JP5020728B2 - Quality control device, method and program - Google Patents

Quality control device, method and program

Info

Publication number
JP5020728B2
JP5020728B2 JP2007181206A JP2007181206A JP5020728B2 JP 5020728 B2 JP5020728 B2 JP 5020728B2 JP 2007181206 A JP2007181206 A JP 2007181206A JP 2007181206 A JP2007181206 A JP 2007181206A JP 5020728 B2 JP5020728 B2 JP 5020728B2
Authority
JP
Japan
Prior art keywords
quality
improvement
degree
range
factor
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
JP2007181206A
Other languages
Japanese (ja)
Other versions
JP2009020600A (en
Inventor
淳治 伊勢
潔 和嶋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nippon Steel Corp
Original Assignee
Nippon Steel Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nippon Steel Corp filed Critical Nippon Steel Corp
Priority to JP2007181206A priority Critical patent/JP5020728B2/en
Publication of JP2009020600A publication Critical patent/JP2009020600A/en
Application granted granted Critical
Publication of JP5020728B2 publication Critical patent/JP5020728B2/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Description

本発明は、鉄鋼プロセス等の製造プロセスの工程を経て製造される製品であって、且つ、複数の品質を満足する必要のある製品の品質を所望の範囲内に管理するのに好適な技術に関する。   The present invention relates to a technique that is suitable for managing the quality of a product that is manufactured through a manufacturing process such as a steel process and that needs to satisfy a plurality of qualities within a desired range. .

製品の品質に関して、最近の顧客からの高度な品質の要求のため、従来さほど重要視されなかった品質も大きな品質トラブルとして扱われるようになってきている。そのため、従来から管理されていた品質に、新たに管理すべき品質が加わることがある。   With regard to product quality, due to the recent demand for high quality from customers, quality that has not been emphasized so far has been treated as a major quality problem. For this reason, the quality to be newly managed may be added to the quality that has been conventionally managed.

従来の品質管理としては、主成分分析を用いたマハラノビス距離による管理を行う統計的品質管理が行われている。しかしながら、従来の品質管理では、迅速に所望の品質を確保できる製品の製造条件を見いだせる技術はなく、十分な品質管理が不可能であった。   As conventional quality control, statistical quality control is performed in which management is performed by Mahalanobis distance using principal component analysis. However, in the conventional quality control, there is no technique for finding the production conditions of a product that can ensure a desired quality quickly, and sufficient quality control is impossible.

これに対し、特許文献１には、散布図や相関係数では捉えられないプロセス制御データと品質データとの相関を解析できるようにする手法が開示されている。   On the other hand, Patent Document 1 discloses a technique that enables analysis of the correlation between process control data and quality data that cannot be captured by a scatter diagram or a correlation coefficient.

特開２００２−１８３２５０号公報JP 2002-183250 A

しかしながら、特許文献１には、品質管理を行う方法については開示されていない。また、従来から管理されていた品質を損なわずに、新たに管理すべき品質を所望の品質に管理し製造することは容易ではなかった。   However, Patent Document 1 does not disclose a method for performing quality control. Moreover, it is not easy to manage and manufacture the quality to be newly managed to a desired quality without deteriorating the quality that has been conventionally managed.

本発明は上記のような問題に鑑みてなされたものであり、製造プロセスの工程を経て製造される製品について複数の品質を所望の範囲内に管理できるようにすることを目的とする。   The present invention has been made in view of the above problems, and an object of the present invention is to make it possible to manage a plurality of qualities within a desired range for a product manufactured through a manufacturing process.

本発明の品質管理装置は、製造プロセスの工程を経て製造される製品について複数の品質を所望の範囲内に管理するための品質管理装置であって、前記製造プロセスから、操業因子の操業条件、及び、該操業因子の操業条件で得られた製品の品質を含むプロセスデータを収集するデータ入力手段と、前記データ入力手段により入力されるプロセスデータを蓄積するデータ蓄積手段と、前記データ蓄積手段に蓄積されたプロセスデータから、所定の品質が所望の範囲内となる操業因子の操業条件の範囲を算出する操業範囲算出手段と、前記データ蓄積手段から取得された、所定の品質を満足する製品である解析対象製品のプロセスデータに基づいて、操業因子と品質との関係式を確率分布相関解析法により求める確率分布相関解析手段と、前記操業範囲算出手段により算出された操業因子の操業条件の範囲と、前記確率分布相関解析手段により求められた関係式とを用いて、操業因子の品質に対する影響度を算出する影響度算出手段と、前記データ蓄積手段から、品質を改善したい製品である品質改善対象製品の操業因子の操業条件を取得する改善対象操業条件設定手段と、前記確率分布相関解析手段により求められた関係式と、前記改善対象操業条件設定手段により取得された品質改善対象製品の操業因子の操業条件とを用いて、品質の改善度合いを算出する品質改善度算出手段と、前記影響度算出手段により算出された影響度と、前記品質改善度算出手段により算出された品質の改善度合いとを用いて、操業因子の改善期待度を算出する改善期待度算出手段とを備えたことを特徴とする。
本発明の品質管理方法は、製造プロセスの工程を経て製造される製品について複数の品質を所望の範囲内に管理するための品質管理方法であって、前記製造プロセスから、操業因子の操業条件、及び、該操業因子の操業条件で得られた製品の品質を含むプロセスデータを収集し、データ蓄積手段に蓄積する手順と、前記データ蓄積手段に蓄積されたプロセスデータから、所定の品質が所望の範囲内となる操業因子の操業条件の範囲を算出する操業範囲算出手順と、前記データ蓄積手段から取得された、所定の品質を満足する製品である解析対象製品のプロセスデータに基づいて、操業因子と品質との関係式を確率分布相関解析法により求める確率分布相関解析手順と、前記操業範囲算出手順により算出された操業因子の操業条件の範囲と、前記確率分布相関解析手順により求められた関係式とを用いて、操業因子の品質に対する影響度を算出する影響度算出手順と、前記データ蓄積手段から、品質を改善したい製品である品質改善対象製品の操業因子の操業条件を取得する改善対象操業条件設定手順と、前記確率分布相関解析手順により求められた関係式と、前記改善対象操業条件設定手順により取得された品質改善対象製品の操業因子の操業条件とを用いて、品質の改善度合いを算出する品質改善度算出手順と、前記影響度算出手順により算出された影響度と、前記品質改善度算出手順により算出された品質の改善度合いとを用いて、操業因子の改善期待度を算出する改善期待度算出手順とを有することを特徴とする。
本発明のプログラムは、製造プロセスの工程を経て製造される製品について複数の品質を所望の範囲内に管理するための品質管理装置としてコンピュータを機能させるためのプログラムであって、前記製造プロセスから、操業因子の操業条件、及び、該操業因子の操業条件で得られた製品の品質を含むプロセスデータを収集するデータ入力手段と、前記データ入力手段により入力されるプロセスデータを蓄積するデータ蓄積手段と、前記データ蓄積手段に蓄積されたプロセスデータから、所定の品質が所望の範囲内となる操業因子の操業条件の範囲を算出する操業範囲算出手段と、前記データ蓄積手段から取得された、所定の品質を満足する製品である解析対象製品のプロセスデータに基づいて、操業因子と品質との関係式を確率分布相関解析法により求める確率分布相関解析手段と、前記操業範囲算出手段により算出された操業因子の操業条件の範囲と、前記確率分布相関解析手段により求められた関係式とを用いて、操業因子の品質に対する影響度を算出する影響度算出手段と、前記データ蓄積手段から、品質を改善したい製品である品質改善対象製品の操業因子の操業条件を取得する改善対象操業条件設定手段と、前記確率分布相関解析手段により求められた関係式と、前記改善対象操業条件設定手段により取得された品質改善対象製品の操業因子の操業条件とを用いて、品質の改善度合いを算出する品質改善度算出手段と、前記影響度算出手段により算出された影響度と、前記品質改善度算出手段により算出された品質の改善度合いとを用いて、操業因子の改善期待度を算出する改善期待度算出手段としてコンピュータを機能させることを特徴とする。 The quality control device of the present invention is a quality control device for managing a plurality of qualities within a desired range for a product manufactured through the steps of the manufacturing process, from the manufacturing process, operating conditions of operating factors, And data input means for collecting process data including product quality obtained under the operating conditions of the operating factors, data storage means for storing process data input by the data input means, and data storage means An operation range calculation means for calculating the range of operation conditions of the operation factor within which the predetermined quality falls within a desired range from the accumulated process data, and a product satisfying the predetermined quality acquired from the data storage means. based on the process data of a certain analysis products, and the probability distribution correlation analyzing means for determining the probability distribution correlation analysis method relationship between operational factors and quality, the Using the range of operating conditions of the operating factor calculated by the operating range calculating means and the relational expression obtained by the probability distribution correlation analyzing means, an influence calculating means for calculating the influence on the quality of the operating factor; The improvement target operation condition setting means for acquiring the operation condition of the operation factor of the quality improvement target product which is a product whose quality is to be improved from the data storage means , the relational expression obtained by the probability distribution correlation analysis means, and the improvement Quality improvement degree calculating means for calculating the degree of quality improvement using the operation conditions of the operation factors of the quality improvement target product acquired by the target operation condition setting means, and the degree of influence calculated by the influence degree calculating means And an improvement expectation degree calculation means for calculating an improvement expectation degree of the operation factor using the improvement degree of the quality calculated by the quality improvement degree calculation means. To.
The quality control method of the present invention is a quality control method for managing a plurality of qualities within a desired range for a product manufactured through the steps of the manufacturing process, from the manufacturing process, operating conditions of operating factors, And a process for collecting the process data including the quality of the product obtained under the operation condition of the operation factor and storing the process data in the data storage means, and the process data stored in the data storage means, a predetermined quality is desired. Based on the operating range calculation procedure for calculating the operating condition range of the operating factor that falls within the range, and the process data of the analysis target product that is a product that satisfies the predetermined quality obtained from the data storage means, the operating factor A probability distribution correlation analysis procedure for obtaining a relational expression between quality and quality by the probability distribution correlation analysis method, a range of operating conditions of the operating factor calculated by the operating range calculation procedure, By using the relational expression obtained by probability distribution correlation analysis procedure, the influence degree calculation procedure for calculating a degree of influence on the quality of operation factors, from the data storage means, of the quality improvement products are products that you want to improve the quality Operation target condition setting procedure for obtaining the operation condition of the operation factor, relational expression obtained by the probability distribution correlation analysis procedure, and operation of the operation factor of the quality improvement target product obtained by the target operation condition setting procedure A quality improvement degree calculation procedure for calculating a quality improvement degree using a condition, an influence degree calculated by the influence degree calculation procedure, and a quality improvement degree calculated by the quality improvement degree calculation procedure. And an improvement expectation degree calculation procedure for calculating an improvement expectation degree of the operation factor.
A program of the present invention is a program for causing a computer to function as a quality management device for managing a plurality of qualities within a desired range for a product manufactured through a manufacturing process, and from the manufacturing process, Data input means for collecting process data including operation conditions of the operation factors and product quality obtained under the operation conditions of the operation factors, and data storage means for storing process data input by the data input means , from the stored process data in the data storage means, and operating range calculation means for calculating the range of the operating conditions of operating factors predetermined quality is within a desired range, obtained from the data storage means, predetermined based on process data analysis products are products to satisfy the quality, probability distributions correlate relationship between operational factors and quality solutions Using the probability distribution correlation analysis means obtained by the method, the range of operation conditions of the operation factor calculated by the operation range calculation means, and the relational expression obtained by the probability distribution correlation analysis means, the quality of the operation factor An influence degree calculating means for calculating an influence degree, an improvement target operation condition setting means for obtaining an operation condition of an operation factor of a quality improvement target product which is a product whose quality is to be improved, and the probability distribution correlation analysis; Quality improvement degree calculating means for calculating the degree of quality improvement using the relational expression obtained by the means and the operating conditions of the operation factor of the quality improvement target product acquired by the improvement target operating condition setting means, The improvement expectation degree of the operation factor is calculated using the influence degree calculated by the influence degree calculating means and the quality improvement degree calculated by the quality improvement degree calculating means. Causing a computer to function as improved expectation calculation means for characterized.

