Design Of Experiment (DOE) Presentation _DWS

合集下载

Introduction to Design of Experiments (DOE)(英文版)(pdf 14页)

Introduction to Design of Experiments (DOE)(英文版)(pdf 14页)

Pat Hammett
Duality of Signal & Noise Factors
Effect of Input Variables on a Process Output Adjustment Factors affects the mean of the process without affecting variation “knob”. Need these to center a process. (especially if Pp is high, Ppk low) Example: tool position knob in machining
Signal Noise Duality Factors location & Factors (dispersion effects) dispersion (location effects)
Dispersion Effect?
Yes No Yes Duality Factor Adjustment Factor No Noise Factor Robust Factor
n
n
n
Use FIRM Approach (next slide)
n
Select levels or settings for each process input variable. Run the experiment at various combinations of levels. Analyze the data for significant main effects (input variable effects) and interaction effects.
Design of Experiments

Design of Experiment(课堂PPT)

Design of Experiment(课堂PPT)
8
Lab 用正交试验设计对设计变量进行筛选
➢ 设计变量12个,因素水平2。 ➢ 采用全析因试验设计需要4096次(212=4096)。 ➢ 采用正交试验设计只需要16次。
9
Lab
设计变量
WingArea Fuselength CruiseVel
WtFuel AR
CLmax sfc
PropEfficiency UltLoadFactor
基准值
135.0 20.0 200.0 139.2 5.6 1.92 1.816e-07 0.945 4.56 141.84 286.56 4.05
水平值
108.0 162.0 16.0 24.0 160.0 240.0 111.36 167.04 4.48 6.72 1.536 2.304 1.4528e-07 2.1792e-07 0.89775 0.99225 3.648 5.472 113.472 170.208 229.248 343.872 3.645 4.455
拉丁超立方抽样是一种修正的蒙特卡罗方法。它覆盖均匀,适用于影响因 素较多的情况,可显著减少试验规模。 特点:
➢ 试验点较均匀; ➢ 样本点是随机的,每次计算结果都不一样; ➢ 试验次数等于水平数; ➢ 试验次数可以是任何数值; ➢ 应用广泛的计算机仿真试验设计,覆盖均匀,适用于影响因素较多的情况,
可显著减少试验规模。
正交试验设计(Orthogonal Arrays)
按照一种已经拟定好的满足正交试验条件的表格(正交表)来安排试验 的试验设计方法。
正交表的形式为LA(pq)
L 代表正交表; A表示表中有A个横行,也即总共所需的试验次数; p表示因素的水平数; q表示因素的个数。
5

DOE试验设计范文

DOE试验设计范文

DOE试验设计范文
DOE(Design of Experiments)试验设计是一种科学的、系统性的方法,用于研究因果关系,找出影响实验结果的因素,并确定最优的因素组合。

通过DOE试验设计,可以减少试验次数,提高试验效率,准确地分析
因素对结果的影响程度,从而优化和改进产品、工艺或系统。

全因子设计是一种最基本的试验设计方法,它考虑了所有可能的因素
和它们的水平,以确定它们对结果的影响。

全因子设计通常包括因子的选取、水平的确定、试验方案的建立和结果的分析等步骤。

通过全因子设计,可以确定每个因素对结果的影响程度,推断最佳因素水平以及交互作用的
影响。

在进行DOE试验设计时,需要考虑以下几个关键要素:
1.确定实验目的:明确实验的研究目的和需要解决的问题,确定关键
的因素和响应变量。

2.选择适当的设计方法:根据实验目的和研究问题选择合适的试验设
计方法,如全因子设计或响应面设计。

3.设计试验方案:确定因子和水平的选取,建立试验方案,包括样本
数量、实验次数、随机化方法等。

4.进行实验:按照设计方案进行实验操作,并记录实验数据。

5.分析数据:利用统计方法对实验数据进行分析,建立数学模型,推
断因素对结果的影响程度和交互作用。

6.进行优化:根据分析结果,确定最佳的因素组合,优化实验结果。

DOE试验设计在品质改善、生产优化、产品创新等方面具有重要的应
用价值,能够帮助企业降低成本、提高效率,提高产品质量和市场竞争力。

因此,掌握和运用DOE试验设计方法是很有必要的,有助于实现科学的实
验研究和数据分析。

#试验设计(DOE)

