JP3809790B2 - Vending machine product configuration adjustment support system, method and recording medium - Google Patents

Description

【０１２８】
全てのロケについて、上記処理を完了すると（ステップＳ６９，ＹＥＳ）、続いて、特に上記粗利率Ｒａと総粗利率Ｒａllの比に基づいて、アンバランスロケを抽出し、これを出力（または表示）する（ステップＳ７０）。すなわち、全体と比較してコスト／利益のバランスが極端にとれていないロケを調整対象候補として抽出／出力する。
【０１２９】
次に、図２６を参照して、商品系のコスト／利益バランス分析部１６の処理内容について説明する。
まず、売上実績情報Ａ４、商品分類情報Ａ３、商品売価、原価情報Ａ６をデータベースから取得する（ステップＳ６７）。
【０１３０】
次に、これら取得したデータに基づいて、各ロケ毎に、上記ロケ別の粗利率Ｒａと、ロケ別、商品系別の粗利率Ｒｋを算出し、更にＲａとＲｋの比を求める （ステップＳ７３）。
【０１３１】
ここで、各ロケ毎における商品系別の粗利率Ｒｋは、以下の式により求める。
【０１３２】
【数２】

【０１３３】
全てのロケについて、上記処理を完了すると（ステップＳ７４，ＹＥＳ）、続いて、特に上記粗利率ＲａとＲｋの比に基づいて、アンバランス商品系を抽出し、これを出力（または表示）する（ステップＳ７５）。すなわち、ステップＳ７３までの処理結果により（これを表示させて）、各ロケ毎に、そのロケ全体の粗利率Ｒａと比較してコスト／利益のバランスが極端に取れていない商品系（アンバランス商品系）が存在するか否かを分析者等が判断することができ、このようなアンバランス商品系を抽出すると共に、このようなアンバランス商品系が存在するロケを、後の処理における調整対象ロケの追加候補とすることができる。
【０１３４】
尚、上記のようにコスト／利益のバランスを分析するのは、自動販売機がオペレータ会社のものであっても、メーカー会社のものであっても有効な意味がある。オペレータ会社とは、自販機を購入（貸し）して、中身商品も商品メーカから仕入れて、自動販売機で商品を販売する会社のことである。特徴は、同一自販機内に複数の商品メーカーの商品が置かれていることと、商品原価がメーカー／品種により差がある（大差があるものも存在する）点である。よって、オペレータ会社にとっては、売価が同じ商品同士であれば、原価が安い商品（商品１個当たりの粗利が高い）が多く売れるのが望ましいが、原価の安い商品は通常不人気商品である場合が多く、売上数が低い為、結局、「粗利×売上数」は低い（儲けが薄い）。一方で、売上数が多い商品（人気商品）は、通常、原価が高い（商品１個当たりの粗利が低い）傾向にある。よって、多く売れても儲けが薄い場合がある（但し、上記のことは、全般的な傾向を言っているのであり、個々の商品毎に見れば、例えば原価が安い商品であっても多く売れる商品があるかもしれない）。
【０１３５】
上述した事情がある故に、特にオペレータ会社にとっては、コスト／利益のバランスを取ることは、重要なことであり、上記アンバランス商品系が存在するロケを、調整対象ロケの追加候補とすることは、非常に意味があることである。
【０１３６】
一方で、メーカー会社の場合、自販機で扱う商品は自社商品だけである。すなわち、１つの自販機に、単一商品メーカーの商品しか置かれていない。しかし、商品原価は品種により差があるので、やはり、コスト／利益のバランスを分析するのは有効な意味があることである。
【０１３７】
調整対象ロケ抽出処理部１０−１は、上述してきた特異ロケ分析部１０−２、ロケ競合分析部５、商品構成バランス分析部４、ロケのコスト／利益バランス分析部１５、商品系のコスト／利益バランス分析部１６の処理結果、及び属性でのロケ分類情報Ｂ３、売上でのロケ分類情報Ｂ４、売上実績情報Ａ４を用いて（但し、これら全てを用いるとは限らない）、図２７に示す処理を実行し、調整対象とすべきロケの候補を選出する。
【０１３８】
以下、図２７〜図２９を参照して、調整対象ロケ抽出処理部１０−１について説明する。
図２７は、調整対象ロケ抽出処理部１０−１における処理手順を説明する為のフローチャート図である。
【０１３９】
まず、上記属性特徴からのロケ分類処理部３−１により作成された、属性によるロケ分類情報Ｂ３と、データベースに格納されている売上実績情報Ａ４とを取得する（ステップＳ７７）。但し、これは一例であり、売上でのロケ分類情報Ｂ４を用いるようにしてもよい。
【０１４０】
次に、ステップＳ７８、ステップＳ７９の処理により、各グループ（ＧｒＬ）毎に、非特異ロケのロケ／商品系売上占有率リストＣ２０（各グループの主力ロケ／メイン商品系表）を作成する（図２９はＧｒＬ１の例を示すが、ＧｒＬ２、ＧｒＬ３についても同様にして作成；また図２９には特異ロケのリストＣ１０も示してある）。
【０１４１】