本発明によれば、製造プロセスの工程を経て製造される製品について複数の品質を所望の範囲内に管理することができる。   ADVANTAGE OF THE INVENTION According to this invention, several quality can be managed within a desired range about the product manufactured through the process of a manufacturing process.

以下、添付図面を参照して、本発明の好適な実施形態について説明する。
図１は、本発明の実施形態に係る品質管理装置の概略構成を示す図である。また、図２は、本発明の実施形態に係る品質管理装置において実行される品質管理方法を説明するためのフローチャートである。以下、図１及び図２を参照して、鉄鋼製品の製造工場において、製品として「薄板コイル」、品質として「材質」及び「表面疵発生量」を対象とし、「材質」を満足しつつ「表面疵発生量」を低減させるように管理する例を説明する。 Preferred embodiments of the present invention will be described below with reference to the accompanying drawings.
FIG. 1 is a diagram showing a schematic configuration of a quality management apparatus according to an embodiment of the present invention. FIG. 2 is a flowchart for explaining a quality management method executed in the quality management apparatus according to the embodiment of the present invention. Hereinafter, referring to FIG. 1 and FIG. 2, in a steel product manufacturing plant, “thin coil” as a product, “material” and “surface flaw generation” as targets, and satisfying “material” An example of managing so as to reduce the “surface flaw generation amount” will be described.

図１に示すように、製品の製造プロセス（工程Ａ〜Ｃ）から収集される、製品を製造する上で設定が必要な変数である各操業因子の操業条件が製品毎に出力される。さらに最終工程Ｃからは、該操業因子の操業条件で得られた製品の品質を検査した結果である品質データが製品毎に出力される。ここで、製品毎に出力される各操業因子の操業条件及び品質データの両者を合わせてプロセスデータと呼ぶ。   As shown in FIG. 1, the operation conditions of each operation factor, which is a variable that needs to be set for manufacturing a product, collected from the product manufacturing process (steps A to C), is output for each product. Furthermore, from the final process C, quality data, which is the result of inspecting the quality of the product obtained under the operating conditions of the operating factor, is output for each product. Here, the operation conditions and quality data of each operation factor output for each product are collectively referred to as process data.

図１において、データ入力部１０１は、製造プロセスからプロセスデータを収集、入力する。   In FIG. 1, a data input unit 101 collects and inputs process data from a manufacturing process.

データ蓄積部１０２は、データ入力部１０１により入力されるプロセスデータを蓄積する。   The data storage unit 102 stores process data input from the data input unit 101.

操業範囲算出部１０３は、データ蓄積部１０２に蓄積されたプロセスデータから、所定の品質が所望の範囲内となり、製品の製造条件として用いることが可能な操業因子Ｘの操業条件の範囲Ｒ（以下、「操業因子Ｘの範囲Ｒ」と称する）を算出する。ここでは、図２のステップＳ２０１に相当する処理を行う。   The operation range calculation unit 103 determines from the process data stored in the data storage unit 102 that the predetermined quality is within a desired range and the operation condition range R of the operation factor X that can be used as a product manufacturing condition (hereinafter referred to as the product condition). , Referred to as “range R of operating factor X”). Here, processing corresponding to step S201 in FIG. 2 is performed.

特徴量算出部１０４は、データ蓄積部１０２から解析対象製品のプロセスデータを取得し、品質に関係する特徴量を算出する。ここでは、図２のステップＳ２０２に相当する処理を行う。特徴量算出部１０４は、プロセスデータが操業条件や品質データが時々刻々と記録された時系列データである場合、プロセスデータを製品の品質管理に必要な時間単位又は製品長さ単位で分割する。以降は、分割した一つ一つの単位を仮想的な製品として取り扱い、品質管理を行う。分割した単位の時系列データの形態をしたプロセスデータから品質に関係する特徴量を算出し、品質管理に用いるプロセスデータとする。特徴量としては、最大、最小、平均、分散、標準偏差等の統計量が主に用いられる。また、既に品質管理の単位のデータとなっている場合には、そのデータをそのまま特徴量として取り扱ってもよい。   The feature amount calculation unit 104 acquires process data of the analysis target product from the data storage unit 102 and calculates a feature amount related to quality. Here, processing corresponding to step S202 of FIG. 2 is performed. When the process data is time-series data in which operation conditions and quality data are recorded every moment, the feature amount calculation unit 104 divides the process data in units of time or product length necessary for product quality management. Thereafter, each divided unit is handled as a virtual product, and quality control is performed. A feature quantity related to quality is calculated from process data in the form of time-series data in divided units, and is used as process data used for quality management. As the feature quantity, statistics such as maximum, minimum, average, variance, standard deviation, etc. are mainly used. Further, when the data is already in quality control units, the data may be handled as it is as a feature amount.

