JP2004206309A - Process arrangement device, process arrangement method, and process arrangement program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To acquire an arrangement content of a plurality of workstations for producing products of a plurality of product numbers as a quasi-optimum or an optimum solution. <P>SOLUTION: The device is equipped with; an initial process arrangement data generation section 2 which assigns random values associated with one of a plurality of workstations to each of a plurality of work elements, a plurality of facility identifiers, and a plurality of workers identifiers acquired from a process arrangement data storage section 1, and finds the arrangement content of the plurality of workstations by the associated work elements, the facility identifiers and the workers identifiers as an initial solution of a preliminary solution; a simulated annealing execution section 3 which finds the optimum solution by generating a new preliminary solution from the preliminary solution by emergent computation; and a process arrangement data evaluation section 4 which finds process arrangement efficiencies, resource shortage numbers, frequencies of transfer between workstations, and excess periods of time, and calculates evaluation values by operating weighted with these factors. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

【００３２】
作業者データは、データテーブルによりなり、以下の（表２）に示すように、複数人の作業者をそれぞれ示す複数の作業者識別名（「作業者」欄のデータ）を含み、これら複数の作業者識別名毎に、当該作業者識別の作業者により実行可能な作業を示す作業要素（「可能作業」欄のデータ）が少なくとも１つ対応付けられている。なお、同表では、作業者が複数の作業要素の作業をこなすことができるようになっている。
【００３３】
【表２】

【００３４】
生産品データは、複数（ここでは２つ）のデータテーブルによりなり、以下の（表３），（表４）に示すように、複数の品番｛（表３）では“Ｘ”，（表４）では“Ｙ”｝別に、当該品番の生産品を１台生産するために必要な複数の作業をそれぞれ示す複数の作業要素毎に、当該作業要素に対して少なくとも１つの他の作業要素が先行関係にあればそれを示す先行要素と、当該作業要素の作業に必要な作業時間と、当該作業要素の作業に作業者が必要であればその作業者数と、当該作業要素の作業に設備が必要であればその設備識別名（「必要設備」欄のデータ）とが対応付けられている。
【００３５】
【表３】

【００３６】
【表４】

【００３７】
ここで、作業要素は、「ねじ締め」「品質チェック」などの生産品の加工内容を意味する。（表３）の作業要素“Ｂ”は、先行要素“Ａ”が完了してから行われることを意味する。また、（表３）から、図２に示すようないわゆるプロジェクトのネットワークを得ることができ、（表４）から、図３に示すようなプロジェクトのネットワークを得ることができる。
【００３８】
入力データは、データテーブルによりなり、以下の（表５）に示すように、生産するべき複数の品番の生産品に対する品番毎に、当該品番の生産品の生産数が対応付けられ、また、それら複数の品番の生産品に対する操業時間と、ワークステーション数と、焼鈍し法温度設定値と、この低温化用の焼鈍し法温度更新係数設定値と、焼鈍し法計算時間設定値とを含んでいる。
【００３９】
【表５】

【００４０】
そして、上記各種データのうち、例えば入力データは、入力装置を介して入力され（図４のＳ１０参照）、工程編成データ記憶部１に記憶される一方、その他のデータは、予め作成され、工程編成データ記憶部１に記憶されている。
【００４１】
図１に戻って、初期工程編成データ生成部２は、工程編成データ記憶部１の入力データから得られる複数の品番を基に、工程編成データ記憶部１の生産品データから該当する各品番の複数の作業要素を読み出すとともに、工程編成データ記憶部１の設備データおよび作業者データからそれぞれ複数の設備識別名および複数の作業者識別名を読み出し、これら複数の作業要素、複数の設備識別名および複数の作業者識別名の個々に対して、上記ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値（ここではワークステーション数内のランダム値）を割り当て、それらランダム値で関連付けられる作業要素、設備識別名および作業者識別名により複数のワークステーションの編成内容を求めて暫定解の初期解とするものである（図４のＳ１１，Ｓ１２参照）。
【００４２】
より具体的には、品番“Ｘ”の生産品については、（表３）のデータテーブルから、以下の（表６）に示すように、作業要素と先行要素とが読み出され、それぞれについてランダム値（「ワークステーション」）が割り当てられる。
【００４３】
【表６】

【００４４】
同様に、品番“Ｙ”の生産品については、（表４）のデータテーブルから、以下の（表７）に示すように、作業要素と先行要素とが読み出され、それぞれについてランダム値が割り当てられる。なお、（表６），（表７）における「移動の有無」欄のデータは、工程編成データ評価部４によって求められる。
【００４５】
【表７】

【００４６】
そして、（表１）のデータテーブルから、以下の（表８）に示すように、設備識別名（「設備」）と保有台数（「台数」）とが読み出され、それぞれの保有台数１台当たりの設備識別名についてランダム値が割り当てられ、（表２）のデータテーブルから、以下の（表９）に示すように、作業者識別名（「作業者」）と作業要素（「可能作業」）とが読み出され、それぞれについてランダム値が割り当てられることにより、（表６）〜（表９）のデータテーブルから成り、図５の編成内容を示す初期解が生成される。ただし、品番“Ｘ”と品番“Ｙ”の作業要素はグループ化され、同一作業要素に対してランダム値が割り当てられている。
【００４７】
【表８】