#试验设计(DOE)
Titleist Avg. =
Pinnacle Avg. =
21
Graph >Boxplot Y - 距离 X - 球,球棒,風
Distance
260
250
240
230
220
210
200
190
-1
1
B a lls
260
250
240
230
220
210
200
190
-1
1
W inds
结论:
Distance
= 41- 45.5 = - 4.5
13
全因子实验示例1-曲奇饼DOE
交互作用 - 一个因子的影响取决于另一个因子的水平
烘烤时间 烘烤温度
(min) (oF)
口味
A
B
AB
6(-)
375 (-) +
41
10(+) 375 (-) -
50
6(-)
450 (+) -
47
10 (+) 450 (+) +
35
#2)对低水平球棒在高水平风速状态下所击出的所有距离重 复相同计算。
210+215= 212.5 2
#3)对高水平球棒在低水平风速状态下所击出的所有距离重 复相同计算。
220+215= 217.5 2
#4)对高水平球棒在高水平风速状态下所击出的所有距离重
复相同计算。
245+240= 242.5
2
相互作用图
A
B
6375 4110375 506
450 47
10
450 35
烘烤时间 烘烤温度

DOE 试验设计

DOE 试验设计

实验设计的目的可能包括: (1)确定哪些参数对响应的影响最大; (2)确定应把有影响的参数设定在什么水平,以使响应达 到或尽可能靠近希望值(On target); (3)确定应把有影响的参数设定在什么水平,以使响应的 分散度(或方差)尽可能减小. (4)确定应把有影响的参数设定在什么水平,以使不可控 参数(噪声参数)对响应的影响尽可能减小. 因此, 在制造过程的开发以及解决过程中出现的问题中 都可以应用实验设计,以改善过程的性能,或者使过程对 于外部波动源(干涉)不那么敏感,即得到一个"稳 健"(Robust)的过程,同时还可节省时间和降低成本. 所以,实验设计对于开发和改善制造过程,提高产品质量 是一个非常重要的工程工具.
3
析因实验的优点 1.与一次只改变一个参数的实验方法相比, 可以减少试验次数(24:8) 2可以观察参数间的相互作用 3.得到的结果适用范围更广——主效应和相 互作用是在各参数各种可能的组合的情况 下得到的,与实际情况较接近.
进行实验设计的步骤 进行实验设计有五个关键步骤: (1) 组成一个小组来设计实验,一般应该把若干有 关的人员组织起来,通过协作来共同进行一个实验 设计. (2) 规划实验: 明确实验的目的或目标 确定系统输出,也就是响应 确定响应的测量方法 通过集思广义(Brainstorming),并利用鱼刺图 (Fishbone diagram),来找出对响应可能有影响 的所有参数. 对上述参数进行分析,筛选,最后选定可能最 有影响的那些参数.将用实验检验它们.
应该指出,进行上述计算和模拟的前提是要能找到描述 影响参数与响应之间关系的工程方程(数学描述),否则便 无法进行上述计算和模拟.即使在这种情况下,也可以 应用实验的方法找到影响参数与响应之间的关系,达到 改进质量的目的. 所以可以说,为获得高质量的产品,进行必要的实验是不 可缺少的.而进行实验是需要付出代价的,往往代价较 高,需要花费较多的人力,物力和时间.所以,如何合 理设计实验,以便能以最小的代价获得尽可能多,而且 可靠的有关产品及其制造过程的知识,从而达到改进质 量的目的,是很重要的,也是很有学问的. 下面以一个实例来引出如何合理设计实验的问题.

DOE(Design of Experiment,试验设计)

DOE(Design of Experiment,试验设计)

DOE出自 MBA智库百科(/)DOE(Design of Experiment,试验设计)目录[隐藏]∙ 1 什么是DOE∙ 2 为什么需要DOE∙ 3 DOE的基本原理∙ 4 DOE实验的基本策略∙ 5 DOE的步骤∙ 6 DOE的作用∙7 DOE的方法[编辑]什么是DOEDOE(Design of Experiment)试验设计,一种安排实验和分析实验数据的数理统计方法;试验设计主要对试验进行合理安排,以较小的试验规模(试验次数)、较短的试验周期和较低的试验成本,获得理想的试验结果以及得出科学的结论。