確率分布相関解析部１０５は、特徴量算出部１０４から特徴量（或いは品質管理の単位のデータ）を取得し、操業因子Ｘと品質Ｑとの関係式Ｑ(Ｘ)を確率分布相関解析法により求める。ここでは、図２のステップＳ２０３に相当する処理を行う。確率分布相関解析法とは、特許文献１に開示されている散布図や相関係数では捉えられない操業因子と品質との間の相関を解析することができる近似式算出方法である。   The probability distribution correlation analysis unit 105 obtains a feature amount (or quality control unit data) from the feature amount calculation unit 104, and obtains a relational expression Q (X) between the operation factor X and the quality Q by the probability distribution correlation analysis method. Ask. Here, processing corresponding to step S203 in FIG. 2 is performed. The probability distribution correlation analysis method is an approximate expression calculation method that can analyze a correlation between an operation factor and quality that cannot be captured by a scatter diagram or a correlation coefficient disclosed in Patent Document 1.

影響度算出部１０６は、操業範囲算出部１０３により算出された操業因子Ｘの範囲Ｒと、確率分布相関解析部１０５により算出された関係式Ｑ(Ｘ)とを用いて、操業因子の品質に対する影響度Ｗを算出する。ここでは、図２のステップＳ２０４に相当する処理を行う。   The influence degree calculation unit 106 uses the range R of the operation factor X calculated by the operation range calculation unit 103 and the relational expression Q (X) calculated by the probability distribution correlation analysis unit 105 to evaluate the quality of the operation factor. The influence degree W is calculated. Here, processing corresponding to step S204 in FIG. 2 is performed.

改善対象操業条件設定部１０７は、データ蓄積部１０２から品質を改善したい製品（品質改善対象製品）の操業因子の操業条件を取得する。   The improvement target operation condition setting unit 107 acquires the operation condition of the operation factor of the product (quality improvement target product) whose quality is to be improved from the data storage unit 102.

品質改善度算出部１０８は、確率分布相関解析部１０５により算出された関係式Ｑ(Ｘ)と、改善対象操業条件設定部１０７により取得された品質改善対象製品の操業因子の操業条件とを用いて、操業因子毎に品質の改善度合いΔＱを算出する。品質の改善度合いΔＱは、例えば操業因子の変更量に応じて得られる品質の最大値を示す。ここでは、図２のステップＳ２０５に相当する処理を行う。   The quality improvement degree calculation unit 108 uses the relational expression Q (X) calculated by the probability distribution correlation analysis unit 105 and the operation condition of the operation factor of the quality improvement target product acquired by the improvement target operation condition setting unit 107. Thus, the quality improvement degree ΔQ is calculated for each operation factor. The quality improvement degree ΔQ indicates, for example, the maximum value of quality obtained according to the change amount of the operation factor. Here, processing corresponding to step S205 in FIG. 2 is performed.

改善期待度算出部１０９は、影響度算出部１０６により算出された影響度Ｗと、品質改善度算出部１０８により算出された品質の改善度合いΔＱとを用いて、品質改善に効果の高い操業因子の改善期待度ＷＱを算出する。改善期待度ＷＱは、品質改善に効果の高い操業因子の優先順位を示す。ここでは、図２のステップＳ２０６に相当する処理を行う。   The improvement expectation degree calculation unit 109 uses the influence degree W calculated by the influence degree calculation unit 106 and the quality improvement degree ΔQ calculated by the quality improvement degree calculation unit 108, so that the operation factor having a high effect on the quality improvement is obtained. The improvement expectation level WQ is calculated. The improvement expectation WQ indicates the priority order of operation factors that are highly effective for quality improvement. Here, processing corresponding to step S206 in FIG. 2 is performed.

品質改善案表示部１１０は、改善期待度算出部１０９により算出された改善期待度ＷＱの大きい順に、変更候補となる操業因子を表示する。ここでは、図２のステップＳ２０７に相当する処理を行う。   The quality improvement plan display unit 110 displays operation factors as change candidates in descending order of the expected improvement WQ calculated by the expected improvement calculation unit 109. Here, processing corresponding to step S207 in FIG. 2 is performed.

次に、図２のフローチャートを参照して、本実施形態に係る品質管理装置において実行される品質管理方法について説明する。データ蓄積部１０２に蓄積されたプロセスデータのうち、解析対象製品のプロセスデータであって、所定の品質を満足するプロセスデータに基づいて、変数である操業因子Ｘの範囲Ｒ（操業因子Ｘの操業条件の範囲Ｒ）を操業因子毎に算出する（ステップＳ２０１）。   Next, a quality management method executed in the quality management apparatus according to this embodiment will be described with reference to the flowchart of FIG. Based on the process data of the analysis target product among the process data stored in the data storage unit 102 and satisfying a predetermined quality, the range R of the operation factor X as a variable (the operation of the operation factor X) A condition range R) is calculated for each operation factor (step S201).

操業因子Ｘの範囲Ｒは、少なくとも操業条件の変更の必要がない所定の品質を満足する範囲である。算出方法の一つとして、操業因子Ｘの操業条件の最小と最大を範囲Ｒとする方法がある。また、別の算出方法として、操業因子Ｘを正規分布もしくは対称な分布とみなしてよい場合に、操業因子Ｘの平均μと標準偏差σを算出し、定数Ｎを用いて平均μを中心とする前後Ｎ倍の標準偏差σを範囲Ｒとする方法がある。このとき、定数Ｎは通常２．０を基準とし、操業因子Ｘに対して参考とならないデータ（いわゆるノイズや外れ値）の含み具合によって１．０〜６．０程度の範囲を取るものとする。また、別の算出方法として、操業因子Ｘを非対称な分布とみなしてよい場合に、分布形状と設備での設定可能範囲に応じて少なくとも９０％以上の累積確率分布となる範囲を覆うものとし、上位及び下位を０〜１０%のいずれかの割合に相当する部分を除くことで、操業因子Ｘの範囲Ｒの累積確率分布が一定の場合に範囲の大きさ（範囲の最大と最小の差）が最小となる場合を範囲Ｒとする方法がある。   The range R of the operating factor X is a range that satisfies at least a predetermined quality that does not require a change in operating conditions. As one of the calculation methods, there is a method in which the minimum and maximum operating conditions of the operating factor X are in the range R. As another calculation method, when the operation factor X may be regarded as a normal distribution or a symmetric distribution, the average μ and the standard deviation σ of the operation factor X are calculated, and a constant N is used to center the average μ. There is a method of setting a standard deviation σ N times before and after as a range R. At this time, the constant N is normally set to 2.0 as a standard, and takes a range of about 1.0 to 6.0 depending on the inclusion of data (so-called noise and outliers) that are not helpful for the operation factor X. . Further, as another calculation method, when the operation factor X may be regarded as an asymmetric distribution, it covers a range where the cumulative probability distribution is at least 90% or more according to the distribution shape and the settable range in the facility, The size of the range (the difference between the maximum and minimum of the range) when the cumulative probability distribution of the range R of the operating factor X is constant by excluding the upper and lower portions corresponding to any ratio of 0 to 10% There is a method of setting the range R to be the minimum.

次に、データ蓄積部１０２に蓄積されているプロセスデータから、品質管理に必要な時間単位又は製品長さ単位で分割し、分割された単位で品質に関係する特徴量（例えば最大、最小、平均、分散、標準偏差等の統計量）を算出する（ステップＳ２０２）。これにより、品質管理の単位となったプロセスデータを得る。   Next, the process data stored in the data storage unit 102 is divided into time units or product length units necessary for quality control, and feature quantities related to quality in the divided units (for example, maximum, minimum, average). , Statistics such as variance and standard deviation) are calculated (step S202). As a result, process data that is a unit of quality control is obtained.