ロケ分類情報Ｂ３（上記の例ではＧｒＬ１、ＧｒＬ２、ＧｒＬ３）に従って、各グループ（ＧｒＬ）毎に、そのグループに属する全てのロケについて、ロケ別の商品系売上占有率ＡＢＣ分析を行い（ステップＳ７８）、更に特異ロケリストＣ９から逆に非特異ロケが分かるので、非特異ロケの各商品系の売上占有率の平均を算出する。例えば図２９の例では、非特異ロケ（ロケＮｏ＝１１、８、６、・・・２２）の各商品系の売上占有率の平均は、商品系Ｃが49.04、商品系Ｎが14.02等となっており、図２８はこれを視覚的に分かり易いように棒グラフ化したものである。そして、各商品系を売上占有率の平均の高い順にソーティングする（図２９において、左から商品系Ｃ、Ｎ、Ｗ、Ｄ・・・の順になっているのは、ソーティングした結果である）（ステップＳ７９）。
【０１４２】
このように、各グループ毎に、そのグループの特徴を表わす上で特異な存在となっている特異ロケを除いたロケ（すなわち、非特異ロケ）の商品系売上占有率の平均を算出し、商品系を売上占有率の平均値の大―小順にソーティングすることにより、グループの売上パターンを明かにし、グループのメイン商品系が何であるか等の売上特徴を掴み易くすることができる（図２９の例ではＧｒＬ１のメイン商品系は商品系Ｃ、Ｎ、Ｗ等であることが分かる）。
【０１４３】
更に、図２９においては、非特異ロケについては総売上の上位から順にソーティングしてあり、図示の例ではロケＮｏ＝１１、８、６、・・・という順にソーティングされている。尚、特異ロケについても図示のように総売上の上位から順にソーティングしてもよいが、特異ロケの場合は単純に総売上の下位のロケを調整対象候補とするとは限らないので、ソーティングしなくてもよい。
【０１４４】
尚、さらに、非特異ロケの中で総売上上位のロケのみを対象として、各商品系の平均売上占有率を算出し、グループGrLiの総売上上位ロケ（主力ロケ）の売上パターン／売上特徴を明らかにすることもできる）。
【０１４５】
上述した処理により、各グループ毎に、非特異ロケのロケ／商品系売上占有率リストＣ２０（各グループの主力ロケ／メイン商品系表）を作成すると、これに基づいて、調整対象とすべきロケの候補（一次候補）を選定する。
【０１４６】
この選定は、非特異ロケについては、各グループ毎に総売上の下位から順に選定してもよいし（図示の例では最下位のロケ（ロケＮｏ＝２２）のみが調整対象に選定されている（図において最も左側の「調整」の項目に○が示してある））、または総売上によるＡＢＣ分析によりＣランクとなったロケを、一次候補に選定する（ステップＳ８１）。
【０１４７】
また、特異ロケについては、例えば分析者等が総合的に判断する。すなわち、図１においても説明してあるように、特異ロケの売上占有率パターンを、そのロケが属するはずのグループGrLi（または／及びグループGrSi）の非特異ロケによる商品系売上占有率の平均（グループの売上パターン；上記ステップＳ７８、Ｓ７９の処理結果）と比較することによって、どのロケの売上実績に問題が存在するかを分析者等が判断することができる。
【０１４８】
これより、例えば、図示の例では、分析者等によりロケＮｏ＝１６及び１９のロケを調整対象とすべきと判断され、指示／入力（図において最も左側の「調整」の項目に○）されている（ステップＳ８０）。
【０１４９】
尚、単純に、特異ロケの中で総売上の低いものを調整対象ロケとしてもよい。また、尚、上記ステップＳ８０、Ｓ８１の処理は、両方行なってもよいし、何れか一方を行なってもよい。
【０１５０】
全てのグループについて上述した処理を行なって、調整対象ロケの一次候補ロケリストを作成すると、この一次候補リストに、上述したアンバランスロケ等を追加していき、二次候補、三次候補、・・・等と候補リストを作成していく。尚、以下に説明するステップＳ８２〜Ｓ８５の全ての処理を行うとは限らない。 （分析者が選択／指示する等して）何れか１つ以上の処理を行なわせるようにしてもよいし、あるいは一次候補をそのまま最終候補リストとしてもよい。
【０１５１】
ステップＳＳ８２〜Ｓ８４の処理には、特に処理順番はないが、ここでは図面に沿って説明すると、商品構成バランス分析部４により作成された調整対象ロケリストへの追加情報Ｃ１６を用いて、一次候補ロケリストに、“商品構成のバランスが（極端に）とれていない”ロケを追加して、二次候補ロケリストを作成する（ステップＳ８２）。
【０１５２】
ロケのコスト／利益バランス分析部１５により求められた、全ロケの平均と比較してコスト／利益のバランスが（極端に）とれていないロケを、調整対象候補として二次候補ロケリストに追加し、三次候補ロケリストを作成する（ステップＳ８３）。
【０１５３】
商品系のコスト／利益バランス分析部１６により求められた、そのロケ全体の粗利率Ｒａと比較してコスト／利益のバランスが（極端に）とれていない商品系（アンバランス商品系）が存在するロケを、調整対象候補として三次候補ロケリストに追加し、四次候補リストを作成する（ステップＳ８４）。
【０１５４】
そして、四次候補リストの中から、ロケ競合分析部５により求められた、強競合のあるロケを除外して、五次候補リストを作成する（ステップＳ８５）。
以上の処理で得たロケリストを最終候補ロケリストＣ２１として調整ロケリストを出力し、調整商品系決定処理部１１に通知する（ステップＳ８６）。
