem, they require too...

다중반응표면 최적화는 다수의 반응변수(품질특성치)를 최적화하는 입력변수의 조건을 찾는 것을 목적으로 한다. 최적화를 위해 제안된 가중평균제곱오차(Weighted Mean Squared Error, WMSE) 최소화법은 평균제곱오차의 구성요소인 제곱편차와 분산에 서로 다른 가중치를 부여하는 방법이다. 지금까지 WMSE 최소화법과 관련하여, 개별 반응변수의 WMSE를 구성한 후 이들의 가중합을 최소화하는 가중합 기반 최소화법이 제안되었다. 그러나 기반법은 목적함수 공간에서 볼록하지 않은 구간이 있고 이 구간에서 가장 선호되는 해가 존재할 경우 해를 찾아내지 못한다는 한계를 지니고 있다. 본 논문에서는 기존의 기반법의 한계점을 극복하기 위하여 Tchebycheff Metric 최소화법을 제안하고자 Multiresponse optimization (MRO) seeks to find the setting of input variables, which optimizes multiple...

Abstract One of the most important issues in multiple response surface optimization (MRSO) is obtaining a satisfactory “compromise” solution considering decision maker (DM)'s preference information on tradeoffs among responses. A promising alternative to incorporate DM's into problem posterior articulation approach, which first generates all (or most) nondominated solutions and then makes DM select best one from set posteriori. However, it has an inefficiency that excessive number almost are...

가장 대표적인 범용센서모델인 다항식비례모형(Rational Function Model)은 물리적 센서모형의 정확도에 견줄 수 있는 특성으로 인하여 상업용 위성영상의 센서모델링 기법에서 많이 쓰이고 있다. RPCs를 이용하여 인공위성 영상의 3차원 위치를 결정할 있지만, 대축척의 지형도 제작시 정확도 측면에서 한계를 가지고 본 연구에서는 QuickBird-2, 영상을 지상기준점의 수량, 분포 및 다항식비례모형의 차수에 따른 분석을 수행하였다. 그 결과 1:25,000 축척의 수평위치 표고 허용오차 범위에 포함 될 가능성을 확인하였다. The Rational Model has been used as a replacement sensor model in most commercial photogrammetric systems due to its capability of maintaining the accuracy physical models. Although satellite...

Multi-Response Surface Optimization aims at finding the optimal setting of input variables considering multiple responses simultaneously. The Weighted Mean Squared Error (WMSE) minimization approach, which imposes a different weight on two components mean squared error, bias and variance, first obtains WMSE for each response then minimizes all WMSEs once. Most methods proposed approach to date are classified into prior preference articulation requires that decision maker (DM) provides...