次に、品質管理に必要な単位で分割されたプロセスデータを用いて、操業因子Ｘと品質Ｑとの相関を確率分布相関解析法にて解析し、関係式Ｑ(Ｘ)を求める（ステップＳ２０３）。確率分布相関解析法では、操業因子Ｘを複数の水準に分割し、各水準において品質データＱの確率分布を推定し、推定した確率分布を基に品質データＱの代表値を計算する。そして、操業因子Ｘの各水準の代表値と品質データＱの代表値との相関を表わす関係式Ｑ(Ｘ)を計算する。   Next, using the process data divided in units necessary for quality control, the correlation between the operation factor X and the quality Q is analyzed by the probability distribution correlation analysis method to obtain the relational expression Q (X) (step S203). ). In the probability distribution correlation analysis method, the operation factor X is divided into a plurality of levels, the probability distribution of the quality data Q is estimated at each level, and the representative value of the quality data Q is calculated based on the estimated probability distribution. Then, a relational expression Q (X) representing the correlation between the representative value of each level of the operation factor X and the representative value of the quality data Q is calculated.

次に、上記ステップＳ２０１で算出された操業因子Ｘの範囲Ｒと、上記ステップＳ２０３で得られた操業因子Ｘと品質Ｑとの関係式Ｑ(Ｘ)とから、各操業因子の影響度Ｗを算出する（ステップＳ２０４）。影響度Ｗは、着目した操業因子で変更可能な品質の範囲の大きさを示す指標であり、操業因子の品質に対する影響の大きさを示す。ｓ番目の操業因子Ｘｓの影響度Ｗｓは、下式（１）に示すように、操業因子Ｘｓの範囲Ｒｓにおける品質Ｑの最大と最小の差を示す。   Next, from the range R of the operating factor X calculated in step S201 and the relational expression Q (X) between the operating factor X and quality Q obtained in step S203, the influence W of each operating factor is calculated. Calculate (step S204). The influence degree W is an index indicating the magnitude of the quality range that can be changed by the focused operation factor, and indicates the magnitude of the influence of the operation factor on the quality. The influence Ws of the s-th operation factor Xs indicates the difference between the maximum and minimum quality Q in the range Rs of the operation factor Xs, as shown in the following equation (1).

Figure 0005020728
Figure 0005020728

また、影響度Ｗｓは、関係式Ｑ(Ｘ)を操業因子Ｘｓで微分した値を用いて表わすと、操業因子Ｘｓの範囲Ｒｓの最大値Ｒｓｍａｘ及び最小値Ｒｓｍｉｎを用いて、下式（２）のようになる。なお、関係式Ｑ(Ｘ)が１次式の場合、（式１）及び（式２）は同等となる。   Further, when the influence Ws is expressed using a value obtained by differentiating the relational expression Q (X) with the operation factor Xs, the following equation (2) is used by using the maximum value Rsmax and the minimum value Rsmin of the range Rs of the operation factor Xs. become that way. When the relational expression Q (X) is a linear expression, (Expression 1) and (Expression 2) are equivalent.

Figure 0005020728
Figure 0005020728

次に、改善対象操業条件Ｘｔ（品質改善対象製品の操業因子の操業条件）に対する各操業因子の品質の改善度合いΔＱｔを算出する（ステップＳ２０５）。品質の改善度合いΔＱｔは、下式（３）に示すように、改善対象操業条件Ｘｔの場合の品質Ｑ(Ｘｔ)から変更候補となる操業条件Ｘｔ'を用いて品質Ｑ(Ｘｔ')との品質の差の大きさとなる。   Next, the quality improvement degree ΔQt of each operation factor with respect to the improvement target operation condition Xt (operation condition of the operation factor of the quality improvement target product) is calculated (step S205). As shown in the following equation (3), the quality improvement degree ΔQt is calculated from the quality Q (Xt) in the case of the operation condition Xt to be improved using the operation condition Xt ′ that is a change candidate and the quality Q (Xt ′). The magnitude of the quality difference.

Figure 0005020728
Figure 0005020728

ここで、図３を参照して、このときの変更候補となる操業条件Ｘｔ'を説明する。目標の品質Ｑobjを実現する操業条件Ｘobjが操業因子Ｘの範囲Ｒ内にある場合（図３（ａ））、Ｘｔ'＝Ｘobjである。それに対して、操業条件Ｘobjが操業因子Ｘの範囲Ｒに満たない場合（図３（ｂ））、Ｘｔ'＝範囲Ｒの下限値Ｒminであり、操業条件Ｘobjが操業因子Ｘの範囲Ｒを超える場合（図３（ｃ））、Ｘｔ'＝範囲Ｒの上限値Ｒmaxである。 Here, with reference to FIG. 3, the operation condition Xt ′ which is a candidate for change at this time will be described. If operating conditions X obj to achieve the quality Q obj of the target is within the range R of operating factors X (FIG. 3 (a)), a Xt '= X obj. On the other hand, when the operating condition X obj is less than the range R of the operating factor X (FIG. 3B), Xt ′ = the lower limit value R min of the range R, and the operating condition X obj is the range of the operating factor X When exceeding R (FIG. 3C), Xt ′ = the upper limit value R max of the range R.

次に、上記ステップＳ２０４で算出された影響度Ｗと、上記ステップＳ２０５で算出された品質の改善度合いΔＱとの積で表わされる改善期待度ＷＱを算出する（ステップＳ２０６）。   Next, an improvement expectation degree WQ represented by the product of the influence degree W calculated in step S204 and the quality improvement degree ΔQ calculated in step S205 is calculated (step S206).

次に、上記ステップＳ２０６で算出された改善期待度ＷＱが大きい操業因子から順に、操業因子の変更案を表示する（ステップＳ２０７）。改善期待度ＷＱが同じ場合は、例えば影響度Ｗの順で順序付ける。この表示に基づいて製品の標準的な操業条件を変更し、各工程の操業者が製造することで、品質を所望の範囲内に変更することができる。   Next, the operation factor change proposals are displayed in order from the operation factor having the highest improvement expectation WQ calculated in step S206 (step S207). When the improvement expectation degree WQ is the same, for example, the order is in the order of the influence degree W. The quality can be changed within a desired range by changing the standard operating conditions of the product based on this display and manufacturing by the operator of each process.

（実施例）
鉄鋼製品の製造工場において、製品として「薄板コイル」、品質として「製品材質」及び最終工程における検査によって計測される「表面疵発生量」、多品種の製品作り分けに必要な変数である操業因子として「加熱炉からの製品の抽出温度（抽出温度）」、「圧延後の捲き取り機直前の温度（捲取温度）」、「焼鈍時の最高温度（焼鈍温度）」を対象として、製品の「材質」を満足しつつ「表面疵発生量」を低減させる操業因子の操業条件を提示する例について説明する。品質の一つである「表面疵発生量」として、「薄板コイル」の表面にある単位面積あたりの表面疵個数を用いた。なお、ここでの表面疵は、鉄鋼鋳片の内部に含まれる介在物、パウダー等を起因として発生するスリバー疵を対象とした。データは、解析対象とした特定品種の製品のうち、連続した期間で製造された製品材質を満足する３０００本の製品に関するデータを用いた。 (Example)
In steel product manufacturing factories, "thin sheet coils" as products, "product materials" as quality, and "surface flaw generation" measured by inspection in the final process, operating factors that are necessary variables for different types of products. As the target of “Product extraction temperature from heating furnace (extraction temperature)”, “Temperature immediately before scraper after rolling (scraping temperature)”, “Maximum temperature during annealing (annealing temperature)” An example will be described in which the operating conditions of operating factors that reduce the “surface flaw generation amount” while satisfying the “material” are presented. As the “surface flaw generation amount” which is one of the qualities, the number of surface flaws per unit area on the surface of the “thin coil” was used. The surface defects here were sliver defects generated due to inclusions, powders and the like contained in the steel slab. As data, data on 3000 products satisfying the product material manufactured in a continuous period among the products of the specific varieties to be analyzed were used.

