CAFE 實驗計劃法軟體-研究分析軟體/新永資訊有限公司

CAFE 實驗計劃法軟體

CAFE 實驗計劃法軟體

  • CAFE 實驗計劃法軟體
  • 編號
  • 類別
    研究分析軟體
  • 介紹
    實驗計劃法軟體CAFE (Computer-Aided Formula Environment)為一套實驗設計軟體,可靠的模型驗證方式 :傳統的DOE或田口方法不區分樣本內、樣本外實驗數據之作法,其誤差被嚴重低估。CAFE採用交叉驗證法,可以在不需要增加實驗數目下,達到區分樣本內、樣本外的效果,使誤差能被準確估計。
  • 價格

CAFE Experimental Planning Method Software

CAFE (Computer-Aided Formula Environment) is a suite of experimental design
CAFE (Computer-Aided Formula Environment) is a software package for experimental design, its user-friendliness and Chinese windows will provide users with a quick and convenient way to
*Model construction: Data sets with experimental factors
 (independent variables) and quality factors (dependent
  variables).
*Modeling the relationship between experimental factors
 (independent variables) and quality factors (dependent
  variables) by statistical methods.
*Model analysis: Analyze the relationship between experimental
  factors (independent variables) and quality factors (dependent
  variables).

*Parameter optimization: Find a model that satisfies specific
 cons traints (experimental factor, quality factor composition)
 and optimizes
 Quality improvement solution (combination of experimental
  factors) to meet specific objectives (composition of
  experimental factors and quality factors, e.g., cost, yield, etc.).
 CAFE has the following advantages over traditional DOE
  software or Taguchi method software:
      •  Reliable model validation: Traditional DOE or
          Taguchi methods do not distinguish between
          in-sample and out-of-sample experimental data.
          CAFE uses cross-validation, which can be performed
          without the need for a sample.CAFE adopts
          cross-validation method, which can achieve the effect of
          distinguishing between in-sample and out-of-sample
          experimental data without increasing the number of
          experiments.CAFE uses cross-validation to distinguish
          between in-sample and out-of-sample experimental
          data without increasing the number of experiments,
          so that the error can be accurately estimated.

      •  Accurate model building capabilities: Traditional DOE
          uses regression analysis.Traditional Taguchi method uses
          effect analysis CAFE uses neural-like networks in artificial
          in telligence.With the same cross-validation method,
          neural-like networks are far more accurate than
          regression analysis and Effectiveness analysis is more
          accurate.
      •  Practical quality improvement model: The traditional
          DOE mixes quality constraints and objectives.
          The traditional DOE method can only deal with a single
          quality objective.
          The traditional Taguchi method can only deal with a
          single quality objective, but in practice, process
          performance and product quality often have many
          limitations(such as cost and some quality characteristics
          up per limit, defect rate and some quality characteristics
          lower limit) and objectives CAFE uses the following
          quality improvement model to solve these problems.
          CAFE uses the following quality improvement models
          to solve these problems.

CAFE has several application examples,
      including:
  •     • Taguchi experiment or factor experiment
  •      • Wire bonding process for copper
  •         conductor wafer packaging
  •         • Optimization of CNC tooling machine
  •           performance
  •       • Optimization of the mold adhesion of /C packag
  •       • Optical sheet molding quality and injection
  •          molding process
  •      • Experiment on reaction surface method
  •      • Enzymatic synthesis of hexyl esters for
  •        optimal adaptation
  •      • Optimization of hydrofluoric acid reaction
  •         kiln operation
  •    • Mixture Design Experiment
  •        • High performance concrete
  •           proportioning design

CAFE實驗計劃法軟體

CAFE (Computer-Aided Formula Environment)為一套實驗設計
軟體,其友善性與中文視窗將可提供使用者快速且便利地
*模型建構:用實驗因子(自變數)與品質因子(因變數)的數據集,
​​​​ ​​​用統計方法建立實驗因子(自變數)與品質因子(因變數)之間的關係
​​​ ​​​​之模型。
*模型分析:分析實驗因子(自變數)與品質因子(因變數)之間的關係。
*參數優化:尋找能滿足特定限制(實驗因子、品質因子組成),並最佳化
 特定目標(實驗因子、品質因子組成,例如成本、良率…)的品質改善方案(實驗因子組合)。
CAFE與傳統的實驗計劃法(DOE)軟體或田口方法軟體相較,有下列優點:
      •可靠的模型驗證方式 :
        傳統的DOE或田口方法不區分樣本內、樣本外實驗數據之作法,
        其誤差被嚴重低估。CAFE採用交叉驗證法,可以在不需要增加
​​​​        ​​​實驗數目下,達到區分樣本內、樣本外的效果,使誤差能被準確
​        ​​​​​​估計。
      •精確的模型建構能力:
        傳統的DOE使用迴歸分析,傳統的田口方法使用效果分析CAFE
        使用人工智慧中的類神經網路。在同樣使用交叉驗證法下, 
        類神經網路遠比迴歸分析、效果分析更準確。
      •實用的品質改展模式:傳統的DOE將品質上的限制與目標混為一談,
        採用不科學的方法整合限制與目標,傳統的田口方法只能處理
        單一品質目標,但實務上,製程效能、產品品質經常有許多限制
       (如成本及部份品質特性上限,不良率及部份品質特性下限)與目標
       (如成本最小化,良率、產率最大化)。
        CAFE採用下列品質改展模式,可以解決上述問題:

CAFE有多個應用實例,包括:
 • 田口實驗或因子實驗:
  •   • 銅導線晶圓封裝之銲線製程
  •    • CNC工具機加工性能之最適化
  •    • IC封裝黏模力之最適化
  •    • 光板成型品質與射出成型製程
  •  • 反應曲面法實驗
  •   • 酵素合成己醇酯類之最適化
  •    • 氫氟酸反應窯操作之最適化
  •  • 配方設計(Mixture Design)實驗
  •   • 高性能混凝土配比設計

DADiSP 6.7 (pronounced day-disp) 數據處理分析軟體

DADiSP(發音為day-disp)是一個交互式圖形工作表-一種面向視覺的軟件包,用於顯示,管理,分析和表示 科學和技術數據。如果您收集,處理,編輯,減少,轉換,顯示或分析數據,DADiSP可以處理和簡化您的數據需求。簡而言之,如果您使用的是技術數據,DADiSP非常適合您。

DADiSP 6.7  (pronounced day-disp) 數據處理分析軟體

XpertRule Knowledge Builder 專家系統建構軟體

Knowledge Builder是一個用於開發和部署基於知識的應用程序和組件的企業實力環境。基於知識的應用程序是包含規則、專業知識、技術訣竅、程序、政策和法規的軟件組件,這些可以統稱為 "業務規則"。

XpertRule Knowledge Builder 專家系統建構軟體

HB-Reg v4.2.2303 貝斯階層回歸軟體

什麼是分層貝斯階層回歸(HB-Reg)?在市場研究數據的分析中,有很多時候研究人員有受訪者,商店或其他實驗單位的樣本,並希望估計每個單位的單獨回歸係數。在過去幾年中,分層貝葉斯(HB)估計在市場營銷研究中發揮著越來越重要的作用。它可以在精度和有效性方面改進參數估計(例如β權重和效用)。我們的HB-Reg軟件用於基於回歸的問題,其中受訪者提供包含連續因變量的多個觀察(病例)。

HB-Reg v4.2.2303 貝斯階層回歸軟體