【００４８】
【表９】

【００４９】
図５において、例えば、品番“Ｘ”の生産品に必要な作業要素“Ｅ”は、作業時間が９分であり、その生産数が６０台であるので、作業要素“Ｅ”の合計作業時間は、９時間（＝９［分／台］×６０［台］）となる。品番“Ｘ”，“Ｙ”の生産品に必要な作業要素“Ａ”は、作業時間が５分であり、それら生産数が１２０台であるので、作業要素“Ａ”の合計作業時間は、１０時間（＝５［分／台］×１２０［台］）となる。
【００５０】
なお、初期工程編成データ生成部２は、複数の設備識別名および複数の作業者識別名を読み出す場合、生産するべき複数の品番の生産品に必要な複数の作業要素に対応する複数の設備識別名および複数の作業者識別名を読み出すような構成でもよい。
【００５１】
図１に示す焼鈍し法実行部３は、暫定解から新たな暫定解を生成しながら最適解を求めるものであり、進化型計算(Evolutionary Computation)および焼鈍し法(Simulated Annealing) などの創発計算(Emaergent Computation) のうち、焼鈍し法によって最適解を求める。なお、この焼鈍し法実行部３の詳細については、次の工程編成データ評価部４の後で説明する。
【００５２】
工程編成データ評価部４は、暫定解の複数のワークステーションに対して、工程編成データ記憶部１の生産品データ内の作業時間を参照して工程編成効率を求め、工程編成データ記憶部１の少なくとも生産品データ内の作業要素に対する設備識別名を参照して資源不足数を求め、工程編成データ記憶部１の生産品データ内の先行要素からワークステーション間移動回数を求め、工程編成データ記憶部１の入力データ内の操業時間に対する各ワークステーションの超過時間を求め、これらを重み付けした演算により評価値を算出するものであって、焼鈍し法実行部３の創発計算に利用される。
【００５３】
ここで、上記工程編成効率は、Σ（各ワークステーションの合計作業時間）／｛（最大合計作業時間）×（合計作業時間が零でないワークステーション数）｝の演算式で算出される。例えば、図４の初期解の工程編成効率は、０．６５｛＝（１５＋１６＋２４＋１０＋１３＋１５）／（２４×６）｝となる。
【００５４】
資源不足数は、設備および作業者の不足数を示し、不足であれば“１”、そうでなければ“０”となる。図５の例では、作業者識別名“ＳＧ７”の作業者に対して“ワークステーション６”が割り当てられているが、その作業者は、“ワークステーション６”に割り当てられている作業要素“Ｍ”の作業を実行できないので、“ワークステーション６”に対する資源不足数は、“１”となる。
【００５５】
ワークステーション間移動回数は、ある作業要素のワークステーションに対する先行要素における作業要素のワークステーションの相違数で求められ、「移動の有無」欄に入力される。例えば、（表６）において、作業要素“Ｂ”のワークステーション間移動回数は、先行要素内の作業要素“Ａ”の“ワークステーション１”と同じであるので、“０”となる。作業要素“Ｄ”のワークステーション間移動回数は、先行要素内の作業要素“Ｂ”の“ワークステーション１”と異なる“ワークステーション４”であるので、“１”となる。作業要素“Ｍ”のワークステーション間移動回数は、先行要素内の作業要素“Ｊ”，“Ｋ”，“Ｌ”の“ワークステーション５”，“ワークステーション５”，“ワークステーション３”と異なる“ワークステーション６”であるので、“３”となる。
【００５６】
超過時間は、ワークステーション毎に求められ、操業時間の２０時間に対する当該ワークステーションの合計作業時間の超過分となる。例えば、“ワークステーション３”の合計作業時間が２４時間であるので、その超過時間は“４”となる。
【００５７】
評価値は、工程編成効率、資源不足数、ワークステーション間移動回数および超過時間の重み付き線形和として算出される。例えば、評価値＝１００×（工程編成効率）＋（−１）×（資源不足数）＋（−１）×（超過時間）＋（−１）×（ワークステーション間移動回数）の重み付き線形和を使用したとすれば、図５の初期解の評価値は、３６（＝１００×０．６５−１−４−１４）となる。
【００５８】
ここで、焼鈍し法実行部３について詳述すると、焼鈍し法実行部３は、工程編成データ評価部４を利用して、初期解とした暫定解の評価値Ｃを算出し（図４のＳ１３参照）、この後、焼鈍し法処理に移行する。
【００５９】
すなわち、暫定解の複数のワークステーションに対して、作業要素、設備識別名および作業者識別名を関連付けている各ランダム値の一部に代えて、ワークステーション数で決まる複数のワークステーションのいずれかに関連付ける新たなランダム値を割り当てることにより、新たな暫定解の候補として複数のワークステーションの編成内容を求める処理と（Ｓ１４）、工程編成データ評価部４を利用して、候補としての暫定解の評価値Ｃ’を算出し（Ｓ１５）、この評価値Ｃ’が前回求めた現在の暫定解の評価値Ｃよりも評価の良い値であれば（Ｓ１６）、候補を新たな暫定解とし（Ｓ１７）、そうでなければ、両評価値の各々を、少ないほど評価が良いエネルギー量に見立て、これら両エネルギー量の差分ΔＣを焼鈍し法温度設定値Ｔで除して除算値を求め、この除算値のマイナス値を変数として指数関数で演算することにより確率を求め、この確率に従って、候補を新たな暫定解とするかまたは現在の暫定解のままとする（Ｓ１８，Ｓ１９）処理とが、所定の探索条件を満足するまで繰り返し実行される。なお、上記各ランダム値の一部は、乱数を用いて選定するようにしてもよく、ワークステーション単位の評価値が最悪値から所定番目の値までの劣悪範囲内に属する各ワークステーションからランダムに選定するようにしてもよい。
【００６０】
そして、探索条件を満足すれば、焼鈍し法温度設定値Ｔを、これに焼鈍し法温度更新係数設定値αを乗ずることにより更新しながら（Ｓ２０）、所定の停止条件を満足するまで上記焼鈍し処理が繰り返される（Ｓ２１）。
【００６１】
工程編成データ出力部５は、焼鈍し法実行部３によって求められた最適解を表示装置に表示出力したり、プリンタなどに印刷出力したり、記憶装置に書込み出力したりするものである。
【００６２】
次に、上記構成の工程編成装置の動作について説明すると、設計者が表示装置に表示された工程編成用の画面を見ながら、入力装置を介して入力データを入力すると（図４のＳ１０）、初期工程編成データ生成部２によって図５の例の編成内容を示す初期解が生成され（Ｓ１１）、その評価値Ｃが工程編成データ評価部４によって算出される（Ｓ１３）。
【００６３】
この後、焼鈍し法実行部３による焼鈍し法処理に移行し、図６〜図９の各例に示すように、暫定解の複数のワークステーションに対して、作業要素、設備識別名および作業者識別名を関連付けている各ランダム値の一部に代えて、ワークステーション数で決まる複数のワークステーションのいずれかに関連付ける新たなランダム値を割り当てることにより、新たな暫定解の候補として複数のワークステーションの編成内容を求める処理が実行される（Ｓ１４）。
【００６４】
続いて、候補としての暫定解の評価値Ｃ’を算出し（Ｓ１５）、評価値Ｃ’が前回求めた現在の暫定解の評価値Ｃよりも評価の良い値であれば（Ｓ１６でＹＥＳ）、候補を新たな暫定解として記憶し（Ｓ１７）、ステップＳ２０に進む。
【００６５】
一方、評価値Ｃ’が前回求めた現在の暫定解の評価値Ｃよりも評価の良い値でなければ（Ｓ１６でＮＯ）、０≦Ｐ≦１範囲内で乱数値Ｐを生成し（Ｓ１８）、両評価値の各々を、少ないほど評価が良いエネルギー量に見立て、これら両エネルギー量の差分ΔＣを焼鈍し法温度設定値Ｔで除して除算値を求め、この除算値のマイナス値（−ΔＣ／Ｔ）を変数として指数関数｛ｅｘｐ（−ΔＣ／Ｔ）｝で演算することにより確率を求め、この確率に従って、候補を新たな暫定解とするか（Ｓ１９でＹＥＳ）、または現在の暫定解のままとする（Ｓ１９でＮＯ）処理が実行される。図４では、ｅｘｐ（−ΔＣ／Ｔ）の演算値が乱数値Ｐ以下であればステップＳ１７に進み、そうでなければステップＳ２０に進むようになっている。
【００６６】
上記焼鈍し法処理は、所定の探索条件、例えば所定の時間経過するまで、繰り返し実行され、探索条件を満足すれば、焼鈍し法温度設定値Ｔを、これに焼鈍し法温度更新係数設定値αを乗ずることにより更新しながら（Ｓ２０）、所定の停止条件、例えば、資源不足数および超過時間が零となり、焼鈍し法温度設定値Ｔが焼鈍し法計算時間設定値を超えるまで、繰り返し実行される。なお、上記探索条件は、所定の時間に限らず、所定の回数、所定の候補受理ないし棄却回数でもよい。また、上記停止条件は、アニーリングを定めた回数、受理がほとんど起こらなくなった回数、同じ状態が何度も生成される状態になる条件でもよい。
【００６７】
続いて、焼鈍し法実行部３によって求められた最適解の作業要素、設備識別名および作業者識別名などが、工程編成データ出力部５によって出力される。
【００６８】
ここで、上記図６の候補は、図５における初期解の複数のワークステーションに対して、作業要素“Ｅ”，“Ｉ”，“Ｋ”，“Ｌ”、設備識別名“Ｓ２”の設備１台，設備識別名“Ｓ３”の設備２台および作業者識別名“ＳＧ３”を関連付けている各ランダム値に代えて、新たなランダム値を割り当てることにより生成されたものである。そして、図６の候補としての暫定解は、工程編成効率が０．７４｛＝９３／（１８×７）｝、資源不足数が１、超過時間が０、ワークステーション間移動回数が１６であるので、その評価値Ｃ’は５７となる。この評価値Ｃ’は現在の暫定解である初期解の評価値Ｃ（＝３６）よりも良いので、ステップＳ１７に進み、図６の候補が暫定解となる。
【００６９】
図７の候補は図６の暫定解から生成され、工程編成効率が０．８６｛＝９３／（１８×６）｝、資源不足数が４、超過時間が０、ワークステーション間移動回数が１６であるので、その候補の評価値Ｃ’は６６となる。そして評価値Ｃ’は現在の暫定解の評価値Ｃ（＝５７）よりも良いので、ステップＳ１７に進み、図７の候補が暫定解となる。
【００７０】
図８の候補は図７の暫定解から生成され、工程編成効率が０．７３｛＝９３／（１６×８）｝、資源不足数が３、超過時間が０、ワークステーション間移動回数が１９であるので、その候補の評価値Ｃ’は５１となる。そして評価値Ｃ’は現在の暫定解の評価値Ｃ（＝６６）よりも良くないので、ステップＳ１８に進むことになる。そして、ステップＳ１８で、例えば０．２５の乱数値Ｐが生成されたとすれば、ｅｘｐ（−ΔＣ／Ｔ）の演算値が０．８６｛＝ｅｘｐ（−｜５１−６６｜／１００）｝となり、乱数値Ｐ以下とならないので、図８の候補は棄却され、ステップＳ２０に進む。
【００７１】
次に、図９の候補が図７の暫定解から生成され、工程編成効率が０．８６｛＝９３／（１８×６）｝、資源不足数が０、超過時間が０、ワークステーション間移動回数が１５であるので、その候補の評価値Ｃ’は５１となる。そして評価値Ｃ’は現在の暫定解の評価値Ｃ（＝６６）よりも良くないので、ステップＳ１８に進むことになる。そして、ステップＳ１８で、例えば０．９０の乱数値Ｐが生成されたとすれば、ｅｘｐ（−ΔＣ／Ｔ）の演算値が０．８６となり、乱数値Ｐ以下となるので、ステップＳ１７に進み、図９の候補が暫定解となり、局所解に陥ることを防止する。
【００７２】
このように最適解探索を続行し、最終的な暫定解を（準）最適解とする。なお、上記処理において、暫定解の少なくとも評価値Ｃの一番高いものを記憶して残すようにし、最終的な一連の最適解探索処理が終了したとき、それまでに記憶した暫定解の評価値Ｃの一番高いものを（準）最適解とすることもできる。
【００７３】
以上、第１実施形態によれば、作業要素、設備識別名および作業者識別名に対して、複数のワークステーションのいずれかに関連付けるランダム値を割り当てて、創発計算で最適化問題を解くことにより、局所解に陥ることなく、複数の品番の生産品を生産するための複数のワークステーションの編成内容を（準）最適解として求めることができる。
【００７４】
（第２実施形態）
本発明による第２実施形態の工程編成装置は、第１実施形態において、工程編成データ評価部４が、工程編成効率を求める場合、工程編成データ記憶部１の生産品データ内の作業時間を参照して、生産するべき複数の品番の生産品に対する品番毎に工程編成効率を求めることを特徴とする。
【００７５】
例えば、図５に示した初期解の場合、品番“Ｘ”の生産品のみを生産する部分を抜き出したものが図１０であり、品番“Ｙ”の生産品のみを生産する部分を抜き出したものが図１１となる。この場合、品番“Ｘ”の工程編成効率は、（表６）に示した作業要素に対するワークステーションの作業時間から求められ、０．４２｛＝（７．５＋２４＋５＋６．５＋７．５）／（２４×５）｝となる。
【００７６】
品番“Ｙ”の工程編成効率も、（表７）に示した作業要素に対するワークステーションの作業時間から求められ、０．５３｛＝７．５＋１６＋５＋６．５＋７．５）／（１６×５）｝となる。
【００７７】
そして、それら品番別の工程編成効率、資源不足数、ワークステーション間移動回数および超過時間の重み付き線形和として、評価値が算出される。
【００７８】
単一品番の生産品をまとめて生産する場合（ロット生産など）や、各品番の生産品の生産量比率が一定ではなく変動する場合は、１品番の生産品だけが工程に流れる時間が長くなったりして変動するが、第２実施形態によれば、品番別に工程編成効率を算出して、各ワークステーションの負荷の平準化を行うので、単一品番の生産品のみを生産する場合でも、編成効率の良い効率的な工程編成を行うことができる。
【００７９】
（第３実施形態）
本発明による第３実施形態の工程編成装置は、第１実施形態において、初期工程編成データ生成部２および焼鈍し法実行部３が、ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当てる場合、作業要素に対して類似する作業要素に同一のランダム値を割り当てることを特徴とする。
【００８０】
（表３），（表４）に示した品番“Ｘ”，“Ｙ”の生産品の場合、これらに必要な作業要素は、“Ａ”，“Ｂ”，“Ｄ”，“Ｇ”，“Ｋ”，“Ｊ”，“Ｍ”である。ここで、多品種生産品の生産ラインでは、同一作業要素の作業を同じ場所で実行する方が管理が容易となる。つまり、品番“Ｘ”の作業要素“Ａ”と品番“Ｙ”の作業要素“Ａ”は、同じワークステーションでさばく方が工程の管理が容易であるので、品番“Ｘ”の作業要素“Ａ”と品番“Ｙ”の作業要素“Ａ”を常に同じワークステーションに割り当てるようにするのである。
【００８１】
第３実施形態によれば、作業要素が類似する品番の作業要素をグループ化することにより、焼鈍し法実行時の組合せ数が減り、高速に工程編成を行うことができる。また、品番が異なっていても同一のワークステーションで行いたい作業要素がある場合に意図した工程編成を行うことができる。
【００８２】
（第４実施形態）
本発明による第４実施形態の工程編成装置は、第３実施形態において、初期工程編成データ生成部２および焼鈍し法実行部３が、作業要素に対して類似する作業要素に同一のランダム値を割り当てる場合、類似する作業要素で決まる作業時間の合計時間が操業時間を超えなければ、その類似する作業要素に同一のランダム値を割り当てることを特徴とする。
【００８３】
第４実施形態でも、例えば、品番“Ｘ”の作業要素“Ａ”と品番“Ｙ”の作業要素“Ａ”を常に同じワークステーションに割り当てるようにするのであるが、このような処理は、それらの作業要素の合計作業時間が操業時間の２０時間を超えない場合に実行されるのである。本例の処理は、合計作業時間が１０時間（＝５［分／台］×１２０［台］）となり２０時間を超えないので、実行されることになる。
【００８４】
第４実施形態によれば、内容が類似する作業要素をグループ化する場合、合計作業時間が操業時間を超えることがないように、工程編成を行うことができる。
【００８５】
（第５実施形態）
本発明による第５実施形態の工程編成装置は、第１実施形態において、焼鈍し法実行部３が、新たな暫定解の候補として複数のワークステーションの編成内容を求める場合、先ず、設備識別名および作業者識別名に対して新たなランダム値を割り当て、続いてランダム値が割り当てられた設備識別名および作業者識別名と整合するように作業要素に対して新たなランダム値を割り当てることを特徴とする。
【００８６】
（表６）〜（表９）に示した各作業要素、各設備識別名および各作業者識別名の個々に対するワークステーションの割当てにおいて、先ず、設備識別名および作業者識別名に対して新たなランダム値を割り当て、続いて設備識別名および作業者識別名に対するランダム値（ワークステーション番号）を制約として、作業要素に対して新たなランダム値を割り当てる。例えば、品番“Ｘ”，“Ｙ”の作業要素“Ａ”は、設備識別名“Ｓ１”の設備を必要とするので、作業要素“Ａ”に対しては、設備識別名“Ｓ１”の“ワークステーション１”しか割り当てることができないようにするのである。
【００８７】
第５実施形態によれば、設備および作業者を必要とする作業要素が、設備および作業者が割り付いたワークステーションに割り付くため、工程編成データ評価部４によって求められる資源不足数を零にすることができ、実行可能解を確実に得ることができる。
【００８８】
（第６実施形態）
本発明による第６実施形態の工程編成装置は、第１実施形態において、焼鈍し法実行部３によって複数の最適解を求め、工程編成データ出力部５から出力することを特徴とする。
【００８９】
つまり、複数の準最適解を提案することで、最終的な配置案の選択に柔軟性を持たせるのである。焼鈍し法実行部３によって求められる最終的な解は、初期解と探索に依存しているため、初期解が同じでも例えば探索回数により解が変動する。もちろん初期解が異なっても最終的な解は変動する。第６実施形態では、その性質を利用し、焼鈍し法実行部３において、例えば複数の初期解から複数の最終的な解を得て、工程編成データ出力部５から出力するように構成される。
【００９０】
最適解を１つだけ出力する構成では、レイアウトなどの関係で実行不可能となる場合があるが、第６実施形態によれば、複数案を提示するので、実行可能な案を得られやすくすることができる。
【００９１】
（第７実施形態）
本発明による第７実施形態の工程編成装置は、第１実施形態において、工程編成データ記憶部１に記憶されている入力データのワークステーション数が、生産するべき複数の品番の生産品に対する全作業要素の合計作業時間を操業時間で除して切り上げた値に余裕率を乗じることにより自動的に設定されることを特徴とする。
【００９２】
（表３），（表４）に示した品番“Ｘ”，“Ｙ”の作業時間と（表５）に示した品番別の生産数とから、品番“Ｘ”，“Ｙ”の合計作業時間として９３時間が得られる。そして、合計作業時間の９３時間を操業時間の２０時間で除し、得られた値４．６（＝９３÷２０）を切り上げると５となり、これに例えば１．５の余裕率を乗じれば、７．５（＝５×１．５）が得られる。この７．５を例えば切り上げて得られる値８がワークステーション数として自動的に設定される。
【００９３】
第７実施形態によれば、適切なワークステーション数を自動的に設定することができ、最適な工程編成を行うことができる。
【００９４】
（第８実施形態）
本発明による第８実施形態の工程編成装置は、第１実施形態において、焼鈍し法実行部３が、停止条件を満足しても、超過時間が零にならなければ、ワークステーション数を所定値増分し、再度、創発計算によって暫定解から新たな暫定解を生成しながら最適解を求めることを特徴とする。
【００９５】
設定されたワークステーション数（＝８）で、実行可能な工程編成結果、すなわち資源不足数および超過時間が零となる工程編成結果が得られない場合、そのワークステーション数を所定の値だけ増分して９などの値に変更し、再度、創発計算によって暫定解から新たな暫定解を生成しながら最適解を求めるのである。
【００９６】
第８実施形態によれば、設定されたワークステーション数が適切でない場合でも、最適な工程編成を行うことができる。
【００９７】
（第９実施形態）
図１２，図１３は本発明による第９実施形態の工程編成装置にて生成される暫定解を求める際の前処理の例を示す図である。
【００９８】
第９実施形態の工程編成装置は、第１実施形態において、初期工程編成データ生成部２および焼鈍し法実行部３が、複数のワークステーションの編成内容を求める場合、前処理として、生産するべき複数の品番の生産品に対する品番別に、複数のワークステーションの編成内容を求めて、各ワークステーションに配分された作業要素のグループを創発計算で使用する作業要素とすることを特徴とする。
【００９９】
前処理として、品番“Ｘ”の作業要素については、図１２の例に示すように、作業要素（Ａ，Ｂ），（Ｅ），（Ｈ），（Ｄ，Ｇ，Ｌ），（Ｋ，Ｊ），（Ｍ）の６つにグループ化するとともに、品番“Ｙ”の作業要素については、図１３の例に示すように、作業要素（Ａ，Ｂ），（Ｃ，Ｆ），（Ｉ），（Ｄ，Ｇ），（Ｋ，Ｊ），（Ｍ）の６つにグループ化するのである。そして、品番別にグループ化された単位で、ワークステーションに割り付けて工程編成を行う。
【０１００】
第９実施形態によれば、品番別に予め工程編成を行って、各品番別に作業要素をグループ化することで、高速に工程編成を行うことができ、また品番別に編成効率の良い工程編成が得られる。また、ワークステーション間移動回数を少なくすることができる。
【０１０１】
（第１０実施形態）
本発明による第１０実施形態の工程編成装置は、第１実施形態において、以下の（表１０）に示すように、工程編成データ記憶部１に記憶されている入力データ内のワークステーション数で決まる複数のワークステーションおよび工程編成データ記憶部１に記憶されている生産品データ内の複数の作業要素が、複数の工程のいずれかに対応付けられ、初期工程編成データ生成部２および焼鈍し法実行部３が、作業要素に対して、ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当てる場合、その作業要素に対応付けられた工程と同じ工程が対応付けられたワークステーションに関連付けるランダム値を割り当てることを特徴とする。
【０１０２】
【表１０】

【０１０３】
つまり、工程に対してワークステーションおよび作業要素を割り当てるのであるが、この場合、作業要素とワークステーションの工程が同じ組合せを見つけ、工程が同じである場合のみ割り当てを許す。例えば、作業要素“Ｆ”には、“工程２”が対応付けられているので、“工程２”が対応付けられた“ワークステーション４”〜“ワークステーション６”のいづれかに関連付けるランダム値が割り当てられることになる。
【０１０４】
第１０実施形態によれば、各ワークステーションと各作業要素が工程に対応付けられることで、各作業要素に割り当て可能なワークステーションの候補数が少なくなり、工程編成を高速に算出することができる。また、全品番の作業要素のワークステーション間の移動の仕方が似通ったものとなり、ワークの流れ方が、品番によって逆行することがなくなる。
【０１０５】
（第１１実施形態）
本発明による第１１実施形態の工程編成装置は、第１実施形態において、初期工程編成データ生成部２および焼鈍し法実行部３が、作業要素に対して、ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当てる場合、当該作業要素に対応付けられた先行要素で示される他の作業要素があれば、この他の作業要素に割り当てられたランダム値で関連付けられるワークステーションと同じか後段のワークステーションに関連付けるランダム値を割り当てることを特徴とする。
【０１０６】
ここでは、先行要素の作業要素に割り当てられたワークステーション番号以上のワークステーション番号に関連付けるランダム値が割り当てられる。例えば、以下の（表１１）の例における作業要素“Ｂ”に対しては、その先行要素が“ワークステーション１”に割り当てられた作業要素“Ａ”であるので、“ワークステーション１”以降のワークステーションに関連付けるランダム値が割り当てられる。
【０１０７】
【表１１】