「製品材質」を目標の範囲内にするためには、「抽出温度」と「捲取温度」と「焼鈍温度」が正しく目標の温度の範囲内に管理されている必要がある。さらに、もう一方の品質である「表面疵発生量」を低減させる必要がある。   In order to bring the “product material” within the target range, it is necessary that the “extraction temperature”, the “scoring temperature”, and the “annealing temperature” are correctly controlled within the target temperature range. Furthermore, it is necessary to reduce the “surface flaw generation amount” which is the other quality.

「製品材質」を目標の範囲内に製造するための必要条件として、「抽出温度」と「捲取温度」と「焼鈍温度」が関連をもってそれぞれの温度の範囲が指定されている。また、「抽出温度」は「製品材質」を満足するために下限値が決められており、一方で操業上は所定の温度以上に高くしたくない。   As a necessary condition for manufacturing the “product material” within the target range, the “extraction temperature”, the “scraping temperature”, and the “annealing temperature” are associated with each temperature range. Further, the “extraction temperature” has a lower limit value in order to satisfy the “product material”, but on the other hand, it is not desired to make it higher than a predetermined temperature in operation.

「表面疵発生量」は「抽出温度」との間に傾向があるが、明確な相関はないことが知られており、操業条件を決めた場合であっても確定的な発生量を見積ることは困難である。   “Surface generation amount” has a tendency to be related to “extraction temperature”, but it is known that there is no clear correlation, and even when operating conditions are determined, a definite generation amount is estimated. It is difficult.

各操業因子Ｘの操業範囲Ｒとして、「抽出温度」は製品の「製品材質」を満足するための最低温度から操業上許される温度の上限を範囲として最小を０、最大を１で表わす。「捲取温度」は用いたデータの累積確率の５％〜９５％の温度範囲であり、最大を１、最小を０とする０〜１の範囲で表わす。「焼鈍温度」は平均をμ、標準偏差をσとしたときの範囲[μ−２σ,μ＋２σ]であり、最大を１、最小を０として表わす（ステップＳ２０１）。   As the operation range R of each operation factor X, the “extraction temperature” represents 0 as the minimum and 1 as the maximum with the upper limit of the temperature allowed for operation from the minimum temperature for satisfying the “product material” of the product. The “scoring temperature” is a temperature range of 5% to 95% of the cumulative probability of the data used, and is expressed in a range of 0 to 1 where 1 is the maximum and 0 is the minimum. “Annealing temperature” is a range [μ−2σ, μ + 2σ] where the average is μ and the standard deviation is σ, and the maximum is 1 and the minimum is 0 (step S201).

各工程の操業の結果得られた温度を品質管理の単位で平均値を計算し、また、最終工程においてスリバー疵を検査した結果から単位面積あたりのスリバー疵個数への換算を行う特徴量の算出を行う（ステップＳ２０２）。   Calculate the average value of the temperature obtained as a result of the operation of each process in the unit of quality control, and calculate the feature value that converts the result of the inspection of the sliver rods in the final process into the number of sliver rods per unit area Is performed (step S202).

操業因子の大小によって「表面疵発生量」の傾向の違いを確率分布の違いとして捉え、操業因子と「表面疵発生量」との相関関係（関係式）を確率分布相関解析法により求めた（ステップＳ２０３）。その結果、図４に示すように、「抽出温度」が「表面疵発生量（表面疵個数）」と相関があることが判明した。図４は、横軸に表面疵個数、縦軸に抽出温度を示し、縦軸を１０の水準に分割し、各水準に該当するデータに対する確率分布を推定した。推定した確率分布に基づいて、累積確率が等しい点を結ぶ等累積確率の等高線（細実線）を示している。ただし、１０の水準のうち、該当するデータが無い水準が１つある。併せて、「抽出温度」の各水準で推定した確率分布の累積確率が８０％となる疵個数の部分に×印を示している。この１０個（計算できない点が１つあるので実際は９個）の×印に対して最小２乗法による直線（＝関係式：太実線）と２次関数の曲線（太鎖線）を示す。図４の最小２乗法による直線（太実線）からスリバー疵個数を低減させるには、「抽出温度」を増加させることである。他の操業因子である「捲取温度」、「焼鈍温度」についても、わずかであるが温度を増加させるとスリバー疵個数が減少する相関が得られた。   The difference in the tendency of “surface flaw generation” depending on the size of the operation factor is regarded as the difference in probability distribution, and the correlation (relational expression) between the operation factor and “surface flaw generation” was obtained by the probability distribution correlation analysis method ( Step S203). As a result, as shown in FIG. 4, it was found that “extraction temperature” has a correlation with “surface flaw generation amount (number of surface flaws)”. In FIG. 4, the horizontal axis indicates the number of surface defects, the vertical axis indicates the extraction temperature, the vertical axis is divided into 10 levels, and the probability distribution for the data corresponding to each level is estimated. Based on the estimated probability distribution, contour lines (thin solid lines) of equal cumulative probabilities connecting points with equal cumulative probabilities are shown. However, of the 10 levels, there is one level for which no corresponding data exists. In addition, a cross is shown in the portion of the number of baskets where the cumulative probability of the probability distribution estimated at each level of “extraction temperature” is 80%. A straight line (= relational expression: thick solid line) and a quadratic function curve (thick chain line) by the least square method are shown with respect to the 10 marks (actually 9 because there is one point that cannot be calculated). In order to reduce the number of sliver rods from the straight line (thick solid line) by the least square method of FIG. 4, it is to increase the “extraction temperature”. As for other operating factors such as “cracking temperature” and “annealing temperature”, a slight correlation was obtained that the number of sliver rods decreased with increasing temperature.

次に、影響度Ｗを求める（ステップＳ２０４）。図４において関係式を示す右下がりの直線から「抽出温度」が０のときに「表面疵個数」が２であり、「抽出温度」が１のときに「表面疵個数」が１であるから、「抽出温度」の影響度Ｗは１．０である。他の操業因子である「捲取温度」の影響度は０．０１、「焼鈍温度」の影響度は０．０１である。   Next, the influence degree W is obtained (step S204). From the right-downward straight line showing the relational expression in FIG. 4, when the “extraction temperature” is 0, the “number of surface defects” is 2, and when the “extraction temperature” is 1, the “number of surface defects” is 1. The influence degree W of “extraction temperature” is 1.0. The degree of influence of “running temperature” as another operating factor is 0.01, and the degree of influence of “annealing temperature” is 0.01.

品質改善対象製品の操業因子の操業条件は、「抽出温度」が０．１、「捲取温度」が０．５、「焼鈍温度」が０．５である。したがって、「表面疵発生量」を低減させるためには、それぞれの操業因子を１まで増加させることであるから、操業因子毎の品質の改善度合いΔＱは、「抽出温度」については０．９、「捲取温度」及び「焼鈍温度」については０．００５となる（ステップＳ２０５）。   The operation conditions of the operation factors of the quality improvement target product are “extraction temperature” of 0.1, “scraping temperature” of 0.5, and “annealing temperature” of 0.5. Therefore, in order to reduce the “surface flaw generation amount”, it is to increase each operation factor to 1, so the quality improvement degree ΔQ for each operation factor is 0.9 for the “extraction temperature”, The “scoring temperature” and “annealing temperature” are 0.005 (step S205).

次に、「表面疵発生量」の改善期待度ＷＱ＝Ｗ×ΔＱを求める（ステップＳ２０６）。「抽出温度」の改善期待度ＷＱは、Ｗ×ΔＱ＝１×０．９＝０．９、「捲取温度」及び「焼鈍温度」の改善期待度ＷＱは、Ｗ×ΔＱ＝０．０１×０．００５＝５．０×１０-5である。 Next, an improvement expectation WQ = W × ΔQ of “surface flaw generation amount” is obtained (step S206). The improvement expectation degree WQ of the “extraction temperature” is W × ΔQ = 1 × 0.9 = 0.9, and the improvement expectation degree WQ of the “scraping temperature” and the “annealing temperature” is W × ΔQ = 0.01 × 0.005 = 5.0 × 10 −5 .

下記の表１に、本実施例における各操業因子の操業条件、影響度Ｗ、品質の改善度合いΔＱ、改善期待度ＷＱの値を示す。   Table 1 below shows values of operation conditions, influence W, quality improvement degree ΔQ, and improvement expectation degree WQ of each operation factor in the present embodiment.