【０１５５】
図２７に示す処理は一例であり、これに限るわけではない。例えば、特異ロケ／非特異ロケに分類せずに、属性によるロケ分類情報Ｂ３または売上でのロケ分類情報Ｂ４により分類されたグループ毎に、上記ステップＳ７８、Ｓ７９、Ｓ８１と同様の処理を行うようにしてもよい。
【０１５６】
あるいは、上記ロケ分類情報Ｂ３、Ｂ４のように複数のロケを分析して調整対象ロケを決定するものに限らず、各ロケ毎に単独で処理を行なって、そのロケを調整対象ロケとすべきか否かを決定するようにしてもよい。すなわち、商品構成バランス分析部４またはロケのコスト／利益バランス分析部１５あるいは商品系のコスト／利益バランス分析部１６による分析結果のみを用いて、アンバランスロケを調整対象ロケとするようにしてもよい。
【０１５７】
次に、調整商品系決定処理部１１は、上記調整対象ロケ抽出処理部１０−１の処理結果と、商品構成バランス分析部４、及び以下に説明する商品系間の相関関係分析部７の処理結果を用いて、調整商品系を決定する処理部を行う。
【０１５８】
以下、図３０〜図３２を参照して、商品系間の相関関係分析部７について説明する。
図３０は、商品系間の相関関係分析部７における処理手順を説明する為のフローチャート図である。
【０１５９】
尚、この処理自体は、既存の分析ツールを利用して例えば図３２に示すような相関係数マトリクス（相関係数と、有意確率に基づく有意水準）が得られるので、説明は簡単に行う。
【０１６０】
まず、データベースに格納されている売上実績情報Ａ４（または、上記売上特徴分析部２の処理で得られた「商品系別売上占有率評価表」Ｃ３）、及びロケ分類情報Ｂ３またはＢ４を取得する（ステップＳ８９）。
【０１６１】
次に、各グループ毎に、グループ内の各商品系間の相関係数計算、有意分析と、相関関係マトリクスの作成処理を行う（ステップＳ９０）。これは、各グループに属するロケで扱う全ての商品系に対して、商品系の売上占有率にて相関係数を求める。
【０１６２】
図３１（ａ）、（ｂ）は商品系相関分析の説明図である。
同図（ａ）に示す式は、相関係数ｒの算出式の一例であり、これは同図（ｂ）に示す商品系Ｕと商品系Ｔとの間の相関係数ｒｕｔの算出式である。
【０１６３】
尚、相関係数は必ず−１〜１の範囲内となる。つまり、正の相関が強ければ１に近い値となり、負の相関が強ければ−１に近い値となり、相関関係が弱ければ、０に近い値となる。
【０１６４】
上述したステップＳ９０の処理を、全ての商品系の組み合わせについて実行し、これらをまとめてソーティング（ある商品系について他商品系との相関係数の値でソートする）し、図３２に示すような相関関係マトリクスＣ１７を作成し表示する（ステップＳ９１）。また、有意性計算の処理で、有意の相関係数を持つ商品系を抽出し、上記マトリクスに標識を表示する（図３２では、数値の右側に＊、＊＊を付して表示する。尚、＊は有意確率５％水準で有意（両側）、＊＊は有意確率１％水準で有意（両側）を意味する）（ステップＳ９２）。
【０１６５】
分析者は、図３２のような相関関係マトリクスを参照することで、どの商品系とどの商品系とがどの程度の相関関係があるか、また相関度合いの順を知ることができる。更に、商品系間の相関関係の有意性を把握することができる。
【０１６６】
ここで、有意性の把握について説明する。有意確率（またはＰ値という）とは、「ある１回の調査により得られたデータから統計量を計算し、その値以上の統計量が得られる確率」のことである。
【０１６７】
例えば、もし有意確率がある値より小さければ“Ｈ０が正しい”ということが疑わしくなり（棄却）、ひいてはＨ１の正当性が支持（採択）されることになる。有意確率がどの程度に小さければＨ０を棄却（Ｈ１を採択）したらよいかという基準が「有意水準」と呼ばれるものであり、これをαで表わす。上記有意確率５％水準とは、α＝０．０５を意味し、これは、同じ状況下で検定を行うと２０回に１回は決定を誤る危険性があることを意味する。あるいは、１回の決定で、Ｈ０が正しいにも係わらず誤ってＨ０を棄却する確率が５（％）である、ということである。
【０１６８】
このように、図３２のような相関関係マトリクスを参照することで、有意性を把握できる。
上述した処理により、グループ内での相関分析が完了すると（ステップＳ９３，ＹＥＳ）、更に以上のグループ内の相関分析と同じようにして、全ロケを対象とした商品系間の相関分析を行う（ステップＳ９４）。そして、この分析結果を、全ロケでの各商品系間の相関関係情報Ｃ１９として出力する。グループ内に設置されている商品系の実態は実際の設置場所特徴を反映しない場合があることに対応して、人の判断により、全体で扱っている（全範囲内の）商品系に対して相関分析を行うことにより上述の不足を補う。
【０１６９】
以上の処理によって商品／商品系間の相関性をつかむことにより、後の調整商品系決定処理部１１における商品／商品系の調整を支援する。つまり、グループ内、または全範囲内のロケにおいて、非常に売れている商品系（メイン商品系）と正の相関関係を持つ商品系が売れるはずであるから、該当商品系の品種が足りなければ品種を増やす、または該当商品系の商品をそのロケで扱って（販売して）いなければ追加する等の調整を行うことで、適切な商品構成とすることが可能となる。