【０１０８】
（表１１）では、“ワークステーション１”に関連付けるランダム値が割り当てられている。同様に、作業要素“Ｇ”に対しては、その先行要素が“ワークステーション３”に割り当てられた作業要素“Ｄ”であるので、“ワークステーション３”以降のワークステーションに関連付けるランダム値が割り当てられる。（表１１）では、“ワークステーション４”に関連付けるランダム値が割り当てられている。
【０１０９】
第１１実施形態によれば、ワークの流れ方が、品番によって逆行することがなくなる。
【０１１０】
（第１２実施形態）
図１４，図１５は本発明による第１２実施形態の工程編成装置にて生成される暫定解の例を示す図である。
【０１１１】
第１２実施形態の工程編成装置は、第１実施形態において、焼鈍し法実行部３が、ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当てる場合、工程編成データ評価部４を利用して現在の暫定解の評価値を算出する際に資源不足数が零であったとき、現在の暫定解において合計作業時間がより長いワークステーションに関連付けられた作業要素のうち、設備が必要でない作業要素があればこれを別のワークステーションに関連付けるようにランダム値を割り当てることを特徴とする。
【０１１２】
図１４の例において、各作業要素のうち、作業要素“Ａ”，“Ｃ”，“Ｅ”，“Ｉ”，“Ｌ”がそれぞれ“Ｓ１”，“Ｓ２”，“Ｓ２”，“Ｓ３”，“Ｓ３”の設備識別名の設備を必要とする。この場合、合計作業時間がより長いワークステーションのうち、例えば“ワークステーション３”に割り当てられた作業要素“Ｅ”，“Ｈ”，“Ｌ”のうち、“Ｅ”，“Ｌ”が設備を必要とし、“Ｈ”が設備を必要としないので、図１５の例に示すように、作業要素“Ｈ”に対して、別のワークステーションに関連付けるようにランダム値が割り当てられる。図１５の例では、“ワークステーション４”に関連付けるランダム値が割り当てられている。
【０１１３】
第１２実施形態によれば、編成効率の高い実行可能解を得ることができるようになる。
【０１１４】
（第１３実施形態）
本発明による第１３実施形態の工程編成装置は、第１２実施形態において、現在の暫定解において合計作業時間が最長のワークステーションに関連付けられた作業要素のうち、設備が必要でない作業要素があればこれを別のワークステーションに関連付けるようにランダム値を割り当てることを特徴とする。
【０１１５】
第１３実施形態によっても、編成効率の高い実行可能解を得ることができるようになる。
【０１１６】
（第１４実施形態）
本発明による第１４実施形態の工程編成装置は、第１２実施形態において、現在の暫定解において合計作業時間が最長のワークステーションに関連付けられた作業要素のうち、設備が必要でない作業要素が複数あればこれら全てを別のワークステーションに関連付けるようにランダム値を割り当てることを特徴とする。
【０１１７】
第１４実施形態によっても、編成効率の高い実行可能解を得ることができるようになる。
【０１１８】
（第１５実施形態）
図１６は本発明による第１５実施形態の工程編成装置にて生成される暫定解の例を示す図である。
【０１１９】
第１５実施形態の工程編成装置は、第１実施形態において、焼鈍し法実行部３が、ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当てる場合、工程編成データ評価部４を利用して現在の暫定解の評価値を算出する際に資源不足数が零であったとき、現在の暫定解において合計作業時間がより長いワークステーションに関連付けられた作業要素のうち、設備が必要である作業要素があればこれら設備および作業要素を別のワークステーションに関連付けるようにランダム値を割り当てることを特徴とする。
【０１２０】
図１４の例の場合、合計作業時間がより長いワークステーションのうち、例えば“ワークステーション３”に割り当てられた作業要素“Ｅ”，“Ｈ”，“Ｌ”のうち、“Ｅ”，“Ｌ”が設備を必要とし、“Ｈ”が設備を必要としないので、少なくともいずれかの設備が必要である作業要素およびその設備の設備識別名に対して、別のワークステーションに関連付けるようにランダム値が割り当てられる。図１６の例では、作業要素“Ｅ”および設備識別名“Ｓ２”に対して、“ワークステーション４”に関連付けるランダム値が割り当てられている。
【０１２１】
第１５実施形態によれば、負荷バランスのとれた暫定解が得られ、編成効率の高い実行可能解を得ることができるようになる。
【０１２２】
（第１６実施形態）
本発明による第１６実施形態の工程編成装置は、第１５実施形態において、現在の暫定解において合計作業時間が最長のワークステーションに関連付けられた作業要素のうち、設備が必要である作業要素があればこれら設備および作業要素を別のワークステーションに関連付けるようにランダム値を割り当てることを特徴とする。
【０１２３】
第１６実施形態によっても、編成効率の高い実行可能解を得ることができるようになる。
【０１２４】
（第１７実施形態）
本発明による第１７実施形態の工程編成装置は、第１５実施形態において、現在の暫定解において合計作業時間がより長いワークステーションに関連付けられた作業要素のうち、設備が必要である作業要素が複数あればこれら複数の設備および作業要素の全てを別のワークステーションに関連付けるようにランダム値を割り当てることを特徴とする。
【０１２５】
図１４の例の場合、作業要素“Ｌ”および設備識別名“Ｓ３”と作業要素“Ｅ”および設備識別名“Ｓ２”とに対しても、別のワークステーションに関連付けるようにランダム値が割り当てられる。
【０１２６】
第１７実施形態によっても、編成効率の高い実行可能解を得ることができるようになる。
【０１２７】
（第１８実施形態）
図１７，図１８は本発明による第１８実施形態の工程編成装置にて生成される暫定解の例を示す図である。
【０１２８】
第１８実施形態の工程編成装置は、第１実施形態において、焼鈍し法実行部３が、ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当てる場合、工程編成データ評価部４を利用して現在の暫定解の評価値を算出する際に資源不足数が零であったとき、現在の暫定解において合計作業時間が最短のワークステーションに関連付けられた作業要素のうち、設備が必要でない作業要素があればそれ単独で、設備が必要である作業要素があれば設備とともに別のワークステーションに関連付けるようにランダム値を割り当てることを特徴とする。
【０１２９】
図１７の例において、各作業要素のうち、作業要素“Ａ”，“Ｃ”，“Ｅ”，“Ｉ”，“Ｌ”がそれぞれ“Ｓ１”，“Ｓ２”，“Ｓ２”，“Ｓ３”，“Ｓ３”の設備識別名の設備を必要とする。この場合、合計作業時間が最短の“ワークステーション５”に割り当てられた作業要素のうち、設備が必要でない作業要素があればそれ単独で、設備が必要である作業要素があれば設備とともに別のワークステーションに関連付けるようにランダム値を割り当てることになる。図１７の例では、作業要素“Ｊ”のみであり、設備が必要でないので、図１８の例に示すように、それ単独で別の“ワークステーション１”に関連付けるランダム値が割り当てられる。
【０１３０】
第１８実施形態によれば、ワークステーション数を減らすことができるので、稼動率の高い実行可能解を得ることができるようになる。また、負荷バランスのとれた解となり、より早い計算時間で解を得ることができる。
【０１３１】
（第１９実施形態）
本発明による第１９実施形態の工程編成装置は、第１実施形態において、初期工程編成データ生成部２および焼鈍し法実行部３が、作業要素に対して、ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当てる場合、その作業要素に必要な設備があればこれら作業要素および設備に同一のランダム値を割り当てることを特徴とする。
【０１３２】
つまり、設備を必要とする作業要素とその設備を一つのグループとして扱うのである。例えば、以下の（表１２）に示すような設備識別名（設備）と作業要素（可能作業）の各組合せをそれぞれ一つのグループ作業要素とみなして初期工程編成を行い、次いでグループ作業要素単位で、焼鈍し法実行部３を実行させるのである。
【０１３３】
【表１２】