Figure 0005020728
Figure 0005020728

改善期待度ＷＱの大きい順に「表面疵発生量」を低減させる効果が大きい操業因子の操業条件の変更案を表示した（ステップＳ２０７）。この表示結果に基づいて、加熱炉を操業している操業者が、対象となる製品５０本に対して、「抽出温度」を操業上許される上限である１となるように操業した。それらの製品をスリバー疵を検査する工程に検査した結果、すべての単位面積あたりのスリバー疵個数が１を下回る結果となった。   The proposal for changing the operating condition of the operating factor having a large effect of reducing the “surface flaw generation amount” is displayed in descending order of the improvement expectation degree WQ (step S207). Based on this display result, the operator operating the heating furnace operated the “extraction temperature” to 50, which is the upper limit allowed for operation, for 50 target products. As a result of inspecting those products in the process of inspecting sliver rods, the number of sliver rods per unit area was less than 1.

なお、本発明は複数の機器から構成されるシステムに適用しても、一つの機器からなる装置に適用してもよい。   The present invention may be applied to a system composed of a plurality of devices or an apparatus composed of a single device.

また、本発明の目的は、上述した実施形態の機能を実現するソフトウェアのプログラムコードを記録した記憶媒体を、システム或いは装置に供給することによっても達成される。この場合、そのシステム或いは装置のコンピュータ（又はＣＰＵやＭＰＵ）が記憶媒体に格納されたプログラムコードを読み出し実行する。   The object of the present invention can also be achieved by supplying a storage medium storing software program codes for realizing the functions of the above-described embodiments to a system or apparatus. In this case, the computer (or CPU or MPU) of the system or apparatus reads and executes the program code stored in the storage medium.

この場合、記憶媒体から読み出されたプログラムコード自体が上述した実施形態の機能を実現することになり、プログラムコード自体及びそのプログラムコードを記憶した記憶媒体は本発明を構成することになる。プログラムコードを供給するための記憶媒体としては、例えば、フレキシブルディスク、ハードディスク、光ディスク、光磁気ディスク、ＣＤ−ＲＯＭ、ＣＤ−Ｒ、磁気テープ、不揮発性のメモリカード、ＲＯＭ等を用いることができる。   In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiments, and the program code itself and the storage medium storing the program code constitute the present invention. As a storage medium for supplying the program code, for example, a flexible disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

本発明の実施形態に係る品質管理装置の概略構成を示す図である。It is a figure which shows schematic structure of the quality control apparatus which concerns on embodiment of this invention. 本発明の実施形態に係る品質管理装置において実行される品質管理方法を説明するためのフローチャートである。It is a flowchart for demonstrating the quality control method performed in the quality control apparatus which concerns on embodiment of this invention. 品質の改善度合いΔＱの算出に必要な変更候補となる操業因子の操業条件Ｘｔ'を求める方法を説明する図である。It is a figure explaining the method of calculating | requiring the operation condition Xt 'of the operation factor used as the change candidate required for calculation of quality improvement degree (DELTA) Q. 表面疵個数と操業因子（加熱炉からの製品の抽出温度）との関係式Ｑ(Ｘ)を確率分布相関解析法により求めた結果を示す図である。It is a figure which shows the result of having calculated | required the relational expression Q (X) of the surface flaw number and the operation factor (the extraction temperature of the product from a heating furnace) by the probability distribution correlation analysis method.

符号の説明Explanation of symbols

１０１ データ入力部
１０２ データ蓄積部
１０３ 操業範囲算出部
１０４ 特徴量算出部
１０５ 確率分布相関解析部
１０６ 影響度算出部
１０７ 改善対象操業条件設定部
１０８ 品質改善度算出部
１０９ 改善期待度算出部
１１０ 品質改善案表示部 DESCRIPTION OF SYMBOLS 101 Data input part 102 Data storage part 103 Operation range calculation part 104 Feature-value calculation part 105 Probability distribution correlation analysis part 106 Influence calculation part 107 Improvement object operation condition setting part 108 Quality improvement degree calculation part 109 Improvement expectation degree calculation part 110 Quality Improvement plan display

Claims (14)