【０１７０】
以上の処理において、商品系の売上占有率の代りに、商品の売上占有率を用いるようにしてもよい。
以下、図３３〜図３５を参照して、調整商品系決定処理部１１における処理について説明する。
【０１７１】
図３３〜図３５は、調整商品系決定処理部１１における処理手順を説明する為のフローチャート図（その１）（その２）（その３）である。
まず、調整対象ロケ抽出処理部１０−１により得られた、調整候補ロケリストＣ２１（最終候補ロケリスト）を表示する等して、分析者等が、調整対象ロケを指定する（ステップＳ９６）。
【０１７２】
そして、指定されたロケについて、ステップＳ９７〜ステップＳ１１８の処理を実行し、そのロケの対象商品系／調整方法リストＣ２２（調整すべき商品系とその調整方法一覧）を作成したら、次の調整対象ロケがある場合には指定して、同様の処理を繰り返し行う。尚、分析者等が指定するのではなく、調整候補ロケリストの上から順番に処理を行うようにしてもよい。
【０１７３】
まず、調整対象ロケ抽出処理部１０−１により得られたロケ／商品系売上占有率リストＣ１０、Ｃ２０（調整対象ロケが属するグループのリスト；例えば調整対象ロケがグループＧｒＬ１に属するとしたら、図２９に示すリスト）の情報を取得する（ステップＳ９７）。
【０１７４】
更に、商品構成バランス分析部４により得られたバランス情報Ｂ６を取得する（ステップＳ９８）。
まず、上記ロケ／商品系売上占有率リスト（例えば図２９）に基づいて、各商品系毎に、調整対象ロケの商品系売上占有率（ＳＲ）と、この調整対象ロケが属するグループの平均商品系売上占有率（Ｇｒ平均）とを比較して、ＳＲがＧｒ平均より高いか低いかによって、各商品系を２つに分類する（ステップＳ１００）。
【０１７５】
但し、その前に、ＳＲ＝０である商品系がある場合には（ステップＳ９９，ＹＥＳ）、この商品系については後述するステップＳ１１３〜Ｓ１１５の処理を行う。
【０１７６】
次に、上記ステップＳ９８で取得したバランス情報に基づいて、バランスがとれていない商品系を調整対象商品系と決定すると共に、その調整方法を決定する（ステップＳ１０１〜Ｓ１１２）。
【０１７７】
図示の通り、ステップＳ１０１に示す条件（アンバランス商品系とする条件；ＳＲ＞＞ＭＲ＞＞ＶＲ、ＭＲ＞＞ＳＲ＞＞ＶＲ）とステップＳ１０７に示す条件とは同じであるが、ステップＳ１００の判定結果によって、その調整方法が異なる。ステップＳ１０３とステップＳ１０９についても同様である。ステップＳ１０５とステップＳ１１１についても一部は同様である。
【０１７８】
まず、ＳＲ＜Ｇｒ平均であって（ステップＳ１００、ＹＥＳ）、ステップＳ１０１に示す条件（ＳＲ＞＞ＭＲ＞＞ＶＲまたはＭＲ＞＞ＳＲ＞＞ＶＲ）を満たす商品系が存在した場合には（ステップＳ１０１，ＹＥＳ）、その商品系を調整対象商品系と決定すると共に、その調整方法を決定する。ステップＳ１０１に示す条件を満たす場合は「その商品系において、既存（現在そのロケで販売している）商品種類が足りない」と考えられるので、「見本数分割で、同系他品を追加」という調整方法に決定する（ステップＳ１０２）。
【０１７９】
ここで、“見本数分割で”とは、例えば、その商品系がコーヒー飲料系Ｃであるとし、既存商品種が“ブラックコーヒー”であったとし、その見本数が‘５’であったとした場合、この見本数‘５’を分割（例えば、３と２）に分割することを意味する。そして、例えば、同系他品として“ミルクコーヒー”を追加するとした場合（但し、これは後の商品決定処理で決めるのであり、ここでは“仮の”話とする）、例えばこの“ミルクコーヒー”に見本数‘２’を割り当てたとすると、“ブラックコーヒー”の見本数は‘３’となる。通常、１つの見本に対しては、１つのコラムが割り当てられるので、“ブラックコーヒー”の在庫最大数が減ることになる。このように、当該調整方法は、例えば現在販売中の“ブラックコーヒー”は、他の商品と完全に入れ替える必要はないものの、取扱い数量を減らして、その分を他の商品に置き換えたほうが良い場合の調整方法である。
【０１８０】
尚、見本数を何対何に分割すべきかは、予め設定されていてもよいし、その都度分析者等が指定するようにしてもよい。
一方、ＳＲ≧Ｇｒ平均であって（ステップＳ１００、ＮＯ）、ステップＳ１０７に示す条件（ステップＳ１０１と同じ）を満たす商品系が存在した場合には （ステップＳ１０７，ＹＥＳ）、その商品系を調整対象商品系と決定すると共に、その調整方法を決定する。この場合、「その商品系は、売上占有率は高いが、商品の種類が少ない」と考えられるので、その調整方法は「同系他品を追加」となる（ステップＳ１０８）。
【０１８１】
また、ＳＲ＜Ｇｒ平均であって（ステップＳ１００、ＹＥＳ）、ステップＳ１０３に示す条件（ＭＲ＞＞ＶＲ＞＞ＳＲ）を満たす商品系が存在した場合には （ステップＳ１０３，ＹＥＳ）、その商品系を調整対象商品系と決定すると共に、その調整方法を決定する。この場合、「その商品系に不人気商品が多いか、あるいはその商品系の商品は、そのロケのユーザ層の好みに合わない」と考えられるので、その調整方法は「見本数分割で、同系他品を追加」か「同系他品で入替」となる（ステップＳ１０４）。