【０１３４】
第１９実施形態によれば、グループ化することにより、高速に工程編成を行うことができ、またワークステーション割当領域において作業要素が必要とする設備の不足がない解を生成できる。
【０１３５】
【発明の効果】
以上のことから明らかなように、請求項１記載の発明の工程編成装置は、複数種類の設備をそれぞれ示す複数の設備識別名毎に、当該設備識別名の設備の保有台数が対応付けられるとともに、当該設備識別名の設備により実行可能な作業を示す作業要素が少なくとも１つ対応付けられた設備データと、複数人の作業者をそれぞれ示す複数の作業者識別名を含む作業者データと、複数の品番別に、当該品番の生産品を１台生産するために必要な複数の作業をそれぞれ示す複数の作業要素毎に、当該作業要素に対して少なくとも１つの他の作業要素が先行関係にあればそれを示す先行要素と、当該作業要素の作業に必要な作業時間と、当該作業要素の作業に作業者が必要であればその作業者数と、当該作業要素の作業に設備が必要であればその設備識別名とが対応付けられた生産品データと、生産するべき複数の品番の生産品に対する品番毎に、当該品番の生産品の生産数が対応付けられ、それら複数の品番の生産品に対する操業時間と、ワークステーション数とを含む入力データとを記憶する記憶手段を備えるとともに、前記記憶手段の入力データから得られる複数の品番を基に、前記記憶手段の生産品データから該当する各品番の複数の作業要素を読み出すとともに、前記記憶手段の設備データおよび作業者データからそれぞれ複数の設備識別名および複数の作業者識別名を読み出し、これら複数の作業要素、複数の設備識別名および複数の作業者識別名の個々に対して、前記ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当て、これらランダム値で関連付けられる作業要素、設備識別名および作業者識別名により複数のワークステーションの編成内容を求めて暫定解の初期解とする初期解生成手段と、創発計算によって暫定解から新たな暫定解を生成しながら最適解を求める創発計算手段と、暫定解の複数のワークステーションに対して、前記記憶手段の生産品データ内の作業時間を参照して工程編成効率を求め、前記記憶手段の少なくとも生産品データ内の作業要素に対する設備識別名を参照して資源不足数を求め、前記記憶手段の生産品データ内の先行要素からワークステーション間移動回数を求め、前記記憶手段の入力データ内の操業時間に対する各ワークステーションの超過時間を求め、これらを重み付けした演算により評価値を算出するものであって、前記創発計算手段の創発計算に利用される評価値算出手段とを備えるので、作業要素、設備識別名および作業者識別名に対して、複数のワークステーションのいずれかに関連付けるランダム値を割り当てて、創発計算で最適化問題を解くことにより、複数の品番の生産品を生産するための複数のワークステーションの編成内容を（準）最適解として求めることができる。
【０１３６】
請求項２記載の発明は、請求項１記載の工程編成装置において、前記記憶手段に記憶されている入力データは、焼鈍し法温度設定値と、この低温化用の焼鈍し法温度更新係数設定値とをさらに含み、前記創発計算手段は、暫定解の複数のワークステーションに対して、作業要素、設備識別名および作業者識別名を関連付けている各ランダム値の一部に代えて、前記ワークステーション数で決まる複数のワークステーションのいずれかに関連付ける新たなランダム値を割り当てることにより、新たな暫定解の候補として複数のワークステーションの編成内容を求める処理と、前記評価値算出手段を利用して、前記候補としての暫定解の評価値を算出し、この評価値が前回求めた現在の暫定解の評価値よりも評価の良い値であれば、前記候補を新たな暫定解とする一方、そうでなければ、前記両評価値の各々を、少ないほど評価が良いエネルギー量に見立て、これら両エネルギー量の差分を焼鈍し法温度設定値で除して除算値を求め、この除算値のマイナス値を変数として指数関数で演算することにより確率を求め、この確率に従って、前記候補を新たな暫定解とするかまたは現在の暫定解のままとする処理とを、焼鈍し処理として所定の探索条件を満足するまで繰り返し、前記探索条件を満足すれば、前記焼鈍し法温度設定値を、これに前記焼鈍し法温度更新係数設定値を乗ずることにより更新しながら、所定の停止条件を満足するまで前記焼鈍し処理を繰り返すので、複数の品番の生産品を生産するための複数のワークステーションの編成内容を（準）最適解として求めることができる。
【０１３７】
請求項３記載の発明は、請求項１または２記載の工程編成装置において、前記評価値算出手段は、前記工程編成効率を求める場合、前記記憶手段の生産品データ内の作業時間を参照して、前記生産するべき複数の品番の生産品に対する品番毎に工程編成効率を求めるので、品番別に工程編成効率が求められ、各ワークステーションの負荷の平準化が行われるから、単一品番の生産品のみを生産する場合でも、編成効率の良い効率的な工程編成を行うことができる。
【０１３８】
請求項４記載の発明は、請求項２記載の工程編成装置において、前記初期解生成手段および前記創発計算手段は、前記ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当てる場合、前記作業要素に対して類似する作業要素に同一のランダム値を割り当てるので、作業要素が類似する品番の作業要素をグループ化することにより、創発計算時の組合せ数が減り、高速に工程編成を行うことができる。また、品番が異なっていても同一のワークステーションで行いたい作業要素がある場合に意図した工程編成を行うことができる。
【０１３９】
請求項５記載の発明は、請求項２記載の工程編成装置において、前記創発計算手段は、前記新たな暫定解の候補として複数のワークステーションの編成内容を求める場合、先ず、設備識別名および作業者識別名に対して新たなランダム値を割り当て、続いてランダム値が割り当てられた設備識別名および作業者識別名と整合するように作業要素に対して新たなランダム値を割り当てるので、設備および作業者を必要とする作業要素が、設備および作業者が割り付いたワークステーションに割り付くため、評価値算出手段によって求められる資源不足数を零にすることができ、実行可能解を確実に得ることができる。
【０１４０】
請求項６記載の発明は、請求項２記載の工程編成装置において、出力手段をさらに備え、前記創発計算手段によって複数の最適解を求め、前記出力手段から出力するので、複数案を提示することにより、実行可能な案を得られやすくすることができる。
【０１４１】
請求項７記載の発明は、請求項２記載の工程編成装置において、前記初期解生成手段および前記創発計算手段は、複数のワークステーションの編成内容を求める場合、前処理として、前記生産するべき複数の品番の生産品に対する品番別に、複数のワークステーションの編成内容を求めて、各ワークステーションに配分された作業要素のグループを前記創発計算で使用する作業要素とするので、高速に工程編成を行うことができ、また品番別に編成効率の良い工程編成が得られる。また、ワークステーション間移動回数を少なくすることができる。
【０１４２】
請求項８記載の発明は、請求項２記載の工程編成装置において、前記記憶手段に記憶されている入力データ内のワークステーション数で決まる複数のワークステーションおよび前記記憶手段に記憶されている生産品データ内の複数の作業要素は、複数の工程のいずれかに対応付けられ、前記初期解生成手段および前記創発計算手段は、作業要素に対して、前記ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当てる場合、その作業要素に対応付けられた工程と同じ工程が対応付けられたワークステーションに関連付けるランダム値を割り当てるので、各作業要素に割り当て可能なワークステーションの候補数が少なくなり、工程編成を高速に算出することができる。また、全品番の作業要素のワークステーション間の移動の仕方が似通ったものとなり、ワークの流れ方が、品番によって逆行することがなくなる。
【０１４３】
請求項９記載の発明は、請求項２記載の工程編成装置において、前記創発計算手段は、前記ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当てる場合、前記評価値算出手段を利用して現在の暫定解の評価値を算出する際に資源不足数が零であったとき、現在の暫定解において合計作業時間がより長いワークステーションに関連付けられた作業要素のうち、設備が必要でない作業要素があればこれを別のワークステーションに関連付けるようにランダム値を割り当てるので、編成効率の高い実行可能解を得ることができるようになる。
【０１４４】
請求項１０記載の発明は、請求項２記載の工程編成装置において、前記創発計算手段は、前記ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当てる場合、前記評価値算出手段を利用して現在の暫定解の評価値を算出する際に資源不足数が零であったとき、現在の暫定解において合計作業時間がより長いワークステーションに関連付けられた作業要素のうち、設備が必要である作業要素があればこれら設備および作業要素を別のワークステーションに関連付けるようにランダム値を割り当てるので、負荷バランスのとれた暫定解が得られ、編成効率の高い実行可能解を得ることができるようになる。
【０１４５】
請求項１１記載の発明は、請求項２記載の工程編成装置において、前記創発計算手段は、前記ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当てる場合、前記評価値算出手段を利用して現在の暫定解の評価値を算出する際に資源不足数が零であったとき、現在の暫定解において合計作業時間が最短のワークステーションに関連付けられた作業要素のうち、設備が必要でない作業要素があればそれ単独で、設備が必要である作業要素があれば設備とともに別のワークステーションに関連付けるようにランダム値を割り当てるので、ワークステーション数を減らすことができ、稼動率の高い実行可能解を得ることができるようになる。また、負荷バランスのとれた解となり、より早い計算時間で解を得ることができる。
【０１４６】
請求項１２記載の発明は、請求項２記載の工程編成装置において、前記初期解生成手段および前記創発計算手段は、作業要素に対して、前記ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当てる場合、その作業要素に必要な設備があればこれら作業要素および設備に同一のランダム値を割り当てるので、高速に工程編成を行うことができ、またワークステーション割当領域において作業要素が必要とする設備の不足がない解を生成できる。
【０１４７】
請求項１３記載の発明は、複数種類の設備をそれぞれ示す複数の設備識別名毎に、当該設備識別名の設備の保有台数が対応付けられるとともに、当該設備識別名の設備により実行可能な作業を示す作業要素が少なくとも１つ対応付けられた設備データと、複数人の作業者をそれぞれ示す複数の作業者識別名を含む作業者データと、複数の品番別に、当該品番の生産品を１台生産するために必要な複数の作業をそれぞれ示す複数の作業要素毎に、当該作業要素に対して少なくとも１つの他の作業要素が先行関係にあればそれを示す先行要素と、当該作業要素の作業に必要な作業時間と、当該作業要素の作業に作業者が必要であればその作業者数と、当該作業要素の作業に設備が必要であればその設備識別名とが対応付けられた生産品データと、生産するべき複数の品番の生産品に対する品番毎に、当該品番の生産品の生産数が対応付けられ、それら複数の品番の生産品に対する操業時間と、ワークステーション数とを含む入力データとを記憶する記憶手段にアクセス可能なコンピュータによって実行される工程編成方法であって、前記記憶手段の入力データから得られる複数の品番を基に、前記記憶手段の生産品データから該当する各品番の複数の作業要素を読み出すとともに、前記記憶手段の設備データおよび作業者データからそれぞれ複数の設備識別名および複数の作業者識別名を読み出し、これら複数の作業要素、複数の設備識別名および複数の作業者識別名の個々に対して、前記ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当て、これらランダム値で関連付けられる作業要素、設備識別名および作業者識別名により複数のワークステーションの編成内容を求めて暫定解の初期解とする初期解生成ステップと、創発計算によって暫定解から新たな暫定解を生成しながら最適解を求める創発計算ステップとを有し、前記創発計算ステップにおいて、暫定解の複数のワークステーションに対して、前記記憶手段の生産品データ内の作業時間を参照して工程編成効率を求め、前記記憶手段の少なくとも生産品データ内の作業要素に対する設備識別名を参照して資源不足数を求め、前記記憶手段の生産品データ内の先行要素からワークステーション間移動回数を求め、前記記憶手段の入力データ内の操業時間に対する各ワークステーションの超過時間を求め、これらを重み付けした演算により評価値を算出するステップが実行されるので、複数の品番の生産品を生産するための複数のワークステーションの編成内容を（準）最適解として求めることができる。
【０１４８】
請求項１４記載の発明の工程編成プログラムは、複数種類の設備をそれぞれ示す複数の設備識別名毎に、当該設備識別名の設備の保有台数が対応付けられるとともに、当該設備識別名の設備により実行可能な作業を示す作業要素が少なくとも１つ対応付けられた設備データと、複数人の作業者をそれぞれ示す複数の作業者識別名を含む作業者データと、複数の品番別に、当該品番の生産品を１台生産するために必要な複数の作業をそれぞれ示す複数の作業要素毎に、当該作業要素に対して少なくとも１つの他の作業要素が先行関係にあればそれを示す先行要素と、当該作業要素の作業に必要な作業時間と、当該作業要素の作業に作業者が必要であればその作業者数と、当該作業要素の作業に設備が必要であればその設備識別名とが対応付けられた生産品データと、生産するべき複数の品番の生産品に対する品番毎に、当該品番の生産品の生産数が対応付けられ、それら複数の品番の生産品に対する操業時間と、ワークステーション数とを含む入力データとを記憶する記憶手段にアクセス可能なコンピュータに対して、前記記憶手段の入力データから得られる複数の品番を基に、前記記憶手段の生産品データから該当する各品番の複数の作業要素を読み出すとともに、前記記憶手段の設備データおよび作業者データからそれぞれ複数の設備識別名および複数の作業者識別名を読み出し、これら複数の作業要素、複数の設備識別名および複数の作業者識別名の個々に対して、前記ワークステーション数で決まる複数のワークステーションのいずれかに関連付けるランダム値を割り当て、これらランダム値で関連付けられる作業要素、設備識別名および作業者識別名により複数のワークステーションの編成内容を求めて暫定解の初期解とする初期解生成機能と、創発計算によって暫定解から新たな暫定解を生成しながら最適解を求める創発計算機能と、暫定解の複数のワークステーションに対して、前記記憶手段の生産品データ内の作業時間を参照して工程編成効率を求め、前記記憶手段の少なくとも生産品データ内の作業要素に対する設備識別名を参照して資源不足数を求め、前記記憶手段の生産品データ内の先行要素からワークステーション間移動回数を求め、前記記憶手段の入力データ内の操業時間に対する各ワークステーションの超過時間を求め、これらを重み付けした演算により評価値を算出するものであって、前記創発計算機能の創発計算に利用される評価値算出機能とを付加するので、複数の品番の生産品を生産するための複数のワークステーションの編成内容を（準）最適解として求めることができる。
【図面の簡単な説明】
【図１】本発明による第１実施形態の工程編成装置の構成図である。
【図２】各品番の生産品についてのプロジェクトのネットワークを示す図である。
【図３】各品番の生産品についてのプロジェクトのネットワークを示す図である。
【図４】同工程編成装置の動作フローチャートである。
【図５】暫定解の例を示す図である。
【図６】暫定解の例を示す図である。
【図７】暫定解の例を示す図である。
【図８】暫定解の例を示す図である。
【図９】暫定解の例を示す図である。
【図１０】本発明による第２実施形態の工程編成装置による工程編成効率の算出の説明図である。
【図１１】同工程編成効率の算出の説明図である。
【図１２】本発明による第９実施形態の工程編成装置にて生成される暫定解を求める際の前処理の例を示す図である。
【図１３】同前処理の例を示す図である。
【図１４】本発明による第１２実施形態の工程編成装置にて生成される暫定解の例を示す図である。
【図１５】同工程編成装置にて生成される暫定解の例を示す図である。
【図１６】本発明による第１５実施形態の工程編成装置にて生成される暫定解の例を示す図である。
【図１７】本発明による第１８実施形態の工程編成装置にて生成される暫定解の例を示す図である。
【図１８】同工程編成装置にて生成される暫定解の例を示す図である。
【符号の説明】
１ 工程編成データ記憶部
２ 初期工程編成データ生成部
３ 焼鈍し法実行部
４ 工程編成データ評価部
５ 工程編成データ出力部[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a process knitting apparatus, a process knitting method, and a process knitting program for obtaining a knitting content of a plurality of workstations for producing a plurality of products of a product number as a solution.
[0002]
[Prior art]
Conventionally, the process organization has been determined by allocating each work element to workstations to which processes, equipment, and workers have been allocated in advance so as to equalize the load on the workstations.
[0003]
For example, in the paper "Development of Line Balancing General Concept and Model" (Measurement and Control, Vol 133, No. 7, pp. 541-546, 1994), when allocating work elements to workstations as balancing of mixed product assembly lines, With the constraint that the working time does not exceed the operating time, a formulation is performed in which the work elements are allocated to the workstations such that the total number of workstations is minimized.
[0004]
Japanese Patent Laid-Open Publication No. Hei 4-21003 (Patent Document 1) discloses a process organization in which the work sequence and the number of processes of each work element are determined and allocated to each process so as not to exceed a preset work time of each process. An apparatus is described. Further, knitting means is provided for allocating so as to minimize the difference of the total time obtained by accumulating each element work time with respect to the work time of each process. According to this process composition device, the work elements are distributed from the first step in the order of the work elements with the earlier work order to the work elements with the later work order, and exceed the predetermined work time of the first step. In this case, by allocating to the next process, an optimal process organization can be obtained.
[0005]
Japanese Patent Application Laid-Open No. 2001-222231 (Patent Document 2) discloses that a work facility plan is created from a plurality of work facilities, a logistics supplementary facility plan is created from a logistics supplementary facility, and the cheapest combination of these combinations is proposed. A production system design apparatus that outputs a production system as a production system is described. In the production system design apparatus, when creating a work equipment plan, a work equipment plan with high knitting efficiency is generated.
[0006]
[Patent Document 1]
JP-A-4-21003
[Patent Document 2]
JP 2001-22211 A
[0007]
[Problems to be solved by the invention]
In the conventional method of allocating each work element to a workstation to which processes, equipment, and workers are pre-allocated, whether the process, equipment, and workers pre-allocated to the workstation are optimal for the workstation. Since it is not taken into account, it is difficult to obtain the optimal knitting content as a solution as a whole. Further, it is assumed that the production line is of a flow shop type, and the work elements are distributed in a state where the respective processes are laid out in series, so that better knitting contents can be obtained by more flexibly allocating the work elements. Even in such a case, such knitting contents cannot be obtained.
[0008]
In the above paper "Development of line balancing general concepts and models", work was carried out without fixing the number of workstations so that the operation rate of each workstation did not exceed 100% and the number of workstations was minimized. The elements are distributed. However, "when a part of the assembly work is performed by a robot in a line where a plurality of types are batch-produced, a problem arises in which work station the robot should be arranged. (Omitted) Since it is difficult to find a possible solution under given conditions, the problem is (omitted).
[0009]
In the process organization device of Patent Document 1, only the distribution of work elements to processes is handled, and the handling of workers and equipment as production resources is not specified.
[0010]
The production system design apparatus of Patent Document 2 can calculate the required number of each facility from a work facility plan with high knitting efficiency, but does not apply to the case of producing a plurality of product numbers. In other words, in a production process for producing a plurality of product numbers, the operation time of each facility changes depending on the number of productions (amount). Therefore, when generating a work equipment plan with high knitting efficiency, it is necessary to create a work equipment plan including the number of equipment by using the number of productions and the work time (tact time) for each work element for each product number. Because there is. Further, many production processes require workers as production resources, and this point cannot be applied to the case of producing a plurality of product numbers.
[0011]
SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and has a process organization apparatus and a process organization capable of finding as a (quasi) optimal solution the contents of a plurality of workstations for producing a plurality of product numbers. It is an object to provide a method and a process organization program.
[0012]
[Means for Solving the Problems]
According to a first aspect of the present invention, there is provided a process organizing apparatus, wherein a plurality of equipment identifiers indicating a plurality of types of equipment are associated with the number of owned equipment having the equipment identifier. Equipment data in which at least one work element indicating a work executable by the equipment having the equipment identifier is associated with, worker data including a plurality of worker identifiers respectively indicating a plurality of workers, and a plurality of product numbers Separately, for each of a plurality of work elements indicating a plurality of works required to produce one product of the product number, if at least one other work element is in a preceding relationship with the work element, the work element is described. The preceding element indicated, the work time required for the work of the work element, the number of workers if work is required for the work element, and the equipment if the work is required for the work element. identification Is associated with the product data, and for each product number of a plurality of product numbers to be produced, the production number of the product of the product number is associated, and the operation time for the product of the plurality of product numbers, And storage means for storing input data including the number of workstations, and based on a plurality of product numbers obtained from the input data of the storage means, a plurality of operations of respective product numbers corresponding to the product data of the storage means. Read out a plurality of facility identifiers and a plurality of worker identifiers from the facility data and the worker data of the storage means, and read the plurality of work elements, the plurality of facility identifiers and the plurality of worker identifiers. Assigns a random value to be associated with any one of a plurality of workstations determined by the number of workstations. Initial solution generation means for finding the organization contents of multiple workstations based on work elements, equipment identifiers, and worker identifiers associated with random values and using them as initial solutions for provisional solutions, and new provisional solutions from provisional solutions by emergent calculations Emergence calculation means for obtaining the optimal solution while generating the, and for a plurality of workstations of the provisional solution, with reference to the work time in the product data of the storage means to determine the process organization efficiency, at least of the storage means The number of resources shortage is obtained by referring to the equipment identifier for the work element in the product data, the number of times of movement between workstations is obtained from the preceding element in the product data in the storage means, and the operation in the input data of the storage means is performed. Calculating an excess time of each workstation with respect to time, and calculating an evaluation value by weighting the excess time; Evaluation value calculation means used for the emergence calculation of the calculation means.
[0013]
According to a second aspect of the present invention, in the process knitting apparatus according to the first aspect, the input data stored in the storage means includes an annealing method temperature set value and an annealing method temperature update coefficient setting for lowering the temperature. Value, and the emergence calculation means replaces a part of each of the random values associating a work element, a facility identifier, and a worker identifier with a plurality of workstations of the provisional solution. By assigning a new random value to be associated with any of the plurality of workstations determined by the number of stations, a process of obtaining the organization contents of the plurality of workstations as candidates for a new provisional solution, and using the evaluation value calculation means Calculating an evaluation value of the provisional solution as the candidate, and if the evaluation value is a value that is better than the evaluation value of the current provisional solution obtained last time, the candidate is On the other hand, if it is not a provisional solution, otherwise, each of the two evaluation values is regarded as an energy amount with a smaller evaluation being better, and the difference between these two energy amounts is divided by the annealing method temperature set value to obtain a division value. Calculating a probability by calculating an exponential function using the minus value of the divided value as a variable, and processing the candidate as a new provisional solution or keeping the current provisional solution in accordance with the probability. The process is repeated until a predetermined search condition is satisfied. If the search condition is satisfied, the annealing method temperature set value is updated by multiplying the annealing method temperature set value by the annealing method temperature update coefficient set value, The annealing process is repeated until the stop condition is satisfied.
[0014]
According to a third aspect of the present invention, in the process knitting apparatus according to the first or second aspect, when the evaluation value calculating means obtains the process knitting efficiency, the evaluation value calculating means refers to a work time in the product data of the storage means. The process organization efficiency is obtained for each product number of a plurality of product numbers to be produced.
[0015]
According to a fourth aspect of the present invention, in the process organizing apparatus according to the second aspect, the initial solution generation means and the emergence calculation means assign a random value to be associated with one of a plurality of workstations determined by the number of workstations. The same random value is assigned to a work element similar to the work element.
[0016]
According to a fifth aspect of the present invention, in the process knitting apparatus according to the second aspect, when the emergence calculation means obtains the knitting contents of a plurality of workstations as candidates for the new provisional solution, first, the equipment identification name and the work Assigning a new random value to the worker identifier, and subsequently assigning a new random value to the work element so as to match the equipment identifier and the worker identifier to which the random value has been assigned. .
[0017]
According to a sixth aspect of the present invention, in the process organization apparatus according to the second aspect, an output unit is further provided, and a plurality of optimal solutions are obtained by the emergence calculation unit and output from the output unit.
[0018]
According to a seventh aspect of the present invention, in the process knitting apparatus according to the second aspect, when the initial solution generating means and the emergence calculating means determine the knitting contents of a plurality of workstations, the initial solution generating means and the emergence calculating means perform pre-processing of the plurality of workstations. The content of the organization of a plurality of workstations is obtained for each product number corresponding to the product number, and a group of work elements allocated to each workstation is used as the work element used in the emergence calculation.
[0019]
The invention according to claim 8 is the process organization apparatus according to claim 2, wherein a plurality of workstations determined by the number of workstations in the input data stored in the storage means and a product stored in the storage means. The plurality of work elements in the data are associated with any of a plurality of steps, and the initial solution generation means and the emergence calculation means determine which of the plurality of workstations is determined by the number of work stations for the work element. When assigning a random value associated with the work element, a random value associated with a workstation associated with the same step as the step associated with the work element is assigned.
[0020]
According to a ninth aspect of the present invention, in the process organizing apparatus according to the second aspect, when the emergence calculation means assigns a random value to be associated with one of a plurality of workstations determined by the number of the workstations, the evaluation value calculation means When the resource shortage number is zero when calculating the evaluation value of the current provisional solution by using, the equipment among the work elements associated with the workstation whose total work time is longer in the current provisional solution is It is characterized by assigning a random value so that any unnecessary work element is associated with another workstation.
[0021]
According to a tenth aspect of the present invention, in the process organizing apparatus according to the second aspect, when the emergence calculation means assigns a random value to be associated with any of a plurality of workstations determined by the number of the workstations, the evaluation value calculation means When the resource shortage number is zero when calculating the evaluation value of the current provisional solution by using, the equipment among the work elements associated with the workstation whose total work time is longer in the current provisional solution is If there is a required work element, a random value is assigned so as to associate the equipment and the work element with another workstation.
[0022]
According to an eleventh aspect of the present invention, in the process knitting apparatus according to the second aspect, when the emergence calculation means assigns a random value to be associated with one of a plurality of workstations determined by the number of the workstations, the evaluation value calculation means When the resource shortage number is zero when calculating the evaluation value of the current provisional solution by using, the equipment among the work elements associated with the workstation with the shortest total work time in the current provisional solution is It is characterized in that a random value is assigned so that there is a work element that is not required alone and a work element that requires equipment is associated with another work station together with the work element.
[0023]
According to a twelfth aspect of the present invention, in the process organizing apparatus according to the second aspect, the initial solution generation means and the emergence calculation means may be configured such that, for a work element, one of a plurality of workstations determined by the number of work stations. When assigning a random value to be associated, if there is equipment required for the work element, the same random value is assigned to the work element and the equipment.
[0024]
In the invention according to claim 13, the number of owned equipment having the equipment identifier is associated with each of a plurality of equipment identifiers respectively indicating a plurality of types of equipment, and the work executable by the equipment having the equipment identifier is performed. Equipment data associated with at least one of the indicated work elements, worker data including a plurality of worker identifiers respectively indicating a plurality of workers, and one product of the product number for each of a plurality of product numbers. For each of a plurality of work elements indicating a plurality of works necessary to perform the work, a preceding element indicating that at least one other work element is in a preceding relationship with the work element, and a work for the work of the work element. Product data in which the required work time, the number of workers if work is needed for the work element, and the equipment identification name if equipment is needed for work on the work element When, For each product number of a plurality of product numbers to be produced, the production number of the product of the product number is associated, and input data including the operation time and the number of workstations for the product products of the plurality of product numbers is stored. A method for organizing processes, which is executed by a computer accessible to a storage unit to perform, based on a plurality of product numbers obtained from input data of the storage unit, a plurality of product numbers of each corresponding product number from the product data of the storage unit. A work element is read, and a plurality of equipment identification names and a plurality of worker identification names are respectively read from the equipment data and the worker data in the storage means, and the plurality of work elements, the plurality of equipment identification names, and the plurality of worker identifications are read. For each name, a random value associated with one of the workstations determined by the number of workstations An initial solution generation step of obtaining the organization contents of a plurality of workstations based on the work elements, equipment identifiers, and operator identifiers associated with these random values and setting the initial content of the provisional solution; and An emergence calculation step of finding an optimal solution while generating a new provisional solution, wherein in the emergence calculation step, a work time in the product data of the storage means is referred to for a plurality of workstations of the provisional solution. To obtain the number of resources shortage by referring to the equipment identifier for at least the work element in the product data in the storage means, and to move from workstation to workstation from the preceding element in the product data in the storage means. The number of times, the excess time of each workstation with respect to the operation time in the input data of the storage means is determined, and these are weighted. The step of calculating an evaluation value is performed by the attached calculation.
[0025]
According to a fourteenth aspect of the present invention, the process scheduling program is executed by the equipment having the equipment identification name while the number of the equipments having the equipment identification name is associated with each of the plurality of equipment identification names indicating the plurality of types of equipment. Equipment data in which at least one work element indicating a possible work is associated, worker data including a plurality of worker identifiers respectively indicating a plurality of workers, and a plurality of product numbers for each product number For each of a plurality of work elements respectively indicating a plurality of works required to produce one of the work elements, a precedent element indicating at least one other work element in a precedent relation to the work element, The work time required for the work of the element, the number of workers if work is needed for the work element, and the equipment identification name if equipment is needed for work of the work element are associated with each other. For each product number corresponding to a plurality of product numbers to be produced, the production number of the product having the product number is associated, and the operation time and the number of workstations for the product products having the plurality of product numbers are associated with each other. For a computer that can access a storage unit that stores input data including the input data, based on a plurality of product numbers obtained from the input data of the storage unit, a plurality of operations of the corresponding product numbers from the product data of the storage unit are performed. Read out a plurality of facility identifiers and a plurality of worker identifiers from the facility data and the worker data of the storage means, and read the plurality of work elements, the plurality of facility identifiers and the plurality of worker identifiers. Allocate a random value to be associated with any of a plurality of workstations determined by the number of workstations, An initial solution generation function that obtains the organization contents of a plurality of workstations based on the work elements, equipment identifiers, and operator identifiers associated with these random values and sets the initial solution of the provisional solution, and a new solution from the provisional solution by emergence calculation An emergence calculation function for finding an optimum solution while generating a provisional solution; and for a plurality of workstations of the provisional solution, a process organization efficiency is determined by referring to a work time in the product data of the storage means. The number of resources shortage is obtained by referring to the equipment identifier for at least the work element in the product data, and the number of times of movement between workstations is obtained from the preceding element in the product data of the storage means. Calculating the excess time of each workstation with respect to the operation time of the work station, and calculating an evaluation value by weighting these work times. It is characterized by adding an evaluation value calculation function used for the emergence calculation of the generation calculation function.
[0026]
BEST MODE FOR CARRYING OUT THE INVENTION
(1st Embodiment)
FIG. 1 is a configuration diagram of a process organizing apparatus according to a first embodiment of the present invention, FIGS. 2 and 3 are diagrams showing a network of projects for products of respective product numbers, and FIG. 4 is an operation flowchart of the process organizing device. 5 to 9 are diagrams showing examples of provisional solutions.
[0027]
The process organizing apparatus according to the first embodiment includes an input device such as a keyboard and a mouse, a display device such as an LCD or a CRT, a storage device such as a hard disk or a data server, and a computer connected to each of these devices. In addition, the knitting contents of a plurality of workstations for producing a plurality of product numbers are obtained as (quasi) optimal solutions.
[0028]
More specifically, as shown in FIG. 1, the above-mentioned storage device and the storage device in the computer constitute a process data storage unit 1 (storage means). As processing function units, an initial process organization data generation unit 2 (initial solution generation unit / function), an annealing method execution unit 3 (emergency calculation unit / function), and a process organization data evaluation unit 4 (evaluation value calculation unit / function) ) And a process data output unit 5 (output means).
[0029]
The process organization data storage unit 1 stores equipment data, worker data, product data, input data, and the like.
[0030]
The equipment data is composed of a data table, and as shown in the following (Table 1), for each of a plurality of equipment identification names (data “S1”, “S2”,... Is associated with the number of owned equipment (“the number”) of the equipment having the equipment identification name, and a work element indicating the work executable by the equipment having the equipment identification name (data “A” in the “possible work” column). , “C”,...) Are associated with at least one.
[0031]
[Table 1]