试验设计源于1920年代研究育种的科学家Dr.Fisher的研究, Dr. Fisher 是大家一致公认的此方法策略的创始者, 但后续努力集其大成, 而使DOE在工业界得以普及且发扬光大者, 则非Dr. Taguchi (田口玄一博士) 莫属。

[编辑]为什么需要DOE∙要为原料选择最合理的配方时(原料及其含量);∙要对生产过程选择最合理的工艺参数时;∙要解决那些久经未决的“顽固”品质问题时;∙要缩短新产品之开发周期时;∙要提高现有产品的产量和质量时;∙要为新或现有生产设备或检测设备选择最合理的参数时等。

另一方面,过程通过数据表现出来的变异,实际上来源于二部分:一部分来源于过程本身的变异,一部分来源于测量过程中产生的变差,如何知道过程表现出来的变异有多接近过程本身真实的变异呢?这就需要进行MSA测量系统分析。

[编辑]DOE的基本原理试验设计的三个基本原理是重复,随机化,以及区组化。

所谓重复,意思是基本试验的重复进行。

重复有两条重要的性质。

第一,允许试验者得到试验误差的一个估计量。

这个误差的估计量成为确定数据的观察差是否是统计上的试验差的基本度量单位。

第二,如果样本均值用作为试验中一个因素的效应的估计量,则重复允许试验者求得这一效应的更为精确的估计量。

如s2是数据的方差,而有n次重复,则样本均值的方差是。

DOE-田口式

DOE-田口式
19



何謂直交表

直交表之符號意義: L代表L型直交表,9代表需要做9次實驗,3代表3 水準,4代表可以擺放4個要因。 L 3
4 9

常用之直交表種類:
L4 2 3 , L8 2 7 , L16 2 15 , L32 2 31






L9 34 , L12 18
L 2 , L 2 3 , L 2
36
3
313

20
代表性直交表暨點線圖(L9)
1 2 3 4 5 6 7 8 9 A 1 1 1 2 2 2 3 3 3 B 1 2 3 1 2 3 1 2 3 C 1 2 3 2 3 1 3 1 2 D 1 2 3 3 1 2 2 3 1 Y Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y8 Y9
11
定義名詞及其重點
特性值分類
要因分類
直交表及變形
直積配置 ANOVA分析
水準及水準值
12
特性值Output分類

以數值形式作分類:

計數值:量測數值不為連續量,一般用“個”代表。


單純計數值:將一個特性區分為良品或不良品,常用在外觀 等,例如:不良個數、故障台數.... 多重計數值:將一個特性區分為優、良、中、可、劣,例如: 外觀可分為好、有一些瑕疵、有很多瑕疵。 單一目標之特性。Ex:某一規定的尺寸或電壓或顏色.... 多重目標之特性,依據不同的需求,只要改變某一要因即可 達成不同產品。Ex:經由三原色加入量的不同即可做出不同的 顏色,此時對顏色而言是有無限多的目標。
1 2 3 4 5 6 7 8 9 A 1 1 1 2 2 2 1 1 1 B 1 2 3 1 2 3 1 2 3 C 1 2 3 2 3 1 3 1 2 D 1 2 3 3 1 2 2 3 1 Y Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y8 Y9

DOE详细讲解ppt课件

DOE详细讲解ppt课件
第36页
实现最优化
第37页
实现最优化
第38页
结果验证
第39页
实现最优化
第40页
部分因子实验
第41页
部分因子实验
第42页
第43页
第44页
第45页
第10页
利用Minitab中设计DOE
实验案例
改进热处理工艺提高钢板断裂强度问题。合金钢板经 热处理后将提高其抗断裂性能,但工艺参数的选择是个 复杂问题。影响断裂强度有4个因子,确认哪些因子影响 确实是显著地,进而确定出最佳的工艺条件,这几个因 子及实验水平如下:
A:加热温度,低水平:820,高水平:860(摄氏度) B: 加热时间,低水平:2,高水平:3(分钟) C:转换时间,低水平:1.4,高水平:1.6(分钟) D:保温时间,低水平:50,高水平:60(分钟)
第4页
6) 分析实验结果:将实验结果进行方差分析,确定实验因 子的重要性及各因子对实验结果的影响程度。
7) 重复性实验:将重要因子或影响实验的主要因素进行评 估,重新进行DOE实验,以确定其实验的真实性。
8) 作出结论:对实验结果进行分析后作出结论。
第5页
名词介绍
• 1、试验因素 试验因素指当试验条件变化,试验考核指标也发生变化
考核指标可以是一个,也可以是多个。前者称为单 指标试验设计,后者称为多指标试验设计。在多指标试 验设计中,一般根据指标的重要程度予以加权,确定为 一个综合性考核指标,以便进行计算。
第8页
4、完全因素位级组合
完全因素位级组合指参与试验的全部因素与全部位 级相互之间的全部组合次数,即全部的试验次数。
C1
实验设计(DOE)
• Design of Experiment • 为什么要进行试验设计?