製造プロセスの工程を経て製造される製品について複数の品質を所望の範囲内に管理するための品質管理装置であって、
前記製造プロセスから、操業因子の操業条件、及び、該操業因子の操業条件で得られた製品の品質を含むプロセスデータを収集するデータ入力手段と、
前記データ入力手段により入力されるプロセスデータを蓄積するデータ蓄積手段と、
前記データ蓄積手段に蓄積されたプロセスデータから、所定の品質が所望の範囲内となる操業因子の操業条件の範囲を算出する操業範囲算出手段と、
前記データ蓄積手段から取得された、所定の品質を満足する製品である解析対象製品のプロセスデータに基づいて、操業因子と品質との関係式を確率分布相関解析法により求める確率分布相関解析手段と、
前記操業範囲算出手段により算出された操業因子の操業条件の範囲と、前記確率分布相関解析手段により求められた関係式とを用いて、操業因子の品質に対する影響度を算出する影響度算出手段と、
前記データ蓄積手段から、品質を改善したい製品である品質改善対象製品の操業因子の操業条件を取得する改善対象操業条件設定手段と、
前記確率分布相関解析手段により求められた関係式と、前記改善対象操業条件設定手段により取得された品質改善対象製品の操業因子の操業条件とを用いて、品質の改善度合いを算出する品質改善度算出手段と、
前記影響度算出手段により算出された影響度と、前記品質改善度算出手段により算出された品質の改善度合いとを用いて、操業因子の改善期待度を算出する改善期待度算出手段とを備えたことを特徴とする品質管理装置。 A quality control device for managing a plurality of qualities within a desired range for a product manufactured through a manufacturing process,
Data input means for collecting process data including the operating conditions of the operating factors and the quality of the product obtained under the operating conditions of the operating factors from the manufacturing process;
Data storage means for storing process data input by the data input means;
From the process data stored in the data storage means, an operation range calculation means for calculating a range of operation conditions of the operation factor that the predetermined quality is within a desired range;
A probability distribution correlation analysis means for obtaining a relational expression between the operating factor and the quality by a probability distribution correlation analysis method based on the process data of the analysis target product which is a product satisfying a predetermined quality obtained from the data storage means; ,
The degree of influence calculation means for calculating the degree of influence on the quality of the operation factor using the range of the operation condition of the operation factor calculated by the operation range calculation means and the relational expression obtained by the probability distribution correlation analysis means; ,
From the data storage means , improvement target operation condition setting means for obtaining the operation condition of the operation factor of the quality improvement target product that is a product whose quality is to be improved;
The quality improvement degree for calculating the degree of quality improvement using the relational expression obtained by the probability distribution correlation analysis means and the operation conditions of the operation factor of the quality improvement target product acquired by the improvement target operation condition setting means A calculation means;
An improvement expectation degree calculating means for calculating an improvement expectation degree of the operation factor using the influence degree calculated by the influence degree calculating means and the improvement degree of the quality calculated by the quality improvement degree calculating means. A quality control device characterized by that. 前記データ蓄積手段から、所定の品質を満足する製品である解析対象製品のプロセスデータを取得して特徴量を算出する特徴量算出手段を備え、
前記確率分布相関解析手段は、前記特徴量算出手段により算出された特徴量に基づいて、前記関係式を求めることを特徴とする請求項１に記載の品質管理装置。 From the data storage means , comprising a feature quantity calculation means for obtaining process data of a product to be analyzed which is a product satisfying a predetermined quality and calculating a feature quantity,
The quality control apparatus according to claim 1, wherein the probability distribution correlation analysis unit obtains the relational expression based on the feature amount calculated by the feature amount calculation unit. 前記改善期待度算出手段により算出された改善期待度の大きい順に変更候補となる操業因子を表示する品質改善案表示手段を更に備えたことを特徴とする請求項１又は２に記載の品質管理装置。   The quality management apparatus according to claim 1, further comprising a quality improvement plan display means for displaying operation factors that are candidates for change in descending order of the expected improvement degree calculated by the improvement expectation degree calculation means. . 前記操業範囲算出手段は、前記データ蓄積手段に蓄積されたプロセスデータに含まれる操業因子の操業条件の最大と最小を、該操業因子の操業条件の範囲として算出することを特徴とする請求項１又は２に記載の品質管理装置。   The operation range calculation means calculates the maximum and minimum of the operation condition of the operation factor included in the process data stored in the data storage means as the range of the operation condition of the operation factor. Or the quality control apparatus according to 2. 前記操業範囲算出手段は、前記データ蓄積手段に蓄積されたプロセスデータに含まれる操業因子の操業条件の平均と標準偏差を算出し、所定の定数を用いて、平均を中心として前後に標準偏差の定数倍の広がりをもつ範囲を、該操業因子の操業条件の範囲として算出することを特徴とする請求項１又は２に記載の品質管理装置。   The operation range calculation means calculates the average and standard deviation of the operation conditions of the operation factors included in the process data accumulated in the data accumulation means, and uses a predetermined constant to calculate the standard deviation around the average. The quality control apparatus according to claim 1 or 2, wherein a range having a constant multiple spread is calculated as a range of operation conditions of the operation factor. 前記操業範囲算出手段は、前記データ蓄積手段に蓄積されたプロセスデータに含まれる操業因子の操業条件の分布を推定し、その推定した分布形状のうち累積確率が所定の確率以上の累積分布となる範囲を、該操業因子の操業条件の範囲として算出することを特徴とする請求項１又は２に記載の品質管理装置。   The operation range calculation means estimates a distribution of operation conditions of operation factors included in the process data accumulated in the data accumulation means, and a cumulative probability of the estimated distribution shape is a cumulative distribution greater than or equal to a predetermined probability. The quality control apparatus according to claim 1 or 2, wherein the range is calculated as a range of operation conditions of the operation factor. 前記影響度算出手段は、前記確率分布相関解析手段により求められた関係式を用いて、前記操業範囲算出手段により算出された操業因子の操業条件の範囲における品質の最大値と最小値の差を影響度として算出することを特徴とする請求項１〜６のいずれか１項に記載の品質管理装置。   The influence degree calculation means uses the relational expression obtained by the probability distribution correlation analysis means to calculate the difference between the maximum value and the minimum value of the quality in the operation condition range of the operation factor calculated by the operation range calculation means. The quality control apparatus according to claim 1, wherein the quality management apparatus calculates the influence degree. 前記影響度算出手段は、前記確率分布相関解析手段により求められた関係式を操業因子で微分した値、前記操業範囲算出手段により算出された操業因子の操業条件の範囲の最大値と最小値を用いて影響度を算出することを特徴とする請求項１〜６のいずれか１項に記載の品質管理装置。   The influence calculation means is a value obtained by differentiating the relational expression obtained by the probability distribution correlation analysis means with an operation factor, and a maximum value and a minimum value of the operation condition range of the operation factor calculated by the operation range calculation means. The quality control apparatus according to claim 1, wherein the influence degree is calculated by using the quality control apparatus. 前記品質改善度算出手段は、前記確率分布相関解析手段により求められた関係式を用いて、前記改善対象操業条件設定手段により取得された品質改善対象製品の操業因子の場合の品質と、変更候補となる操業因子の場合の品質との差を品質の改善度合いとして算出することを特徴とする請求項１〜８のいずれか１項に記載の品質管理装置。   The quality improvement degree calculation means uses the relational expression obtained by the probability distribution correlation analysis means, the quality in the case of the operation factor of the quality improvement target product obtained by the improvement target operation condition setting means, and the change candidate The quality control apparatus according to claim 1, wherein a difference from the quality in the case of an operating factor is calculated as a quality improvement degree. 前記改善期待度算出手段は、前記影響度算出手段により算出された影響度と、前記品質改善度算出手段により算出された品質の改善度合いとの積を改善期待度として算出することを特徴とする請求項１〜９のいずれか１項に記載の品質管理装置。   The improvement expectation degree calculation means calculates the product of the influence degree calculated by the influence degree calculation means and the improvement degree of quality calculated by the quality improvement degree calculation means as an improvement expectation degree. The quality control apparatus of any one of Claims 1-9. 前記製品は鉄鋼製品であることを特徴とする請求項１〜１０のいずれか１項に記載の品質管理装置。   The quality control apparatus according to claim 1, wherein the product is a steel product. 前記品質の一つが表面疵発生量であることを特徴とする請求項１１に記載の品質管理装置。   The quality control apparatus according to claim 11, wherein one of the qualities is a surface flaw generation amount. 製造プロセスの工程を経て製造される製品について複数の品質を所望の範囲内に管理するための品質管理方法であって、
前記製造プロセスから、操業因子の操業条件、及び、該操業因子の操業条件で得られた製品の品質を含むプロセスデータを収集し、データ蓄積手段に蓄積する手順と、
前記データ蓄積手段に蓄積されたプロセスデータから、所定の品質が所望の範囲内となる操業因子の操業条件の範囲を算出する操業範囲算出手順と、
前記データ蓄積手段から取得された、所定の品質を満足する製品である解析対象製品のプロセスデータに基づいて、操業因子と品質との関係式を確率分布相関解析法により求める確率分布相関解析手順と、
前記操業範囲算出手順により算出された操業因子の操業条件の範囲と、前記確率分布相関解析手順により求められた関係式とを用いて、操業因子の品質に対する影響度を算出する影響度算出手順と、
前記データ蓄積手段から、品質を改善したい製品である品質改善対象製品の操業因子の操業条件を取得する改善対象操業条件設定手順と、
前記確率分布相関解析手順により求められた関係式と、前記改善対象操業条件設定手順により取得された品質改善対象製品の操業因子の操業条件とを用いて、品質の改善度合いを算出する品質改善度算出手順と、
前記影響度算出手順により算出された影響度と、前記品質改善度算出手順により算出された品質の改善度合いとを用いて、操業因子の改善期待度を算出する改善期待度算出手順とを有することを特徴とする品質管理方法。 A quality control method for managing a plurality of qualities within a desired range for a product manufactured through a manufacturing process.
A procedure for collecting process data including the operating conditions of the operating factors and the quality of the product obtained under the operating conditions of the operating factors from the manufacturing process, and storing the data in the data storage means;
From the process data stored in the data storage means, an operating range calculation procedure for calculating the operating condition range of the operating factor that the predetermined quality is within a desired range;
A probability distribution correlation analysis procedure for obtaining a relational expression between an operation factor and quality by a probability distribution correlation analysis method based on process data of an analysis target product that is a product satisfying a predetermined quality acquired from the data storage means; ,
Using the range of operating conditions of the operating factor calculated by the operating range calculating procedure and the relational expression obtained by the probability distribution correlation analyzing procedure, an influence calculating procedure for calculating the influence on the quality of the operating factor; ,
From the data storage means , an improvement target operation condition setting procedure for obtaining operation conditions of an operation factor of a quality improvement target product that is a product whose quality is to be improved;
The quality improvement degree for calculating the degree of quality improvement using the relational expression obtained by the probability distribution correlation analysis procedure and the operation condition of the operation factor of the quality improvement target product acquired by the improvement target operation condition setting procedure Calculation procedure,
An improvement expectation degree calculation procedure for calculating an improvement expectation degree of an operation factor using the influence degree calculated by the influence degree calculation procedure and the quality improvement degree calculated by the quality improvement degree calculation procedure; Quality control method characterized by 製造プロセスの工程を経て製造される製品について複数の品質を所望の範囲内に管理するための品質管理装置としてコンピュータを機能させるためのプログラムであって、
前記製造プロセスから、操業因子の操業条件、及び、該操業因子の操業条件で得られた製品の品質を含むプロセスデータを収集するデータ入力手段と、
前記データ入力手段により入力されるプロセスデータを蓄積するデータ蓄積手段と、
前記データ蓄積手段に蓄積されたプロセスデータから、所定の品質が所望の範囲内となる操業因子の操業条件の範囲を算出する操業範囲算出手段と、
前記データ蓄積手段から取得された、所定の品質を満足する製品である解析対象製品のプロセスデータに基づいて、操業因子と品質との関係式を確率分布相関解析法により求める確率分布相関解析手段と、
前記操業範囲算出手段により算出された操業因子の操業条件の範囲と、前記確率分布相関解析手段により求められた関係式とを用いて、操業因子の品質に対する影響度を算出する影響度算出手段と、
前記データ蓄積手段から、品質を改善したい製品である品質改善対象製品の操業因子の操業条件を取得する改善対象操業条件設定手段と、
前記確率分布相関解析手段により求められた関係式と、前記改善対象操業条件設定手段により取得された品質改善対象製品の操業因子の操業条件とを用いて、品質の改善度合いを算出する品質改善度算出手段と、
前記影響度算出手段により算出された影響度と、前記品質改善度算出手段により算出された品質の改善度合いとを用いて、操業因子の改善期待度を算出する改善期待度算出手段としてコンピュータを機能させるためのプログラム。 A program for causing a computer to function as a quality control device for managing a plurality of qualities within a desired range for a product manufactured through a manufacturing process.
Data input means for collecting process data including the operating conditions of the operating factors and the quality of the product obtained under the operating conditions of the operating factors from the manufacturing process;
Data storage means for storing process data input by the data input means;
From the process data stored in the data storage means, an operation range calculation means for calculating a range of operation conditions of the operation factor that the predetermined quality is within a desired range;
A probability distribution correlation analysis means for obtaining a relational expression between the operating factor and the quality by a probability distribution correlation analysis method based on the process data of the analysis target product which is a product satisfying a predetermined quality obtained from the data storage means; ,
The degree of influence calculation means for calculating the degree of influence on the quality of the operation factor using the range of the operation condition of the operation factor calculated by the operation range calculation means and the relational expression obtained by the probability distribution correlation analysis means; ,
From the data storage means , improvement target operation condition setting means for obtaining the operation condition of the operation factor of the quality improvement target product that is a product whose quality is to be improved;
The quality improvement degree for calculating the degree of quality improvement using the relational expression obtained by the probability distribution correlation analysis means and the operation conditions of the operation factor of the quality improvement target product acquired by the improvement target operation condition setting means A calculation means;
Using the influence degree calculated by the influence degree calculation means and the quality improvement degree calculated by the quality improvement degree calculation means, the computer functions as an improvement expectation degree calculation means for calculating an improvement expectation degree of the operation factor Program to let you.
JP2007181206A 2007-07-10 2007-07-10 Quality control device, method and program Expired - Fee Related JP5020728B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2007181206A JP5020728B2 (en) 2007-07-10 2007-07-10 Quality control device, method and program