[0032]
The worker data is composed of a data table, and includes a plurality of worker identifiers (data in the “worker” column) respectively indicating a plurality of workers as shown in (Table 2) below. For each worker identification name, at least one work element (data in the “possible work” column) indicating a work executable by the worker of the worker identification is associated. In addition, in the table, the worker can work on a plurality of work elements.
[0033]
[Table 2]

[0034]
The product data is composed of a plurality of (two in this case) data tables, and as shown in the following (Table 3) and (Table 4), "X", (Table 4) ), At least one other work element precedes the work element for each of a plurality of work elements, each indicating a plurality of works required to produce one product of the product number, in “Y”｝. If there is a relationship, the preceding element indicating the relationship, the work time required for the work element, the number of workers if work is required for the work element, and equipment for the work of the work element If necessary, the equipment identification name (data in the “necessary equipment” column) is associated with the equipment identification name.
[0035]
[Table 3]

[0036]
[Table 4]

[0037]
Here, the work element means a processing content of a product such as “screw tightening” and “quality check”. The work element “B” in (Table 3) means that the operation is performed after the preceding element “A” is completed. Further, a so-called project network as shown in FIG. 2 can be obtained from (Table 3), and a project network as shown in FIG. 3 can be obtained from (Table 4).
[0038]
The input data is composed of a data table, and as shown in the following (Table 5), for each product number corresponding to a plurality of product numbers to be produced, the production number of the product of the product number is associated with the product number. Including the operation time, the number of workstations, the annealing method temperature set value, the annealing method temperature update coefficient setting value for lowering the temperature, and the annealing method calculation time setting value for a plurality of product numbers. I have.
[0039]
[Table 5]

[0040]
Of the various data, for example, input data is input via an input device (see S10 in FIG. 4) and stored in the process data storage unit 1, while other data is created in advance and the process It is stored in the composition data storage unit 1.
[0041]
Returning to FIG. 1, based on a plurality of product numbers obtained from the input data of the process composition data storage unit 1, the initial process composition data generation unit 2 uses the product data of the process composition data storage unit 1 In addition to reading a plurality of work elements, a plurality of equipment identification names and a plurality of worker identification names are read from the equipment data and the worker data in the process scheduling data storage unit 1, respectively. Assign a random value (here, a random value within the number of workstations) associated with one of the plurality of workstations determined by the number of workstations to each of the plurality of worker identifiers, and perform an operation associated with the random value The first provisional solution was determined by determining the organization of multiple workstations using the elements, equipment identifiers, and operator identifiers. It is to a solution (see S11, S12 of FIG. 4).
[0042]
More specifically, for the product with the product number “X”, the work element and the preceding element are read from the data table of (Table 3) as shown in (Table 6) below, and the A value ("workstation") is assigned.
[0043]
[Table 6]

[0044]
Similarly, for the product of the product number “Y”, the work element and the preceding element are read from the data table of (Table 4) as shown in (Table 7) below, and a random value is assigned to each of them. Can be Note that the data in the “presence / absence of movement” column in (Table 6) and (Table 7) are obtained by the process organization data evaluation unit 4.
[0045]
[Table 7]

[0046]
Then, as shown in the following (Table 8), the equipment identification name (“equipment”) and the number of owned vehicles (“the number of vehicles”) are read out from the data table of (Table 1), and the number of each owned vehicle is one. A random value is assigned to the equipment identification name per hit, and from the data table of (Table 2), as shown in the following (Table 9), the worker identification name (“worker”) and the work element (“possible work”) ) Are read and random values are assigned to the respective data, thereby generating an initial solution consisting of the data tables of (Table 6) to (Table 9) and indicating the knitting contents of FIG. However, the work elements of the part number “X” and the part number “Y” are grouped, and a random value is assigned to the same work element.
[0047]
[Table 8]

[0048]
[Table 9]