DOE(Design of Experiment,试验设计)

DOE(Design of Experiment,试验设计)

DOE出自 MBA智库百科(/)DOE(Design of Experiment,试验设计)目录[隐藏]∙ 1 什么是DOE∙ 2 为什么需要DOE∙ 3 DOE的基本原理∙ 4 DOE实验的基本策略∙ 5 DOE的步骤∙ 6 DOE的作用∙7 DOE的方法[编辑]什么是DOEDOE(Design of Experiment)试验设计,一种安排实验和分析实验数据的数理统计方法;试验设计主要对试验进行合理安排,以较小的试验规模(试验次数)、较短的试验周期和较低的试验成本,获得理想的试验结果以及得出科学的结论。

试验设计源于1920年代研究育种的科学家Dr.Fisher的研究, Dr. Fisher 是大家一致公认的此方法策略的创始者, 但后续努力集其大成, 而使DOE在工业界得以普及且发扬光大者, 则非Dr. Taguchi (田口玄一博士) 莫属。

[编辑]为什么需要DOE∙要为原料选择最合理的配方时(原料及其含量);∙要对生产过程选择最合理的工艺参数时;∙要解决那些久经未决的“顽固”品质问题时;∙要缩短新产品之开发周期时;∙要提高现有产品的产量和质量时;∙要为新或现有生产设备或检测设备选择最合理的参数时等。

另一方面,过程通过数据表现出来的变异,实际上来源于二部分:一部分来源于过程本身的变异,一部分来源于测量过程中产生的变差,如何知道过程表现出来的变异有多接近过程本身真实的变异呢?这就需要进行MSA测量系统分析。

[编辑]DOE的基本原理试验设计的三个基本原理是重复,随机化,以及区组化。

所谓重复,意思是基本试验的重复进行。

重复有两条重要的性质。

第一,允许试验者得到试验误差的一个估计量。

这个误差的估计量成为确定数据的观察差是否是统计上的试验差的基本度量单位。

第二,如果样本均值用作为试验中一个因素的效应的估计量,则重复允许试验者求得这一效应的更为精确的估计量。

如s2是数据的方差,而有n次重复,则样本均值的方差是。

试验设计(Design of Experiments)简介53页PPT

试验设计(Design of  Experiments)简介53页PPT

试验设计(Design of Experiments)简介
41、实际上,我们想要的不是针对犯 罪的法 律,而 是针对 疯狂的 法律。 ——马 克·吐温 42、法律的力量应当跟随着公民,就 像影子 跟随着 身体 步。— —杰弗 逊 44、人类受制于法律,法律受制于情 理。— —托·富 勒
45、法律的制定是为了保证每一个人 自由发 挥自己 的才能 ,而不 是为了 束缚他 的才能 。—— 罗伯斯 庇尔
31、只有永远躺在泥坑里的人,才不会再掉进坑里。——黑格尔 32、希望的灯一旦熄灭,生活刹那间变成了一片黑暗。——普列姆昌德 33、希望是人生的乳母。——科策布 34、形成天才的决定因素应该是勤奋。——郭沫若 35、学到很多东西的诀窍,就是一下子不要学很多。——洛克