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2007181206A JP5020728B2 (en) 2007-07-10 2007-07-10 Quality control device, method and program

Publications (2)

Publication Number Publication Date
JP2009020600A JP2009020600A (en) 2009-01-29
JP5020728B2 true JP5020728B2 (en) 2012-09-05

Family

ID=40360211

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2007181206A Expired - Fee Related JP5020728B2 (en) 2007-07-10 2007-07-10 Quality control device, method and program

Country Status (1)

Country Link
JP (1) JP5020728B2 (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010225029A (en) * 2009-03-25 2010-10-07 Fuji Electric Holdings Co Ltd Manufacture history analysis support device and manufacture history analysis support method
JP5481560B2 (en) * 2010-07-30 2014-04-23 日立ビークルエナジー株式会社 Non-aqueous electrolyte secondary battery
JP5704040B2 (en) * 2011-10-06 2015-04-22 新日鐵住金株式会社 Product quality management method and product quality management device
JP7159758B2 (en) * 2018-09-27 2022-10-25 株式会社ジェイテクト Molding condition determination support device and injection molding machine

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3576335B2 (en) * 1996-10-31 2004-10-13 松下電器産業株式会社 Method and apparatus for extracting abnormalities in process processing steps
JP2000252179A (en) * 1999-03-04 2000-09-14 Hitachi Ltd Semiconductor manufacturing process stabilization support system
JP4102019B2 (en) * 2000-12-12 2008-06-18 新日本製鐵株式会社 Analysis apparatus, method, and computer-readable storage medium in manufacturing process
JP3733057B2 (en) * 2001-11-16 2006-01-11 新日本製鐵株式会社 Operation analysis apparatus in manufacturing process, method thereof, and computer-readable storage medium
JP4365600B2 (en) * 2002-03-08 2009-11-18 Ｊｆｅスチール株式会社 Steel product quality design apparatus and steel product manufacturing method
JP4299508B2 (en) * 2002-07-17 2009-07-22 新日本製鐵株式会社 Operation and quality related analysis device in manufacturing process, related analysis method, and computer-readable storage medium
JP2006318263A (en) * 2005-05-13 2006-11-24 Sharp Corp Information analysis system, information analysis method and program
JP4613751B2 (en) * 2005-08-26 2011-01-19 住友金属工業株式会社 Manufacturing condition calculation method, quality adjustment method, steel manufacturing method, manufacturing condition calculation device, quality adjustment system, and computer program
JP4681426B2 (en) * 2005-11-15 2011-05-11 新日本製鐵株式会社 Apparatus and method for analyzing relation between operation and quality in manufacturing process, computer program, and computer-readable recording medium

Also Published As

Publication number Publication date
JP2009020600A (en) 2009-01-29

Similar Documents

Publication Publication Date Title
JP4855353B2 (en) Product quality improvement condition analysis apparatus, analysis method, computer program, and computer-readable recording medium
JP5091604B2 (en) Distribution evaluation method, product manufacturing method, distribution evaluation program, and distribution evaluation system
JP5020728B2 (en) Quality control device, method and program
JP5604945B2 (en) Quality prediction apparatus, quality prediction method, computer program, and computer-readable recording medium
CN111242464A (en) Ceramic bathroom quality management method based on SPC
JP2009265699A (en) Processing time prediction device, method, program, and computer-readable storage medium
JP4365600B2 (en) Steel product quality design apparatus and steel product manufacturing method
JP5003362B2 (en) Product quality control method and control device
WO2018142678A1 (en) Method for visualizing manufacturing results, image processing device, and program
WO2019123792A1 (en) Service life evaluating device and service life evaluating method
JP5973380B2 (en) Production index extraction apparatus and program
JP3733057B2 (en) Operation analysis apparatus in manufacturing process, method thereof, and computer-readable storage medium
JP3738738B2 (en) Steel product quality control equipment
JP4623020B2 (en) Optimal quality design support device
JP5251890B2 (en) Operation analysis apparatus, operation analysis method, computer program, and computer-readable storage medium
KR20200007083A (en) Quality Analysis Device and Quality Analysis Method
JP2004288144A (en) Apparatus and method for analyzing operation result of manufacturing process, and computer-readable storage medium
JP7124321B2 (en) Information processing device, data management system, data management method and program
JP4910757B2 (en) Process data prediction system and process management apparatus using the same
JP2004277835A (en) System for supporting orders for product of steel material
US10706574B2 (en) Method and apparatus for identifying a gap between objects in an image
CN112016216B (en) Method for determining press-fitting curve inflection point
JP4102019B2 (en) Analysis apparatus, method, and computer-readable storage medium in manufacturing process
JP5220458B2 (en) Output value prediction method, apparatus, and program for the method
JP2009238190A (en) Maintenance plan support system

Legal Events

Date Code Title Description
A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20090916

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20110926

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20111004

A521 Request for written amendment filed

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20111201

TRDD Decision of grant or rejection written
A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

Effective date: 20120522

A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20120613

R151 Written notification of patent or utility model registration

Ref document number: 5020728

Country of ref document: JP

Free format text: JAPANESE INTERMEDIATE CODE: R151

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20150622

Year of fee payment: 3

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20150622

Year of fee payment: 3

S533 Written request for registration of change of name

Free format text: JAPANESE INTERMEDIATE CODE: R313533

FPAY Renewal fee payment (event date is renewal date of database)

Free format text: PAYMENT UNTIL: 20150622

Year of fee payment: 3

R350 Written notification of registration of transfer

Free format text: JAPANESE INTERMEDIATE CODE: R350

S533 Written request for registration of change of name

Free format text: JAPANESE INTERMEDIATE CODE: R313533

R350 Written notification of registration of transfer

Free format text: JAPANESE INTERMEDIATE CODE: R350

LAPS Cancellation because of no payment of annual fees