[0049]
In FIG. 5, for example, the work element “E” required for the product with the product number “X” has a work time of 9 minutes and the number of products produced is 60, so the total work time of the work element “E” is Is 9 hours (= 9 [minutes / unit] × 60 [units]). The work element “A” required for the products of the product numbers “X” and “Y” has a work time of 5 minutes and a production number of 120 units. Therefore, the total work time of the work element “A” is This is 10 hours (= 5 [minutes / unit] × 120 [units]).
[0050]
When reading out a plurality of equipment identification names and a plurality of worker identification names, the initial process scheduling data generation unit 2 outputs a plurality of equipment identification numbers corresponding to a plurality of work elements required for a product having a plurality of product numbers to be produced. The configuration may be such that a name and a plurality of worker identification names are read.
[0051]
The annealing method execution unit 3 shown in FIG. 1 finds an optimal solution while generating a new provisional solution from the provisional solution, and performs emergent calculations such as Evolutionary Computation and annealing method (Simulated Annealing). (Emaergent Computation) Find the optimal solution by the annealing method. The details of the annealing method execution unit 3 will be described after the next process organization data evaluation unit 4.
[0052]
The process organization data evaluation unit 4 obtains the process organization efficiency for a plurality of provisional solution workstations by referring to the work time in the product data of the process organization data storage unit 1, At least the number of resources shortage is obtained by referring to the equipment identifier for the work element in the product data, and the number of times of movement between workstations is obtained from the preceding element in the product data in the process data storage 1. The excess time of each workstation with respect to the operation time in the input data of No. 1 is calculated, and an evaluation value is calculated by a calculation in which the excess time is weighted. The evaluation value is used for the emergence calculation of the annealing method execution unit 3.
[0053]
Here, the process scheduling efficiency is calculated by an arithmetic expression of {(total work time of each workstation) / {(maximum total work time) × (number of workstations whose total work time is not zero)}. For example, the process organization efficiency of the initial solution in FIG. 4 is 0.65 {= (15 + 16 + 24 + 10 + 13 + 15) / (24 × 6)}.
[0054]
The resource shortage number indicates the shortage number of facilities and workers, and is “1” if the number is insufficient and “0” otherwise. In the example of FIG. 5, “Workstation 6” is assigned to the worker with the worker identification name “SG7”, but the worker is assigned the work element “M” assigned to “Workstation 6”. Since the operation of "" cannot be executed, the resource shortage number for "workstation 6" is "1".
[0055]
The number of times of movement between workstations is obtained from the number of differences between the work station of the work element and the work station of the work element of the work element of the work element, and is input in the "movement presence / absence" column. For example, in (Table 6), the number of movements between the work elements of the work element “B” is the same as “Workstation 1” of the work element “A” in the preceding element, and thus becomes “0”. The number of times of movement of the work element “D” between workstations is “1” because it is “Workstation 4” different from “Workstation 1” of the work element “B” in the preceding element. The number of times of movement of the work element “M” between the workstations is different from “Workstation 5”, “Workstation 5”, and “Workstation 3” of the work elements “J”, “K”, and “L” in the preceding element. Since it is “Workstation 6”, it becomes “3”.
[0056]
The excess time is calculated for each workstation, and is the excess of the total work time of the workstation with respect to 20 hours of operation time. For example, since the total work time of “Workstation 3” is 24 hours, the excess time is “4”.
[0057]
The evaluation value is calculated as a weighted linear sum of the process organization efficiency, the number of resource shortages, the number of times of movement between workstations, and the excess time. For example, a weighted linear of evaluation value = 100 × (process organization efficiency) + (− 1) × (resource shortage number) + (− 1) × (excess time) + (− 1) × (number of times of movement between workstations) If the sum is used, the evaluation value of the initial solution in FIG. 5 is 36 (= 100 × 0.65-1-4-14).
[0058]
Here, the annealing method execution unit 3 will be described in detail. The annealing method execution unit 3 calculates the evaluation value C of the provisional solution as the initial solution using the process organization data evaluation unit 4 (see FIG. 4). After that, the process proceeds to the annealing method.
[0059]
In other words, for a plurality of workstations of the provisional solution, one of the plurality of workstations determined by the number of workstations, instead of a part of each random value that associates the work element, the equipment identifier, and the worker identifier. By assigning a new random value to be associated with, a process of obtaining the organization contents of a plurality of workstations as candidates for a new provisional solution (S14), and using the process composition data evaluation unit 4 to generate a provisional solution as a candidate The evaluation value C ′ is calculated (S15). If the evaluation value C ′ is a value that is better than the evaluation value C of the current provisional solution obtained last time (S16), the candidate is set as a new provisional solution (S17). ), Otherwise, each of the two evaluation values is regarded as an energy amount with a smaller evaluation value, and the difference ΔC between these two energy amounts is divided by the annealing method temperature set value T. A probability is obtained by calculating an exponential function using a minus value of the divided value as a variable, and according to the probability, a candidate is set as a new provisional solution or remains at the current provisional solution (S18). , S19) are repeatedly executed until a predetermined search condition is satisfied. A part of each of the above random values may be selected using a random number, and the evaluation value for each workstation is randomly selected from each workstation belonging to the inferior range from the worst value to a predetermined value. You may make it select.
[0060]
If the search condition is satisfied, the annealing method temperature set value T is updated by multiplying the annealing method temperature set value T by the annealing method temperature update coefficient set value α (S20), and the annealing is performed until a predetermined stop condition is satisfied. The processing is repeated (S21).
[0061]
The process scheduling data output unit 5 outputs the optimal solution obtained by the annealing method executing unit 3 on a display device, prints it on a printer, or writes and outputs it on a storage device.
[0062]
Next, the operation of the process organizing apparatus having the above configuration will be described. When the designer inputs input data via the input device while looking at the process organizing screen displayed on the display device (S10 in FIG. 4), An initial solution indicating the knitting contents in the example of FIG. 5 is generated by the initial process knitting data generation unit 2 (S11), and the evaluation value C is calculated by the process knitting data evaluation unit 4 (S13).
[0063]
After that, the process shifts to the annealing method processing by the annealing method execution unit 3 and, as shown in each example of FIGS. 6 to 9, the work element, the equipment identification name and the work By assigning a new random value to be associated with one of a plurality of workstations determined by the number of workstations instead of a part of each of the random values associated with the A process for determining the composition of the station is executed (S14).
[0064]
Subsequently, an evaluation value C 'of the provisional solution as a candidate is calculated (S15), and if the evaluation value C' is a value that is better than the evaluation value C of the current provisional solution obtained last time (YES in S16). Are stored as new provisional solutions (S17), and the process proceeds to step S20.
[0065]
On the other hand, if the evaluation value C ′ is not a value that is better than the evaluation value C of the current provisional solution obtained last time (NO in S16), a random number value P is generated within the range of 0 ≦ P ≦ 1 (S18). Each of the two evaluation values is regarded as an energy amount having a smaller evaluation value, and the difference ΔC between the two energy amounts is annealed and divided by a method temperature set value T to obtain a division value, and a minus value (−−) of the division value is obtained. The probability is obtained by calculating the exponential function {exp (−ΔC / T)} using ΔC / T) as a variable, and according to the probability, the candidate is set as a new provisional solution (YES in S19) or the current provisional The process of keeping the solution (NO in S19) is executed. In FIG. 4, if the calculated value of exp (−ΔC / T) is equal to or smaller than the random number value P, the process proceeds to step S17; otherwise, the process proceeds to step S20.
[0066]
The above-mentioned annealing method processing is repeatedly executed until a predetermined search condition, for example, a predetermined time elapses, and when the search condition is satisfied, the annealing method temperature set value T and the annealing method temperature update coefficient set value While updating by multiplying by α (S20), it is repeatedly executed until a predetermined stop condition, for example, the resource shortage number and the excess time become zero and the annealing method temperature set value T exceeds the annealing method calculation time set value. Is done. The search condition is not limited to a predetermined time, but may be a predetermined number of times, a predetermined number of candidates accepted or rejected. Further, the stop condition may be a condition in which annealing is performed a predetermined number of times, the number of times that reception hardly occurs, or a condition in which the same state is generated many times.
[0067]
Subsequently, the work element of the optimal solution, the equipment identification name, the worker identification name, and the like obtained by the annealing method execution unit 3 are output by the process data output unit 5.
[0068]
Here, the candidates in FIG. 6 correspond to the equipments having the work elements “E”, “I”, “K”, “L” and equipment identification name “S2” for the plurality of workstations of the initial solution in FIG. This is generated by assigning a new random value in place of each random value that associates one device, two devices with the equipment identifier “S3”, and the worker identifier “SG3”. The provisional solution as a candidate in FIG. 6 has a process organization efficiency of 0.74 {= 93 / (18 × 7)}, a resource shortage of 1, an excess time of 0, and the number of times of movement between workstations is 16. Therefore, the evaluation value C ′ is 57. Since the evaluation value C ′ is better than the evaluation value C (= 36) of the initial solution that is the current provisional solution, the process proceeds to step S17, and the candidate in FIG. 6 becomes the provisional solution.
[0069]
The candidate in FIG. 7 is generated from the provisional solution in FIG. 6, and the process organization efficiency is 0.86 ｛= 93 / (18 × 6)｝, the resource shortage is 4, the excess time is 0, and the number of times of movement between workstations is 16 Therefore, the evaluation value C ′ of the candidate is 66. Since the evaluation value C ′ is better than the current provisional solution evaluation value C (= 57), the process proceeds to step S17, and the candidate in FIG. 7 becomes the provisional solution.
[0070]
The candidates in FIG. 8 are generated from the provisional solution in FIG. 7, and the process organization efficiency is 0.73 {= 93 / (16 × 8)}, the resource shortage is 3, the excess time is 0, and the number of times of movement between workstations is 19 Therefore, the evaluation value C ′ of the candidate is 51. Since the evaluation value C ′ is not better than the current provisional solution evaluation value C (= 66), the process proceeds to step S18. Then, in step S18, if a random value P of, for example, 0.25 is generated, the calculated value of exp (−ΔC / T) is 0.86 ｛= exp (− | 51−66 | / 100)}. , Is not less than or equal to the random number value P, the candidate in FIG. 8 is rejected, and the process proceeds to step S20.
[0071]
Next, the candidates in FIG. 9 are generated from the provisional solution in FIG. 7, the process organization efficiency is 0.86 ｛= 93 / (18 × 6)｝, the number of resource shortages is 0, the excess time is 0, and the transfer between workstations is performed. Since the number of times is 15, the candidate evaluation value C ′ is 51. Since the evaluation value C ′ is not better than the current provisional solution evaluation value C (= 66), the process proceeds to step S18. Then, in step S18, if a random value P of, for example, 0.90 is generated, the calculated value of exp (−ΔC / T) becomes 0.86, which is equal to or less than the random value P, and the process proceeds to step S17. The candidates in FIG. 9 are provisional solutions and are prevented from falling into local solutions.
[0072]
In this way, the search for the optimum solution is continued, and the final provisional solution is set as a (quasi) optimum solution. In the above process, at least the provisional solution having the highest evaluation value C is stored and retained, and when the final series of optimal solution search processing is completed, the provisional solution evaluation values stored so far are stored. The one with the highest C can also be the (sub) optimal solution.
[0073]
As described above, according to the first embodiment, a random value to be associated with any of a plurality of workstations is assigned to a work element, a facility identifier, and a worker identifier, and an optimization problem is solved by emergent calculation. In addition, the knitting contents of a plurality of workstations for producing a plurality of product numbers can be determined as (quasi) optimal solutions without falling into local solutions.
[0074]
(2nd Embodiment)
In the process organization device according to the second embodiment of the present invention, in the first embodiment, when the process organization data evaluation unit 4 obtains the process organization efficiency, the operation time in the product data of the process organization data storage unit 1 is referred to. Then, the process organization efficiency is obtained for each product number of a plurality of product numbers to be produced.
[0075]
For example, in the case of the initial solution shown in FIG. 5, FIG. 10 shows a portion where only the product of the product number “X” is produced, and FIG. 10 shows a portion where only the product of the product number “Y” is produced. Is shown in FIG. In this case, the process organization efficiency of the product number “X” is obtained from the work time of the workstation for the work elements shown in (Table 6), and 0.42 ｛= (7.5 + 24 + 5 + 6.5 + 7.5) / (24 × 5) It becomes｝.
[0076]
The process organization efficiency of the product number “Y” is also obtained from the work time of the workstation for the work elements shown in (Table 7), and is 0.53 {= 7.5 + 16 + 5 + 6.5 + 7.5) / (16 × 5)}. Become.
[0077]
Then, an evaluation value is calculated as a weighted linear sum of the process organization efficiency, the resource shortage number, the number of times of movement between workstations, and the excess time for each product number.
[0078]
When a single product number is produced together (lot production, etc.) or when the production volume ratio of each product number is not constant and fluctuates, only one product number has a longer flow time in the process. Although it fluctuates, according to the second embodiment, the process organization efficiency is calculated for each product number, and the load on each workstation is leveled. Therefore, even when only a product with a single product number is produced, In addition, efficient process knitting with good knitting efficiency can be performed.
[0079]
(Third embodiment)
In the process organization device according to the third embodiment of the present invention, in the first embodiment, the initial process organization data generation unit 2 and the annealing method execution unit 3 associate a random number with any of a plurality of workstations determined by the number of workstations. When assigning a value, the same random value is assigned to a work element similar to the work element.
[0080]
In the case of the products with the product numbers “X” and “Y” shown in (Table 3) and (Table 4), the work elements required for these are “A”, “B”, “D”, “G”, “K”, “J”, and “M”. Here, in the production line of a multi-product product, it is easier to manage the work of the same work element at the same place. In other words, the work element “A” of the part number “X” and the work element “A” of the part number “Y” can be easily managed in the same workstation, so that the work element “A” of the part number “X” can be easily managed. "And the work element" A "of the product number" Y "are always assigned to the same workstation.
[0081]
According to the third embodiment, by grouping work elements having part numbers having similar work elements, the number of combinations at the time of executing the annealing method is reduced, and the process can be organized at high speed. In addition, even if the part numbers are different, the intended process can be organized when there is a work element desired to be performed on the same workstation.
[0082]
(Fourth embodiment)
According to the process composition device of the fourth embodiment of the present invention, in the third embodiment, the initial process composition data generation unit 2 and the annealing method execution unit 3 assign the same random value to a work element similar to the work element. When the assignment is performed, the same random value is assigned to the similar work element unless the total time of the work time determined by the similar work element exceeds the operation time.
[0083]
In the fourth embodiment, for example, the work element “A” of the product number “X” and the work element “A” of the product number “Y” are always assigned to the same workstation. Is executed when the total work time of the work element does not exceed the operation time of 20 hours. The process of this example is executed because the total work time is 10 hours (= 5 [minutes / unit] × 120 [units]) and does not exceed 20 hours.
[0084]
According to the fourth embodiment, when work elements having similar contents are grouped, the process can be organized so that the total work time does not exceed the operation time.
[0085]
(Fifth embodiment)
In the process knitting apparatus according to the fifth embodiment of the present invention, in the first embodiment, when the annealing method executing unit 3 obtains the knitting contents of a plurality of workstations as candidates for a new provisional solution, first, the equipment identification name is used. And assigning a new random value to the worker identifier, and subsequently assigning a new random value to the work element so as to match the equipment identifier and the worker identifier to which the random value has been assigned. And
[0086]
In assigning a workstation to each of the work elements, each equipment identifier, and each worker identifier shown in Tables 6 to 9, first, a new equipment identifier and a new worker identifier are assigned to the equipment identifiers and the worker identifiers. A random value is assigned, and then a new random value is assigned to a work element, with the random value (workstation number) for the equipment identifier and the worker identifier as a constraint. For example, the work element “A” of the product numbers “X” and “Y” requires the equipment with the equipment identification name “S1”. Only the workstation 1 "can be assigned.
[0087]
According to the fifth embodiment, since the work elements requiring equipment and workers are allocated to the workstations to which the equipment and workers are allocated, the resource shortage number required by the process organization data evaluation unit 4 is reduced to zero. And a feasible solution can be reliably obtained.
[0088]
(Sixth embodiment)
The sixth embodiment according to the present invention is characterized in that, in the first embodiment, a plurality of optimal solutions are obtained by the annealing method executing unit 3 and output from the process knitting data output unit 5.
[0089]
In other words, by proposing a plurality of sub-optimal solutions, flexibility is provided in selecting the final placement plan. Since the final solution obtained by the annealing method execution unit 3 depends on the initial solution and the search, even if the initial solution is the same, the solution varies depending on, for example, the number of searches. Of course, even if the initial solutions are different, the final solution will vary. In the sixth embodiment, the annealing method execution unit 3 is configured to obtain a plurality of final solutions from a plurality of initial solutions and output the final solutions from the process scheduling data output unit 5 by utilizing the property. .
[0090]
In a configuration in which only one optimal solution is output, execution may not be possible due to a layout or the like. However, according to the sixth embodiment, a plurality of alternatives are presented, so that an executable alternative can be easily obtained. be able to.
[0091]
(Seventh embodiment)
According to the process organization device of the seventh embodiment of the present invention, in the first embodiment, the number of workstations of the input data stored in the process organization data storage unit 1 is equal to the total number of work items for a plurality of product numbers to be produced. It is automatically set by multiplying a value obtained by dividing the total working time of the element by the operating time and rounding up by a margin.
[0092]
Based on the work times of the product numbers “X” and “Y” shown in (Table 3) and (Table 4) and the number of productions by product number shown in (Table 5), the total work of the product numbers “X” and “Y” 93 hours are obtained. Then, 93 hours of the total work time is divided by 20 hours of the operation time, and the obtained value 4.6 (= 93 ÷ 20) is rounded up to 5. If this is multiplied by a margin rate of 1.5, for example, , 7.5 (= 5 × 1.5). The value 8 obtained by rounding up this 7.5, for example, is automatically set as the number of workstations.
[0093]
According to the seventh embodiment, an appropriate number of workstations can be automatically set, and optimal process organization can be performed.
[0094]
(Eighth embodiment)
The process organizing apparatus according to the eighth embodiment of the present invention is characterized in that, in the first embodiment, the annealing method execution unit 3 sets the number of workstations to a predetermined value if the excess time does not become zero even if the stop condition is satisfied. It is characterized in that the optimal solution is obtained while generating a new provisional solution from the provisional solution by emergence calculation again.
[0095]
If the set number of workstations (= 8) does not provide an executable process formation result, that is, a process formation result in which the resource shortage and excess time become zero, the number of workstations is incremented by a predetermined value. Then, the optimal solution is calculated while generating a new provisional solution from the provisional solution by emergence calculation again.
[0096]
According to the eighth embodiment, optimal process organization can be performed even when the set number of workstations is not appropriate.
[0097]
(Ninth embodiment)
FIG. 12 and FIG. 13 are diagrams showing examples of pre-processing when obtaining a provisional solution generated by the process composition device of the ninth embodiment according to the present invention.
[0098]
In the process organization device of the ninth embodiment, in the first embodiment, when the initial process organization data generation unit 2 and the annealing method execution unit 3 obtain the organization contents of a plurality of workstations, they should be produced as preprocessing. It is characterized in that the contents of organization of a plurality of workstations are obtained for each product number of a plurality of product numbers, and a group of work elements allocated to each workstation is used as a work element used in emergence calculation.
[0099]
As the pre-processing, for the work element of the product number “X”, as shown in the example of FIG. 12, the work elements (A, B), (E), (H), (D, G, L), (K, J) and (M) are grouped into six, and the work elements with the product number "Y" are, as shown in the example of FIG. 13, work elements (A, B), (C, F), (I) ), (D, G), (K, J), and (M). Then, the process is organized by allocating to the workstation in units grouped by product number.
[0100]
According to the ninth embodiment, the process is organized in advance for each product number, and the work elements are grouped for each product number, so that the process can be organized at a high speed, and the process organization with good knitting efficiency can be obtained for each product number. Can be Further, the number of times of movement between workstations can be reduced.
[0101]
(Tenth embodiment)
The process organization device of the tenth embodiment according to the present invention is determined by the number of workstations in the input data stored in the process organization data storage unit 1 as shown in the following (Table 10) in the first embodiment. A plurality of workstations and a plurality of work elements in the product data stored in the process data storage unit 1 are associated with any of the plurality of processes, and the initial process data generation unit 2 and the execution of the annealing method are performed. When the unit 3 assigns a random value associated with one of a plurality of workstations determined by the number of workstations to the work element, the work unit assigns a random value to the work station associated with the same process as the process associated with the work element. It is characterized in that a random value to be associated is assigned.
[0102]
[Table 10]

[0103]
That is, a work station and a work element are assigned to a process. In this case, the work element and the work station find the same combination, and the assignment is permitted only when the work is the same. For example, since “step 2” is associated with the work element “F”, a random value associated with any of “workstation 4” to “workstation 6” associated with “step 2” is assigned. Will be done.
[0104]
According to the tenth embodiment, since each workstation and each work element are associated with a process, the number of workstation candidates that can be assigned to each work element is reduced, and a process organization can be calculated at high speed. . In addition, the manner in which the work elements of all product numbers are moved between the workstations is similar, and the flow of the work does not reverse according to the product number.
[0105]
(Eleventh embodiment)
According to the eleventh embodiment of the process knitting apparatus of the present invention, in the first embodiment, the initial process knitting data generator 2 and the annealing method executor 3 are configured such that a plurality of workstations determined by the number of workstations for a work element. When assigning a random value to be associated with any of the work elements, if there is another work element indicated by the preceding element associated with the work element, the same as the workstation associated with the random value assigned to this other work element Or assigning a random value to be associated with a subsequent workstation.
[0106]
Here, a random value associated with a workstation number equal to or greater than the workstation number assigned to the work element of the preceding element is assigned. For example, with respect to the work element “B” in the following example of (Table 11), since the preceding element is the work element “A” assigned to “Workstation 1”, the work element after “Workstation 1” A random value is assigned to associate with the workstation.
[0107]
[Table 11]