Design of Experiments

Design of Experiments

Design of ExperimentsDesign of experiments (DOE) is a statistical methodology that is used to determine the effect of various factors on a process or product. The objective of DOE is to identify the factors that have the greatest impact on the outcome of a process or product and to optimize these factors to achieve the desired outcome. DOE is widely used in manufacturing, engineering, and scientific research to improve the quality and efficiency of processes and products. In this essay, wewill discuss the importance of DOE, its applications, and the steps involved in designing an effective experiment. The importance of DOE cannot be overstated. In today's competitive business environment, it is essential to optimize processesand products to remain competitive. DOE provides a systematic approach to identify the critical factors that affect the quality and efficiency of processes and products. By optimizing these factors, organizations can improve their performance, reduce costs, and increase customer satisfaction. DOE is also useful inidentifying the sources of variation in a process or product and eliminating them to improve quality. DOE has numerous applications in manufacturing, engineering, and scientific research. In manufacturing, DOE is used to optimize production processes, reduce defects, and improve product quality. In engineering, DOE isused to design and optimize products, reduce costs, and improve performance. In scientific research, DOE is used to identify the critical factors that affect the outcome of experiments and to optimize these factors to achieve the desired outcome. The first step in designing an effective experiment is to define the problem or objective. The problem or objective should be clearly stated and should include the desired outcome. The next step is to identify the factors that may affect the outcome of the experiment. These factors are called independent variables or factors. The factors should be identified based on prior knowledge or experience, and should be relevant to the problem or objective. Once the factors have been identified, the next step is to determine the levels of each factor. The levels represent the range of values that the factor can take. The levels shouldbe selected based on prior knowledge or experience, and should be relevant to the problem or objective. The number of levels for each factor depends on the complexity of the problem and the resources available. The next step is to designthe experiment. The design of the experiment should be based on the objectives of the experiment and the resources available. There are several types ofexperimental designs, including full factorial, fractional factorial, and response surface designs. The choice of design depends on the complexity of the problem and the resources available. After the experiment has been designed, the next step is to conduct the experiment. The experiment should be conducted in a controlled environment to minimize the effects of extraneous variables. The data should be collected and recorded in a systematic manner to ensure accuracy and reliability. The data should be analyzed using statistical methods to determine the effect of each factor on the outcome of the experiment. The final step is to draw conclusions and make recommendations based on the results of the experiment. The conclusions should be based on statistical analysis and should be relevant to the problem or objective. The recommendations should be practical and feasible, and should be based on the resources available. In conclusion, DOE is a powerful methodology that can be used to optimize processes and products. It is widely used in manufacturing, engineering, and scientific research to improve quality, reduce costs, and increase customer satisfaction. The steps involved in designing an effective experiment include defining the problem or objective, identifying the factors that may affect the outcome of the experiment, determining the levels of each factor, designing the experiment, conducting the experiment, analyzing the data, and drawing conclusions and making recommendations. By following these steps, organizations can optimize their processes and products and remain competitive in today's business environment.。

Design of Experiments

Design of Experiments

Design of ExperimentsDesign of Experiments (DOE) is a systematic and efficient approach to planning, conducting, and analyzing experiments to understand the relationship between input factors and output responses. It is a crucial tool in various fields such as engineering, manufacturing, pharmaceuticals, and agriculture. By systematically varying the input factors and observing the corresponding changes in the output responses, researchers can gain valuable insights into the underlying processesand optimize the performance of the system. One of the key benefits of DOE is its ability to provide a comprehensive understanding of the factors that influence a particular process or system. By carefully designing the experiments to systematically vary the input factors while keeping other variables constant, researchers can identify the most significant factors that affect the output responses. This allows for the development of predictive models that can be usedto optimize the system and improve its performance. Additionally, the systematic approach of DOE helps in minimizing the number of experiments required, thereby saving time and resources. Another important aspect of DOE is its ability to identify interactions between the input factors. In many real-world systems, the input factors do not act independently, and their combined effects can have a significant impact on the output responses. By using factorial designs and other DOE techniques, researchers can systematically investigate the interactions between the input factors and gain a deeper understanding of how they collectively influence the system. This knowledge is invaluable for making informed decisions about process optimization and product development. Furthermore, DOE enables researchers to quantify the variability in the system and assess the robustness of the process. By conducting experiments at different levels of the input factors, researchers can estimate the effects of variability and identify the optimal operating conditions that minimize the impact of variability on the output responses. This is particularly important in manufacturing and quality control, where minimizing variability is essential for ensuring consistent product quality. In addition to its technical benefits, DOE also fosters a culture of innovationand continuous improvement within organizations. By systematically exploring the relationships between input factors and output responses, researchers areencouraged to think critically and creatively about process optimization. This can lead to the development of new and improved products, processes, and technologies that drive organizational growth and competitiveness in the market. Despite its numerous advantages, implementing DOE can be challenging for organizations, especially those that lack the necessary expertise and resources. Designing and conducting experiments requires a deep understanding of statistical principles and experimental design, as well as access to specialized software and equipment. Additionally, the process of analyzing and interpreting the experimental data can be complex and time-consuming, requiring the skills of experienced statisticians and researchers. Moreover, the success of DOE depends heavily on the careful selection of input factors and the design of the experiments. If the input factors are not chosen appropriately or if the experiments are poorly designed, theresults may not provide meaningful insights into the system. This highlights the importance of proper training and education in the principles of experimental design and statistical analysis for researchers and engineers involved in DOE. In conclusion, Design of Experiments is a powerful and versatile tool that offers a systematic approach to understanding and optimizing complex systems. Its ability to identify significant factors, interactions, and variability, as well as itsrole in fostering innovation and continuous improvement, makes it an invaluable asset for organizations across various industries. However, successful implementation of DOE requires a deep understanding of statistical principles, experimental design, and data analysis, as well as the necessary resources and expertise. By overcoming these challenges and embracing the principles of DOE, organizations can gain a competitive edge and drive innovation in their respective fields.。