【０１８２】
同様にして、ＳＲ＜Ｇｒ平均であって（ステップＳ１００、ＹＥＳ）、ステップＳ１０５に示す条件を満たす商品系が存在した場合には（ステップＳ１０５，ＹＥＳ）、「同系他品で入替」を調整方法とする（ステップＳ１０６）。ＳＲ≧Ｇｒ平均であって（ステップＳ１００、ＮＯ）、ステップＳ１０９に示す条件を満たす商品系が存在した場合には（ステップＳ１０９，ＹＥＳ）、「見本数分割で、他の商品系の商品を追加」を調整方法とする（ステップＳ１１０）。ＳＲ≧Ｇｒ平均であって（ステップＳ１００、ＮＯ）、ステップＳ１１１に示す条件を満たす商品系が存在した場合には（ステップＳ１１１，ＹＥＳ）、「他の商品系の商品を追加するものとし、その為に商品種を減らす」を調整方法とする（ステップＳ１１２）。
【０１８３】
更に、ＳＲ＜Ｇｒ平均である場合において（ステップＳ１００、ＹＥＳ）、更にＳＲがＧｒ平均よりある程度以上小さい商品系が存在した場合には（ステップＳ２０１，ＹＥＳ）（図示のステップＳ２０１の例では「ＳＲ＜１／２Ｇｒ平均」を条件しているが、このようなＧｒ平均に乗ずる係数は１／２に限るものではなく、任意の値（但し、当然、１未満）を用いればよい）、その商品系を調整対象商品系と決定すると共に、その調整方法を決定する。この場合は、「同系他品を追加」となる（ステップＳ２０２）。
【０１８４】
尚、図３４では特に示していないが、上記ステップＳ１０１、Ｓ１０３、Ｓ１０５、Ｓ１０７、Ｓ１０９、Ｓ１１１の判定において“ＮＯ”の判定であった場合には、他の条件について判定していき、全ての条件で（例えばＳＲ＜Ｇｒ平均（ステップＳ１００、ＹＥＳ）と判定された商品系については、ステップＳ１０１、Ｓ１０３、Ｓ１０５の条件全てで）“ＮＯ”となった場合には、その商品系は調整対象商品系ではないものと判断する。
【０１８５】
上述したように、バランスがとれていない商品系を調整対象商品系と決定すると共に、アンバランスとなっている理由（品種不足の商品系、見本の無駄使い多い商品系、売れない商品系等）を判断して、これに応じて調整方法を決定することができ、各ロケ毎に適切な調整対象商品系及び調整方法の決定が行なえる。
【０１８６】
一方、上記ステップＳ９９の判定でＹＥＳ（ＳＲ＝０）であった商品系については、以下の処理を行う。
ここで、どのような商品系であれ、全く売れない（ＳＲ＝０）商品系というものは殆ど考えられないので、ＳＲ＝０ということは、そのロケではその商品系の商品を全く扱っていない（販売していない）ということになるが、単にそのロケでは不人気だから販売していないのか、それとも販売すれば売れる可能性が高いにも係わらず扱っていないのかによって、対応が全く異なることになる。
【０１８７】
よって、商品系間の相関関係分析部７によって得られた相関関係マトリクス （グループ内Ｃ１７、または／及び、全範囲Ｃ１９）の情報を取得して（ステップＳ１１３）、その商品系が、グループ内または全ロケ中で売筋商品系となっている商品系のどれか１つ以上との間で、（強い）正の相関関係があるか否かを判別する（ステップＳ１１４）。上述してある通り、売れている商品系（メイン商品系）と正の相関関係を持つ商品系は売れるはずであるから、正の相関関係があると判別された場合には（ステップＳ１１４、ＹＥＳ）、この商品系は売れるはず（売れる可能性が高い）であるにも係わらずそのロケでは販売していなかった（必要な商品が置かれていなかった）と判断されるので、当該商品系を追加することを調整方法とする（ステップＳ１１５）。そうではない場合には（ステップＳ１１４、ＮＯ）、単にそのロケでは不人気である等が原因であると判断し、そのままとする。
【０１８８】
更に、上述した処理により決定した各調整対象商品系及びその調整方法に対して、その調整数を決定する（ステップＳ１１６）。例えば、「同系他品を追加する」と決めた場合、その商品種数、または各商品種の取扱い数量（例えば１コラム、２コラムといったコラム単位で）を決定する。
【０１８９】
以上の処理により、ある商品系について、調整対象商品系とすると決定した場合には、最終的には全ての調整対象ロケ毎の全ての調整対象商品系及びその調整方法（商品の追加、除去、入替、見本再配分）をまとめたリストとなる対象商品系／調整方法リストに、当該商品系名及びその調整方法を追加する処理を行う （ステップＳ１１７）。
【０１９０】
上述した処理は、ある調整対象ロケについて、全ての商品系についての処理が完了するまで行ない（ステップＳ１１８，ＹＥＳ）、続いて、未処理の調整対象ロケがある場合には（ステップＳ１１９，ＮＯ）次の調整対象ロケについて同様の処理を行なっていく。
【０１９１】
そして、全ての調整対象ロケについての処理が完了したら（ステップＳ１１９、ＹＥＳ）、完成した商品系／調整方法リスト（調整対象ロケ別の調整対象商品系、その調整方法、及び調整数のリスト）を出力する（調整商品決定処理部１２に通知する）（ステップＳ１２０）。
【０１９２】