[0108]
In (Table 11), a random value associated with “Workstation 1” is assigned. Similarly, the work element “G” is assigned a random value to be associated with the work station “work station 3” or later because the preceding element is the work element “D” assigned to “work station 3”. Can be In (Table 11), a random value associated with “workstation 4” is assigned.
[0109]
According to the eleventh embodiment, the flow of the work does not reverse depending on the product number.
[0110]
(Twelfth embodiment)
FIGS. 14 and 15 are diagrams showing examples of provisional solutions generated by the process organizing apparatus according to the twelfth embodiment of the present invention.
[0111]
In the process organization device according to the twelfth embodiment, in the first embodiment, when the annealing method execution unit 3 assigns a random value to be associated with any of a plurality of workstations determined by the number of workstations, the process organization data evaluation unit 4 When the resource shortage number is zero when calculating the evaluation value of the current provisional solution by using, the equipment among the work elements associated with the workstation whose total work time is longer in the current provisional solution is It is characterized by assigning a random value so that any unnecessary work element is associated with another workstation.
[0112]
In the example of FIG. 14, among the work elements, the work elements “A”, “C”, “E”, “I”, and “L” are “S1”, “S2”, “S2”, and “S3”, respectively. , "S3" is required. In this case, among the work stations having a longer total work time, for example, among the work elements “E”, “H”, and “L” assigned to “work station 3”, “E” and “L” install equipment. Since “H” does not require equipment, a random value is assigned to the work element “H” so as to be associated with another workstation, as shown in the example of FIG. In the example of FIG. 15, a random value associated with “workstation 4” is assigned.
[0113]
According to the twelfth embodiment, a feasible solution with high knitting efficiency can be obtained.
[0114]
(Thirteenth embodiment)
In the twelfth embodiment, the process organizing apparatus of the thirteenth embodiment according to the twelfth embodiment is configured such that, among the work elements associated with the workstation having the longest total work time in the current provisional solution, a work element that does not require equipment is provided. It is characterized in that a random value is assigned so as to associate this with another workstation.
[0115]
According to the thirteenth embodiment, a feasible solution with high knitting efficiency can be obtained.
[0116]
(14th embodiment)
According to the process organization apparatus of the fourteenth embodiment of the present invention, in the twelfth embodiment, among the work elements associated with the workstation having the longest total work time in the current provisional solution, there are a plurality of work elements that do not require equipment. For example, a random value is assigned so that all of them are associated with another workstation.
[0117]
According to the fourteenth embodiment, a feasible solution with high knitting efficiency can be obtained.
[0118]
(Fifteenth embodiment)
FIG. 16 is a diagram showing an example of a provisional solution generated by the process organizing apparatus of the fifteenth embodiment according to the present invention.
[0119]
According to the process organization device of the fifteenth embodiment, in the first embodiment, when the annealing method execution unit 3 assigns a random value to be associated with one of a plurality of workstations determined by the number of workstations, the process organization data evaluation unit 4 When the resource shortage number is zero when calculating the evaluation value of the current provisional solution by using, the equipment among the work elements associated with the workstation whose total work time is longer in the current provisional solution is If there is a required work element, a random value is assigned so as to associate the equipment and the work element with another workstation.
[0120]
In the example of FIG. 14, among the workstations having a longer total work time, for example, among the work elements “E”, “H”, and “L” assigned to “workstation 3”, “E” and “L” are used. Since "" requires equipment and "H" does not require equipment, at least a work element for which any equipment is required and the equipment identifier of that equipment are random values so as to be associated with another workstation. Is assigned. In the example of FIG. 16, a random value associated with “workstation 4” is assigned to the work element “E” and the equipment identifier “S2”.
[0121]
According to the fifteenth embodiment, a provisional solution with a good load balance can be obtained, and an executable solution with high knitting efficiency can be obtained.
[0122]
(Sixteenth embodiment)
According to the process organization device of the sixteenth embodiment of the present invention, in the fifteenth embodiment, among the work elements associated with the workstation having the longest total work time in the current provisional solution, there is a work element requiring equipment. For example, a random value is assigned so as to associate these facilities and work elements with another workstation.
[0123]
According to the sixteenth embodiment, a feasible solution with high knitting efficiency can be obtained.
[0124]
(Seventeenth embodiment)
The process organizing apparatus according to the seventeenth embodiment of the present invention is different from the fifteenth embodiment in that, in the fifteenth embodiment, among the work elements associated with the workstation having the longer total work time in the current provisional solution, a plurality of work elements requiring equipment are provided. If any, a random value is assigned so that all of the plurality of facilities and work elements are associated with another workstation.
[0125]
In the case of the example of FIG. 14, a random value is assigned to the work element “L” and the equipment identifier “S3” and also to the work element “E” and the equipment identifier “S2” so as to be associated with another workstation. Can be
[0126]
According to the seventeenth embodiment, a feasible solution with high knitting efficiency can be obtained.
[0127]
(Eighteenth embodiment)
FIGS. 17 and 18 are diagrams showing examples of provisional solutions generated by the process organizing apparatus according to the eighteenth embodiment of the present invention.
[0128]
According to the process organization device of the eighteenth embodiment, in the first embodiment, when the annealing method execution unit 3 assigns a random value to be associated with any of a plurality of workstations determined by the number of workstations, the process organization data evaluation unit 4 When the resource shortage number is zero when calculating the evaluation value of the current provisional solution by using, the equipment among the work elements associated with the workstation with the shortest total work time in the current provisional solution is It is characterized in that a random value is assigned so that there is a work element that is not required alone and a work element that requires equipment is associated with another work station together with the work element.
[0129]
In the example of FIG. 17, among the work elements, the work elements “A”, “C”, “E”, “I”, and “L” are “S1”, “S2”, “S2”, and “S3”, respectively. , "S3" is required. In this case, among the work elements assigned to the “workstation 5” having the shortest total work time, if there is a work element that does not require equipment, it is used alone, and if there is a work element that requires equipment, another work element is added together with the equipment. You will assign a random value to associate with the workstation. In the example of FIG. 17, only the work element “J” is used, and no equipment is required. Therefore, as shown in the example of FIG. 18, a random value to be associated with another “workstation 1” by itself is assigned.
[0130]
According to the eighteenth embodiment, the number of workstations can be reduced, so that an executable solution with a high operating rate can be obtained. In addition, a load-balanced solution is obtained, and a solution can be obtained in a shorter calculation time.
[0131]
(19th embodiment)
According to the nineteenth embodiment of the process knitting apparatus according to the present invention, in the first embodiment, the initial process knitting data generation unit 2 and the annealing method execution unit 3 are provided with a plurality of work stations determined by the number of work stations for work elements. When assigning a random value to be associated with any one of the above, if there is equipment required for the work element, the same random value is assigned to the work element and the equipment.
[0132]
In other words, work elements that require equipment and the equipment are treated as one group. For example, each combination of the equipment identifier (equipment) and work element (possible work) as shown in the following (Table 12) is regarded as one group work element, and the initial process is organized. Then, the annealing method execution unit 3 is executed.
[0133]
[Table 12]

[0134]
According to the nineteenth embodiment, by grouping, a process can be organized at a high speed, and a solution without a shortage of equipment required by a work element in a workstation assignment area can be generated.
[0135]
【The invention's effect】
As is apparent from the above, the process organization device according to the first aspect of the present invention relates to a plurality of equipment identifiers indicating a plurality of types of equipment, in which the number of owned equipment having the equipment identifier is associated with the equipment identification name. A piece of equipment data associated with at least one work element indicating a work executable by the equipment having the equipment identifier, worker data including a plurality of worker identifiers respectively indicating a plurality of workers, For each of the product numbers, for each of a plurality of work elements indicating a plurality of works required to produce one product of the product number, if at least one other work element is in a leading relationship with the work element, The preceding element indicating that, the work time required for the work of the work element, the number of workers if work is required for the work element, and the number of workers if work is required for the work element The setting For each product number corresponding to a plurality of product numbers to be produced, the production number of the product having the corresponding product number is associated with the product data associated with the identification name, and the operation time for the product having the plurality of product numbers is associated with the product number. And storage means for storing input data including the number of workstations, and based on a plurality of product numbers obtained from the input data of the storage means, a plurality of product numbers corresponding to the product numbers of the storage means. And a plurality of facility identifiers and a plurality of worker identifiers are read from the facility data and the worker data in the storage means, respectively, and the plurality of work elements, the plurality of facility identifiers and the plurality of workers are read. Assign a random value to each of the distinguished names to associate with one of a plurality of workstations determined by the number of workstations Initial solution generating means for obtaining the organization contents of a plurality of workstations based on the work elements, equipment identifiers, and operator identifiers associated with these random values, and providing an initial solution of the provisional solution, and a new provisional solution from the provisional solution by emergence calculation Emergence calculation means for obtaining an optimal solution while generating a solution; and for a plurality of provisional solution workstations, a process organization efficiency is obtained by referring to a work time in the product data of the storage means, and At least the number of resources shortage is obtained by referring to the equipment identifier for the work element in the product data, the number of times of movement between workstations is obtained from the preceding element in the product data of the storage means, Obtain the excess time of each workstation with respect to the operating time, calculate the evaluation value by a calculation weighting these, Since evaluation value calculation means used for emergence calculation of emergence calculation means is provided, a random value to be associated with any of a plurality of workstations is assigned to a work element, equipment identifier, and worker identifier, and emergence is performed. By solving the optimization problem by calculation, the knitting contents of a plurality of workstations for producing a plurality of product numbers can be obtained as (quasi) optimal solutions.
[0136]
According to a second aspect of the present invention, in the process knitting apparatus according to the first aspect, the input data stored in the storage means includes an annealing method temperature set value and an annealing method temperature update coefficient setting for lowering the temperature. Value, and the emergence calculation means replaces a part of each of the random values associating a work element, a facility identifier, and a worker identifier with a plurality of workstations of the provisional solution. By assigning a new random value to be associated with any of the plurality of workstations determined by the number of stations, a process of obtaining the organization contents of the plurality of workstations as candidates for a new provisional solution, and using the evaluation value calculation means Calculating an evaluation value of the provisional solution as the candidate, and if the evaluation value is a value that is better than the evaluation value of the current provisional solution obtained last time, the candidate is On the other hand, if it is not a provisional solution, otherwise, each of the two evaluation values is regarded as an energy amount with a smaller evaluation being better, and the difference between these two energy amounts is divided by the annealing method temperature set value to obtain a division value. Calculating a probability by calculating an exponential function using the minus value of the divided value as a variable, and processing the candidate as a new provisional solution or keeping the current provisional solution in accordance with the probability. The process is repeated until a predetermined search condition is satisfied. If the search condition is satisfied, the annealing method temperature set value is updated by multiplying the annealing method temperature set value by the annealing method temperature update coefficient set value, The annealing process is repeated until the stop condition is satisfied, so that the knitting contents of a plurality of workstations for producing a plurality of product numbers can be obtained as (quasi) optimal solutions. That.
[0137]
According to a third aspect of the present invention, in the process knitting apparatus according to the first or second aspect, when the evaluation value calculating means obtains the process knitting efficiency, the evaluation value calculating means refers to a work time in the product data of the storage means. Since the process organization efficiency is obtained for each product number for a plurality of product numbers to be produced, the process organization efficiency is obtained for each product number, and the load of each workstation is leveled. Even when only a single product is produced, efficient process knitting with good knitting efficiency can be performed.
[0138]
According to a fourth aspect of the present invention, in the process organizing apparatus according to the second aspect, the initial solution generation means and the emergence calculation means assign a random value to be associated with one of a plurality of workstations determined by the number of workstations. Since the same random value is assigned to a work element similar to the work element, by grouping work elements having part numbers having similar work elements, the number of combinations at the time of emergence calculation is reduced, and the process organization can be performed at high speed. It can be carried out. In addition, even if the part numbers are different, the intended process can be organized when there is a work element desired to be performed on the same workstation.
[0139]
According to a fifth aspect of the present invention, in the process knitting apparatus according to the second aspect, when the emergence calculation means obtains the knitting contents of a plurality of workstations as candidates for the new provisional solution, first, the equipment identification name and the work A new random value is assigned to the worker identifier, and then a new random value is assigned to the work element so that it matches the equipment identifier and the worker identifier to which the random value has been assigned. The work elements requiring operators are assigned to the equipment and the workstations assigned to the workers, so that the number of resource shortages required by the evaluation value calculation means can be reduced to zero, and a feasible solution can be reliably obtained. Can be.
[0140]
According to a sixth aspect of the present invention, in the process organization device according to the second aspect, an output unit is further provided, and a plurality of optimal solutions are obtained by the emergence calculation unit and output from the output unit. Thereby, it is possible to easily obtain an executable plan.
[0141]
According to a seventh aspect of the present invention, in the process knitting apparatus according to the second aspect, when the initial solution generating means and the emergence calculating means determine the knitting contents of a plurality of workstations, the initial solution generating means and the emergence calculating means perform pre-processing of the plurality of workstations to be produced. Since the content of the organization of a plurality of workstations is obtained for each product number of the product of the above product number and the group of the work elements allocated to each workstation is used as the work element used in the emergence calculation, the process is organized at high speed. In addition, process knitting with good knitting efficiency can be obtained for each product number. Further, the number of times of movement between workstations can be reduced.
[0142]
The invention according to claim 8 is the process organization apparatus according to claim 2, wherein a plurality of workstations determined by the number of workstations in the input data stored in the storage means and a product stored in the storage means. The plurality of work elements in the data are associated with any of a plurality of steps, and the initial solution generation means and the emergence calculation means determine which of the plurality of workstations is determined by the number of work stations for the work element. When assigning a random value to be associated with a work element, a random value to be assigned to a workstation associated with the same process as the process associated with that work element is assigned, so that the number of workstation candidates that can be assigned to each work element is small. Thus, the process organization can be calculated at high speed. In addition, the manner in which the work elements of all product numbers are moved between the workstations is similar, and the flow of the work does not reverse according to the product number.
[0143]
According to a ninth aspect of the present invention, in the process organizing apparatus according to the second aspect, when the emergence calculation means assigns a random value to be associated with one of a plurality of workstations determined by the number of the workstations, the evaluation value calculation means When the resource shortage number is zero when calculating the evaluation value of the current provisional solution by using, the equipment among the work elements associated with the workstation whose total work time is longer in the current provisional solution is If a work element that is not necessary is assigned a random value so as to be associated with another workstation, a workable solution with high organization efficiency can be obtained.
[0144]
According to a tenth aspect of the present invention, in the process organizing apparatus according to the second aspect, when the emergence calculation means assigns a random value to be associated with any of a plurality of workstations determined by the number of the workstations, the evaluation value calculation means When the resource shortage number is zero when calculating the evaluation value of the current provisional solution by using, the equipment among the work elements associated with the workstation whose total work time is longer in the current provisional solution is If necessary work elements are assigned random values to associate these equipment and work elements with another workstation, a load-balanced provisional solution can be obtained, and a workable solution with high knitting efficiency can be obtained. become able to.
[0145]
According to an eleventh aspect of the present invention, in the process knitting apparatus according to the second aspect, when the emergence calculation means assigns a random value to be associated with one of a plurality of workstations determined by the number of the workstations, the evaluation value calculation means When the resource shortage number is zero when calculating the evaluation value of the current provisional solution by using, the equipment among the work elements associated with the workstation with the shortest total work time in the current provisional solution is Random values are assigned to work elements that do not need to be used alone and work elements that require equipment to be associated with equipment and to another workstation, so that the number of workstations can be reduced and the operation rate is high. A feasible solution can be obtained. In addition, a load-balanced solution is obtained, and a solution can be obtained in a shorter calculation time.
[0146]
According to a twelfth aspect of the present invention, in the process organizing apparatus according to the second aspect, the initial solution generation means and the emergence calculation means may be configured such that, for a work element, one of a plurality of workstations determined by the number of work stations. When assigning a random value to be associated, if there is equipment required for the work element, the same random value is assigned to the work element and the equipment, so that the process can be organized at high speed and the work element can be assigned in the workstation assignment area. A solution that does not lack the required equipment can be generated.
[0147]
In the invention according to claim 13, the number of owned equipment having the equipment identifier is associated with each of a plurality of equipment identifiers respectively indicating a plurality of types of equipment, and the work executable by the equipment having the equipment identifier is performed. Equipment data associated with at least one of the indicated work elements, worker data including a plurality of worker identifiers respectively indicating a plurality of workers, and one product of the product number for each of a plurality of product numbers. For each of a plurality of work elements indicating a plurality of works necessary to perform the work, a preceding element indicating that at least one other work element is in a preceding relationship with the work element, and a work for the work of the work element. Product data in which the required work time, the number of workers if work is needed for the work element, and the equipment identification name if equipment is needed for work on the work element When, For each product number of a plurality of product numbers to be produced, the production number of the product of the product number is associated, and input data including the operation time and the number of workstations for the product products of the plurality of product numbers is stored. A method for organizing processes, which is executed by a computer accessible to a storage unit to perform, based on a plurality of product numbers obtained from input data of the storage unit, a plurality of product numbers of each corresponding product number from the product data of the storage unit. A work element is read, and a plurality of equipment identification names and a plurality of worker identification names are respectively read from the equipment data and the worker data in the storage means, and the plurality of work elements, the plurality of equipment identification names, and the plurality of worker identifications are read. For each name, a random value associated with one of the workstations determined by the number of workstations An initial solution generation step of obtaining the organization contents of a plurality of workstations based on the work elements, equipment identifiers, and operator identifiers associated with these random values and setting the initial content of the provisional solution; and An emergence calculation step of finding an optimal solution while generating a new provisional solution, wherein in the emergence calculation step, a work time in the product data of the storage means is referred to for a plurality of workstations of the provisional solution. To obtain the number of resources shortage by referring to the equipment identifier for at least the work element in the product data in the storage means, and to move from workstation to workstation from the preceding element in the product data in the storage means. The number of times, the excess time of each workstation with respect to the operation time in the input data of the storage means is determined, and these are weighted. Since the step of calculating the evaluation value by the attached calculation is executed, the knitting contents of a plurality of workstations for producing a plurality of product numbers can be obtained as (quasi) optimal solutions.
[0148]
According to a fourteenth aspect of the present invention, the process scheduling program is executed by the equipment having the equipment identification name while the number of the equipments having the equipment identification name is associated with each of the plurality of equipment identification names indicating the plurality of types of equipment. Equipment data in which at least one work element indicating a possible work is associated, worker data including a plurality of worker identifiers respectively indicating a plurality of workers, and a plurality of product numbers for each product number For each of a plurality of work elements respectively indicating a plurality of works required to produce one of the work elements, a precedent element indicating at least one other work element in a precedent relation to the work element, The work time required for the work of the element, the number of workers if work is needed for the work element, and the equipment identification name if equipment is needed for work of the work element are associated with each other. For each product number corresponding to a plurality of product numbers to be produced, the production number of the product having the product number is associated, and the operation time and the number of workstations for the product products having the plurality of product numbers are associated with each other. For a computer that can access a storage unit that stores input data including the input data, based on a plurality of product numbers obtained from the input data of the storage unit, a plurality of operations of the corresponding product numbers from the product data of the storage unit are performed. Read out a plurality of facility identifiers and a plurality of worker identifiers from the facility data and the worker data of the storage means, and read the plurality of work elements, the plurality of facility identifiers and the plurality of worker identifiers. Allocate a random value to be associated with any of a plurality of workstations determined by the number of workstations, An initial solution generation function that obtains the organization contents of a plurality of workstations based on the work elements, equipment identifiers, and operator identifiers associated with these random values and sets the initial solution of the provisional solution, and a new solution from the provisional solution by emergence calculation An emergence calculation function for finding an optimum solution while generating a provisional solution; and for a plurality of workstations of the provisional solution, a process organization efficiency is determined by referring to a work time in the product data of the storage means. The number of resources shortage is obtained by referring to the equipment identifier for at least the work element in the product data, and the number of times of movement between workstations is obtained from the preceding element in the product data of the storage means. Calculating the excess time of each workstation with respect to the operation time of the work station, and calculating an evaluation value by weighting these work times. Since the evaluation value calculation function used for the emergence calculation of the generation calculation function is added, the organization contents of a plurality of workstations for producing products of a plurality of product numbers can be obtained as (quasi) optimal solutions.
[Brief description of the drawings]
FIG. 1 is a configuration diagram of a process organizing apparatus according to a first embodiment of the present invention.
FIG. 2 is a diagram showing a project network for a product of each product number.
FIG. 3 is a diagram showing a network of projects for products of each product number.
FIG. 4 is an operation flowchart of the process organization device.
FIG. 5 is a diagram showing an example of a provisional solution.
FIG. 6 is a diagram showing an example of a provisional solution.
FIG. 7 is a diagram showing an example of a provisional solution.
FIG. 8 is a diagram showing an example of a provisional solution.
FIG. 9 is a diagram showing an example of a provisional solution.
FIG. 10 is an explanatory diagram of calculation of the process knitting efficiency by the process knitting apparatus according to the second embodiment of the present invention.
FIG. 11 is an explanatory diagram of calculating the process organization efficiency.
FIG. 12 is a diagram showing an example of a pre-process for obtaining a provisional solution generated by the process composition device of the ninth embodiment according to the present invention.
FIG. 13 is a diagram showing an example of the preprocessing.
FIG. 14 is a diagram showing an example of a provisional solution generated by the process composition device of the twelfth embodiment according to the present invention.
FIG. 15 is a diagram showing an example of a provisional solution generated by the process composition device.
FIG. 16 is a diagram showing an example of a provisional solution generated by the process composition device of the fifteenth embodiment according to the present invention.
FIG. 17 is a diagram showing an example of a provisional solution generated by the process organizing apparatus of the eighteenth embodiment according to the present invention.
FIG. 18 is a diagram showing an example of a provisional solution generated by the process composition device.
[Explanation of symbols]
1 Process data storage
2 Initial process data generation unit
3 Annealing method execution part
4 Process data evaluation section
5 Process data output section