  1. 1、下载文档前请自行甄别文档内容的完整性,平台不提供额外的编辑、内容补充、找答案等附加服务。
  2. 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
  3. 3、如文档侵犯您的权益,请联系客服反馈,我们会尽快为您处理(人工客服工作时间:9:00-18:30)。
The Disadvantage and Advantage is ✓ Time ? ✓ Cost ? ✓ Optimize ?
Sika China
What‘s the Block for DOE?
„Traditional“- One Factoe At a Time(OFAT) Experimentation
Different Operator
Different Machine
Different Raw Material
X1
X2
Proce
X3
ss
Input
X4
Xk
Process
Y1
Y2
Y3
Proce
ss

Output
Ym
Uncontrollable Factors Continuous
OFAT Example(Target is Elasticity maximize) ✓ Fix the conc.A as 16, select conc.B from 2.40, 2.55, 2.70, 2.85, 3.00 ✓ Fix the conc. B is 3.00 when elasticity is maximize of step 1 ✓ Select Cont A from 15,16,17,18,19,20,21, ✓ The conc. A is 15 when elasticity is maximize of step 3 ✓ So we got the conclusion is conc. A is 15 & conc. B is 3.00 ✓ But the output from the chart should be conc. A is 21 & conc. B is 2.25 ✓ Why the difference?
✓ Analysis of Unstructured Data ✓ The Best Educated Guess Method ✓ One Factoe At a Time(OFAT) Experimentation
Sika China
What‘s the Block for DOE?
„Traditional“- Analysis of Unstuctured Data
Sika China
DOE (Design of Experiments)
Sika China
Sika China
Sika China
What‘s the Block for DOE?
Today, three „Traditional“ approaches to experimentation are still in use:
Objective is often to achieve an optimim of one or more of the responses ✓ Vary one factor from low to high ✓ Record and Fix this factor at the optimum setting ✓ Process is repeated for each controllable factor ✓ Result provides the apparent optimum result
Sika China
Room Temperature Barometric Pressure
Relative Humidity
Basic Definitions of Experimentation
A Simple Process Model
Uncontrollable Factors Categorical
Sika China
What‘s the Block for DOE?
„Traditional“- The Best Educated Guess Method
Based on Enough experience and knowledge of underlying processes and the result of previous experiments to provide the basis for selecting the coditions for the next tri
Sika China
What‘s the Block for DOE?
OFAT Conclusion
✓ Fail to find the true optimum setting of input factors ✓ Fails to identify interactions ✓ Is inefficient relative to designed experiments ✓ Is sensitive to baseline (Starting) condition
What‘s the Block for DOE?
Sika China Conc. B
OFAT Example(Target is Elasticity maximize)
Contour Plot of Elasticity vs Conc. B, Conc. A
16 3.00
2.85
2.70
2.55
2.40 15 16 17 18 19 20 21 Conc. A
E lastic ity < 35.0
35.0 - 37.5 37.5 - 40.0 40.0 - 42.5 42.5 - 45.0 45.0 - 47.5 47.5 - 50.0
> 50.0
Sika China
What‘s the Block for DOE?
Valid for: Experimental units-Obtained through a valid sampling plan Experimental factors-measurable but uncontrolled Objective is a study relationship between response and uncontrolled factors
相关文档
最新文档