調整商品決定処理部１２は、上記ステップＳ１２０の出力と、ロケ制約分析部６、ロケ／商品売筋を分析するロケ／商品売上クロスＡＢＣ分析部８または／及びロケ／商品収益筋を分析するロケ／商品粗利クロスＡＢＣ分析部１４、及び商品売上変動時系列分析部９による処理結果に基づいて、調整商品を決定する。
【０１９３】
まず、図３６を参照して、ロケ制約分析部６による処理について説明する。
まず、上記データベースに格納されているロケ設置情報の中には、各ロケ毎 （更にそのロケに属する各自動販売機毎）にそれが特定メーカーの商品のみを専用に扱う自動販売機であるか否かを示す情報（本例では、これを一種の“制約”と捉らえ、制約有無情報と呼ぶ）が含まれてるものとする。
【０１９４】
ロケ制約分析部６は、このロケ設置情報Ａ１に含まれる制約有無情報を取得して（ステップＳ１２２）、全ロケについて処理を行うまで（ステップＳ１２５）、各ロケ毎に、それぞれ、そのロケには上記“制約”があるか否かを判定し（ステップＳ１２３）、制約あるロケの一覧情報Ｃ１１を作成する。すなわち、“制約”があるロケについては（ステップＳ１２３、ＹＥＳ）、そのロケのロケＩＤ及び制約情報（どの特定メーカーの商品専用の自動販売機であるのかを示す情報）を一覧表Ｂ１に順次追加していき（ステップＳ１２４）、全てのロケについての上記処理が完了した時点（ステップＳ１２５，ＹＥＳ）、一覧表Ｂ１が完成する。
【０１９５】
更に制約は設置場所内の全自販機か、それとも１部の自販機かをチェックするようにしてもよい。
次に、図３７を参照して、ロケ／商品売筋を分析するロケ／商品売上クロスＡＢＣ分析部８、ロケ／商品収益筋を分析するロケ／商品粗利クロスＡＢＣ分析部１４の処理手順について説明する。尚、上記各クロスＡＢＣ分析部８、１４は、基本的な処理の流れは同じであるので、いずれも図３７を参照して説明する。
【０１９６】
概略的には、いずれも、クロスＡＢＣ分析手法（公知の手法）を利用して処理を行う。ここで、簡単に説明しておくならば、クロスＡＢＣ分析手法は、既に説明してあるＡＢＣ分析手法と、クロス分析手法とを組み合わせた手法である。クロス分析手法とは、データを集計、表にあらわすとき、結果を単純集計するのではなく、要因別に結果の内訳をクロス（交差）させて、作表や分析を行う手法 （勿論、公知の手法）である。例えば、集計したデータを要因別に縦軸方向／横軸方向に並べた表（クロス集計表）を作成し、全体の傾向や問題の特質を捉らえることができる。そして、この縦軸方向／横軸方向にＡＢＣ分析（各要因別のＡ、Ｂ、Ｃランク）を適用したものが、クロスＡＢＣ分析である。
【０１９７】
本例では、上記“要因別”として、ロケと、商品とを用い、例えばロケを横軸方向、商品（商品名）を縦軸方向とし、売上（または粗利）により分析を行う。
まず、ロケ／商品売上クロスＡＢＣ分析部８について説明する。
【０１９８】
まず、処理対象となる商品の商品ＩＤ、分析範囲（全範囲、グループＧｒＬ１のみ等といった範囲）、分析期間データを取得する（ステップＳ１２７）。これは、その都度、分析者等が指定するようにしてもよい。
【０１９９】
次に、売上実績情報Ａ４を取得して、上記処理対象となる商品について、分析範囲内における各ロケ別の売上、その累計（総売上）を算出する。更に総売上を取扱ロケ数（その商品を販売しているロケの数）で割った値を算出し、これをその商品の平均売上とする（すなわち、総売上が大きくても、それは取扱ロケ数が多い為かもしれない（当然、その逆も有り得る）ので、その商品が売筋商品であるか否かを判断するには、上記平均売上を指標とする必要がある）。
【０２００】
上述した処理を全ての処理対象商品について行う。そして、平均売上の上位順で商品をソートする。更に、各ロケ毎に、そのロケで扱っている全商品の売上の合計を求め、この売上合計の上位順でロケをソートする。
【０２０１】
更に、上述してあるＡＢＣ分析手法を用いて、平均売上の上位順に累計占有率を求め、ＡＢＣランク付けを行なう。同様に、売上合計の上位順に累計占有率を求め、ＡＢＣランク付けを行なう。
【０２０２】
これより、例えば、図３８に示すようなロケ／商品マトリクス表が作成される（ステップＳ１２８）。
そして、ロケ／商品マトリクス表、ＡＢＣ分析結果（ロケ／商品売上クロスＡＢＣマトリクスＢ７）を出力／表示する（ステップＳ１２９）。
【０２０３】
これにより、指定範囲内の主力ロケ／メイン商品（売上筋から判断したもの）を明らかにし、後の調整商品決定処理部１２における調整商品の検索、追加・撤去商品の決定処理に用いられる。
【０２０４】
次に、ロケ／商品粗利クロスＡＢＣ分析部１４について説明する。
まず、処理対象となる商品の商品ＩＤ、分析範囲（全範囲、グループＧｒＬ１のみ等といった範囲）、分析期間データを取得する（ステップＳ１２７）。これは、その都度、分析者等が指定するようにしてもよい。
【０２０５】
次に、商品原価、売価情報Ａ６を取得して、上記処理対象となる商品について、分析範囲内における各ロケ別の粗利（収益・利益）、その累計（総粗利（収益・利益））を算出する。更に総粗利（収益・利益）を取扱ロケ数（その商品を販売しているロケの数）で割った値を算出し、これをその商品の平均粗利（収益・利益）とする。