Claims (14)

For each of a plurality of equipment identifiers indicating a plurality of types of equipment, the number of owned equipment of the equipment identifier is associated with, and at least one work element indicating work executable by the equipment of the equipment identifier is associated. Attached equipment data,
Worker data including a plurality of worker identifiers respectively indicating a plurality of workers,
For each of a plurality of product numbers, for each of a plurality of work elements indicating a plurality of works required to produce one product of the product number, at least one other work element has a precedence relationship with the work element. If there is a preceding element that indicates this, the work time required for the work of the work element, the number of workers if work is needed for the work element, and equipment required for the work of the work element. If the product data associated with the equipment identifier,
For each product number of a plurality of product numbers to be produced, the production number of the product of the product number is associated, and input data including the operation time and the number of workstations for the product products of the plurality of product numbers is stored. Storage means for
Based on a plurality of product numbers obtained from the input data of the storage unit, a plurality of work elements of each corresponding product number are read from the product data of the storage unit, and a plurality of work elements are read from the facility data and the worker data of the storage unit. The plurality of equipment identifiers and the plurality of worker identifiers are read out, and for each of the plurality of work elements, the plurality of equipment identifiers and the plurality of worker identifiers, any one of the plurality of workstations determined by the number of workstations. Initial solution generating means for assigning random values to be associated with each other, obtaining the organization contents of a plurality of workstations based on the work elements, equipment identifiers and operator identifiers associated with these random values, and as an initial solution of a provisional solution,
Emergence calculation means for finding an optimal solution while generating a new provisional solution from the provisional solution by emergence calculation;
For a plurality of provisional workstations, determine the process organization efficiency by referring to the work time in the product data of the storage means, and refer to the equipment identifier for at least the work element in the product data of the storage means. To determine the number of resources shortage, obtain the number of times of movement between workstations from the preceding element in the product data of the storage means, obtain the excess time of each workstation with respect to the operating time in the input data of the storage means, A process organizing apparatus for calculating an evaluation value by a weighted operation, comprising: an evaluation value calculation unit used for emergence calculation by the emergence calculation unit. The input data stored in the storage means further includes an annealing method temperature set value, and an annealing method temperature update coefficient setting value for lowering the temperature,
The emergence calculation means,
For the plurality of workstations of the provisional solution, instead of a part of each random value that associates the work element, the equipment identifier and the worker identifier, any of the plurality of workstations determined by the number of workstations By assigning a new random value to be associated, a process of obtaining the organization contents of a plurality of workstations as a candidate of a new provisional solution, and calculating an evaluation value of the provisional solution as the candidate using the evaluation value calculation means Then, if this evaluation value is a value that is better than the evaluation value of the current provisional solution obtained last time, the candidate is set as a new provisional solution, otherwise, each of the two evaluation values is The smaller the value, the better the evaluation of energy, and the difference between these two amounts of energy is annealed and divided by the method temperature set value to obtain a division value, and the minus value of this division value is used as a variable A probability is obtained by calculating with an exponential function, and according to the probability, a process of setting the candidate as a new provisional solution or keeping the current provisional solution is a predetermined process as an annealing process. Repeat until
If the search condition is satisfied, the annealing process is repeated until the predetermined stopping condition is satisfied, while updating the annealing method temperature set value by multiplying the annealing method temperature set value by the annealing method temperature update coefficient setting value. The process knitting apparatus according to claim 1, wherein: The evaluation value calculation unit, when obtaining the process organization efficiency, refers to a work time in the product data of the storage unit and obtains the process organization efficiency for each product number of the plurality of product numbers to be produced. 3. The process organizing apparatus according to claim 1, wherein: When assigning a random value to be associated with any of a plurality of workstations determined by the number of workstations, the initial solution generation unit and the emergence calculation unit assign the same random value to a work element similar to the work element 3. The process knitting apparatus according to claim 2, wherein: When the emergence calculation means seeks the organization contents of a plurality of workstations as candidates for the new provisional solution, first, a new random value is assigned to the equipment identifier and the worker identifier, and then the random value is 3. The process organizing apparatus according to claim 2, wherein a new random value is assigned to the work element so as to match the assigned equipment identifier and worker identifier. 3. The apparatus according to claim 2, further comprising an output unit, wherein the emergence calculation unit obtains a plurality of optimal solutions and outputs the plurality of optimal solutions from the output unit. When the initial solution generation means and the emergence calculation means determine the organization content of a plurality of workstations, as preprocessing, determine the organization content of a plurality of workstations for each product number of the plurality of product numbers to be produced. 3. The apparatus according to claim 2, wherein a group of work elements allocated to each workstation is used as a work element used in the emergence calculation. A plurality of workstations determined by the number of workstations in the input data stored in the storage means and a plurality of work elements in the product data stored in the storage means are associated with any of a plurality of processes. And
When assigning a random value to be associated with any of a plurality of workstations determined by the number of workstations to the work element, the initial solution generation means and the emergence calculation means are the same as the steps associated with the work element 3. The process organizing apparatus according to claim 2, wherein a random value to be associated with the workstation associated with the process is assigned. In the case where the emergence calculation means assigns a random value to be associated with any of the plurality of workstations determined by the number of workstations, when calculating the evaluation value of the current provisional solution using the evaluation value calculation means, the resource shortage occurs. When the number is zero, a work element associated with a workstation with a longer total working time in the current provisional solution and which does not require equipment is randomly assigned to be associated with another workstation. 3. The apparatus according to claim 2, wherein values are assigned. In the case where the emergence calculation means assigns a random value to be associated with any of the plurality of workstations determined by the number of workstations, when calculating the evaluation value of the current provisional solution using the evaluation value calculation means, the resource shortage occurs. When the number is zero, if any of the work elements associated with the workstation having a longer total work time in the current provisional solution require work equipment, these work equipment and work elements are transferred to another work station. 3. The apparatus according to claim 2, wherein a random value is assigned so as to be associated with the process. When the emergence calculation means assigns a random value to be associated with any of the plurality of workstations determined by the number of workstations, when calculating the evaluation value of the current provisional solution using the evaluation value calculation means, the resource shortage occurs. When the number is zero, among the work elements associated with the workstation that has the shortest total work time in the current provisional solution, any work elements that do not require equipment alone require work equipment. 3. The process organizing apparatus according to claim 2, wherein if there is, a random value is assigned so as to be associated with another workstation together with the equipment. The initial solution generation means and the emergence calculation means, when assigning a random value to be associated with any of the plurality of workstations determined by the number of workstations to the work element, if there is equipment required for the work element, 3. The process organizing apparatus according to claim 2, wherein the same random value is assigned to the work element and the equipment. For each of a plurality of equipment identifiers indicating a plurality of types of equipment, the number of owned equipment of the equipment identifier is associated with, and at least one work element indicating work executable by the equipment of the equipment identifier is associated. Attached equipment data,
Worker data including a plurality of worker identifiers respectively indicating a plurality of workers,
For each of a plurality of product numbers, for each of a plurality of work elements indicating a plurality of works required to produce one product of the product number, at least one other work element has a precedence relationship with the work element. If there is a preceding element that indicates this, the work time required for the work of the work element, the number of workers if work is needed for the work element, and equipment required for the work of the work element. If the product data associated with the equipment identifier,
For each product number of a plurality of product numbers to be produced, the production number of the product of the product number is associated, and input data including the operation time and the number of workstations for the product products of the plurality of product numbers is stored. A process organizing method executed by a computer accessible to a storage unit,
Based on a plurality of product numbers obtained from the input data of the storage unit, a plurality of work elements of each corresponding product number are read from the product data of the storage unit, and a plurality of work elements are read from the facility data and the worker data of the storage unit. The plurality of equipment identifiers and the plurality of worker identifiers are read out, and for each of the plurality of work elements, the plurality of equipment identifiers and the plurality of worker identifiers, any one of the plurality of workstations determined by the number of workstations. Assigning a random value to be associated with the work element, the work element associated with the random value, the equipment identification name and the operator identification name, determine the contents of a plurality of workstations by the initial solution generation step as an initial solution of a provisional solution,
An emergence calculation step of finding an optimal solution while generating a new provisional solution from the provisional solution by emergence calculation,
In the emergence calculation step, for a plurality of provisional solution workstations, a process organization efficiency is obtained by referring to a work time in the product data of the storage means, and at least a work element in the product data of the storage means is obtained. The number of resources shortage is obtained by referring to the equipment identifier for the above, the number of times of movement between workstations is obtained from the preceding element in the product data of the storage means, and the excess of each workstation with respect to the operation time in the input data of the storage means. A step of calculating time and calculating an evaluation value by an operation in which the time is weighted; For each of a plurality of equipment identifiers indicating a plurality of types of equipment, the number of owned equipment of the equipment identifier is associated with, and at least one work element indicating work executable by the equipment of the equipment identifier is associated. Attached equipment data,
Worker data including a plurality of worker identifiers respectively indicating a plurality of workers,
For each of a plurality of product numbers, for each of a plurality of work elements indicating a plurality of works required to produce one product of the product number, at least one other work element has a precedence relationship with the work element. If there is a preceding element that indicates this, the work time required for the work of the work element, the number of workers if work is needed for the work element, and equipment required for the work of the work element. If the product data associated with the equipment identifier,
For each product number of a plurality of product numbers to be produced, the production number of the product of the product number is associated, and input data including the operation time and the number of workstations for the product products of the plurality of product numbers is stored. Computer that can access the storage means
Based on a plurality of product numbers obtained from the input data of the storage unit, a plurality of work elements of each corresponding product number are read from the product data of the storage unit, and a plurality of work elements are read from the facility data and the worker data of the storage unit. The plurality of equipment identifiers and the plurality of worker identifiers are read out, and for each of the plurality of work elements, the plurality of equipment identifiers and the plurality of worker identifiers, any one of the plurality of workstations determined by the number of workstations. An initial solution generation function that assigns random values to be associated with the work items, obtains the organization contents of a plurality of workstations by the work elements, equipment identifiers, and operator identifiers associated with these random values, and provides an initial solution of a provisional solution,
An emergence calculation function that seeks the optimal solution while generating a new provisional solution from the provisional solution by emergence calculation,
For a plurality of provisional workstations, determine the process organization efficiency by referring to the work time in the product data of the storage means, and refer to the equipment identifier for at least the work element in the product data of the storage means. To determine the number of resources shortage, obtain the number of times of movement between workstations from the preceding element in the product data of the storage means, obtain the excess time of each workstation with respect to the operating time in the input data of the storage means, A process scheduling program for calculating an evaluation value by a weighted calculation, and adding an evaluation value calculation function used for emergence calculation of the emergence calculation function.

Images