【０２０６】
上述した処理を全ての処理対象商品について行う。そして、平均粗利（収益・利益）の上位順で商品をソートする。更に、各ロケ毎に、そのロケで扱っている全商品の粗利（収益・利益）の合計を求め、この粗利（収益・利益）合計の上位順でロケをソートする。
【０２０７】
更に、上述してあるＡＢＣ分析手法を用いて、平均粗利（収益・利益）の上位順に累計占有率を求め、ＡＢＣランク付けを行なう。同様に、粗利（収益・利益）合計の上位順に累計占有率を求め、ＡＢＣランク付けを行なう。
【０２０８】
これより、例えば、図３８に示すようなロケ／商品マトリクス表が作成される（ステップＳ１２８）。
そして、ロケ／商品マトリクス表、ＡＢＣ分析結果（ロケ／商品粗利ＡＢＣマトリクス）を出力／表示する（ステップＳ１２９）。これらは、具体的には図示しないが、図３８のロケ、商品ごとにＡＢＣ分析結果を表示すれば得られる。
【０２０９】
これにより、指定範囲内のロケ／商品（収益筋から判断したもの）を明らかにし、後の調整商品決定処理部１２における調整商品の検索、追加・撤去商品の決定処理に用いられる。
【０２１０】
図３９は、商品売上変動時系列分析部９における処理手順を説明する為の図である。
まず、処理対象となる商品の商品ＩＤ、分析範囲（全範囲、グループＧｒＬ１のみ等といった範囲）、分析期間データを取得する（ステップＳ１３１）。これは、分析者等が指定するようにしてもよい。
【０２１１】
これより、処理対象商品の分析範囲、分析期間中の売上集計を行ない、これに基づいて変化率（変動率）の計算処理を行う（ステップＳ１３２）。売上集計の単位は、何でもよいが、例えば図４０に示すように、“月”単位で集計して、各月の売上を、その前月の売上と比較した変化率を算出する。
【０２１２】
そして、ステップＳ１３２の処理で得た商品売上、変化率の結果（商品売上／変化率表Ｃ２３）を、例えば図４０に示すように表示する。また、必要に応じて、図４１に示すような棒グラフ（あるいは、折線グラフなど)表示する（ステッ プＳ１３３）。
【０２１３】
以上の処理を、例えば分析範囲内のロケで扱っている商品全てについて、あるいはユーザが指定する任意の（複数の）商品について、繰り返し実行する。
以上の処理により、例えば分析者が、上記表示内容を参照して、各商品が成長期、成熟期（安定期）、衰退期（特定場所に対して）等のどの段階にあるかを把握することができ、これは後の調整商品決定処理部１２の処理において活用される。すなわち、商品の追加、または入替に際して、成長期、成熟期（安定期）にある商品を追加商品または入替商品とし、衰退期の商品を追加商品または入替商品候補から除外することにより、適切な商品追加／入替が行なえるようになる。
【０２１４】
図４２は、調整商品決定処理部１２における処理手順を説明する為のフローチャート図である。
まず、上記調整商品系決定処理部１１の処理で得られた、商品系／調整方法リストＣ２２（調整対象ロケ別の調整対象商品系、その調整方法、及び調整数のリスト）のデータを取得する。また、上記ロケ／商品売上クロスＡＢＣ分析部８またはロケ／商品粗利クロスＡＢＣ分析部１４により得られた、ロケ／商品ＡＢＣマトリクス表（売筋または収益筋によるもの）を取得する（ステップＳ１３６）。
【０２１５】
そして、（図には示されていないが）例えば上記商品系／調整方法リストの上から順番に、ある調整対象ロケ、その調整対象商品系を処理対象として、ステップＳ１３７〜Ｓ１５５の処理を実行する。
【０２１６】
まず、処理対象となったある調整対象ロケのある調整対象商品系に関して、この調整対象商品系の調整方法が何であるかをステップＳ１３７、ステップＳ１５０の処理で判別する（調整方法は商品系／調整方法リストに載っている）。尚、調整方法が、商品入替である場合には、商品除去の処理と商品追加の処理を順次に行う。
【０２１７】
まず、調整方法が、商品追加であった場合（ステップＳ１３７、ＹＥＳ）の処理について説明する。
この場合、まず、ロケ制約分析部６により得られた一覧表Ｂ１のデータを取得して（ステップＳ１３８）、現在処理対象となっている調整対象ロケに“制約”があるか否かを判別する（ステップＳ１３９）。
【０２１８】
“制約”有りの場合（ある特定メーカーの商品のみを扱っている場合）には （ステップＳ１３９，ＹＥＳ）、ロケ／商品ＡＢＣマトリクス表の上位から、処理の対象となっている商品系に属する商品であって、特定メーカーの商品で且つ調整対象ロケに置いていない商品を検索し、必要な追加商品候補（例えば、調整数が‘２’であった場合は、最上位と２番目の商品）を抽出する（ステップＳ１４１）。
【０２１９】
次に、商品売上変動時系列分析部９により得られた商品売上／変化率表Ｃ２３を用いて、上記抽出した追加商品候補が、衰退期の商品（除外すべき商品）であるか否かを判定する（ステップＳ１４２）。これは、例えば「売上変化率が負、且つ、その絶対値が大」である場合は、衰退期の商品であるとし、追加商品候補から除外する。また、これと共に、追加商品候補の数が上記調整数（例えば‘２’）を満たさなくなるので、ステップＳ１４０の処理に戻り（ステップＳ１４２、ＹＥＳ）、次の追加商品候補を検索する。
【０２２０】
ステップＳ１４２の判定がＮＯの場合、「調整商品の情報」（上記追加商品候補）を作成し（ステップＳ１４３）、再度、上記調整数を満たしているか否かを確認した後（ステップＳ１４４）、上記「調整商品（追加）の情報」を、まとめてリストにし、これを調整方法提示（表示）処理部１３に通知する。また、この段階でリストを表示して、分析者等に確認させ、必要であれば修正させるようにしてもよい（ステップＳ１５５）。
【０２２１】
一方、“制約”なしの場合には（ステップＳ１３９，ＮＯ）、ロケ／商品ＡＢＣマトリクス表の上位から、処理の対象となっている商品系に属する商品であって、調整対象ロケに置いていない商品を検索し、必要な追加商品候補を抽出する（ステップＳ１４５）。
【０２２２】
次に、商品売上変動時系列分析部９により得られた商品売上／変化率表Ｃ２３を用いて、上記抽出した追加商品候補が、衰退期の商品（除外すべき商品）であるか否かを判定する（ステップＳ１４６）。これは、例えば「売上変化率が負、且つ、その絶対値が大」である場合は、衰退期の商品であるとし、追加商品候補から除外する。また、これと共に、追加商品候補の数が上記調整数を満たさなくなるので、ステップＳ１４５の処理に戻り（ステップＳ１４７、ＹＥＳ）、次の追加商品候補を検索する。
【０２２３】
ステップＳ１４７の判定がＮＯの場合、「調整商品の情報」（上記追加商品候補）を作成し（ステップＳ１４８）、再度、上記調整数を満たしているか否かを確認した後（ステップＳ１４９）、上記「調整商品（追加）の情報」を、まとめてリストにし、これを調整方法提示（表示）処理部１３に通知する。また、この段階でリストを表示して、分析者等に確認させ、必要であれば修正させるようにしてもよい（ステップＳ１５５）。
【０２２４】
一方、上記ステップＳ１３７の判定がＮＯであった場合、次に、「商品の撤去」であるか否かを判定する（ステップＳ１５０）。「商品の撤去」ではない場合には（ステップＳ１５０，ＮＯ）、「商品の見本缶数の再分割」であるものとし、商品の見本缶数の再分割の情報を、上記商品系／調整方法リストから得て（ステップＳ１５４）、分析者等が当系の個々の商品の見本数情報を参照し、再分割する商品（通常、見本数の多い商品）を決定し、これを調整方法提示（表示）処理部１３に通知する。また、この段階で表示して、分析者等に確認させ、必要であれば修正させるようにしてもよい（ステップＳ１５５）。
【０２２５】
「商品の撤去」である場合には（ステップＳ１５０，ＹＥＳ）、ロケ／商品ＡＢＣマトリクス表における対象自販機(ロケ)列に置いてある商品の下位から対象商品系の商品を検索し、除去する商品候補を見出す（ステップＳ１５１）。
【０２２６】
そして、必要な数だけ、除去する商品候補を抽出したら（ステップＳ１５２，ＹＥＳ）、調整商品（撤去）の情報を作成し（ステップＳ１５３）、これをステップＳ１５５で調整方法提示（表示）処理部１３に通知し、また、表示する。
【０２２７】
調整方法提示（表示）処理部１３は、上記のように通知された商品の追加・撤去リスト（調整対象ロケ別の、調整商品及び調整方法（商品の追加、除去、入替、見本再配分）の情報）を、調整対象自販機に対する調整方法施策案として、ディスプレイ画面、プリンタ等に出力する。
【０２２８】
尚、商品売上変動時系列分析部９により得られた商品売上／変化率表を用いて、上記のように衰退期の商品を除外するだけでなく、成長期、成熟期（安定期）にある商品を追加商品または入替商品候補とするようにもできる。
【０２２９】
この調整方法施策案の出力形式は、例えば、図４３に示すように、ロケ／商品マトリクス表における対象ロケ−対象商品に対応するフィールドに調整方法のマーク（商品の追加、除去、入替、見本再配分）を付けて提示するものである（勿論、調整ロケ、調整商品及び調整方法（商品の追加、除去、入替、見本再配分）の情報を載せる他の単独のテーブル等でもよい）。
【０２３０】
図４３には、ロケ／商品ＡＢＣマトリクス表において、対象ロケ−対象商品に対応するフィールドに、調整方法のマークとして、商品の追加の場合には1)、入替えの場合であって追加する商品には2)／削除する商品にはバツ（×）、見本減の場合には下向き矢印（↓）、削除の場合にはバツ（×）を示している。
【０２３１】
例えば、ロケＮｏ．＝ｎのロケを例にすると、「商品３８」が調整対象商品系Ｃにおける追加商品であり、「商品３４」が調整対象商品系Ｗにおける追加商品であることが分かる。また、「商品２０」と「商品５」とが入替えであって、商品５を撤去して、「商品２０」を追加することが分かる。また「商品１７」は撤去である。
【０２３２】
次に、以下、上記調整商品系／調整商品決定処理に係わる他の実施例について説明する。この他の実施例による処理は、基本的には、図１における調整商品系決定処理部１１及び調整商品決定処理部１２による処理に代えて、行なわれるものである。
【０２３３】
この他の実施例では、調整対象ロケ抽出処理部１０−１により求められた各調整対象ロケ毎に、出来るだけ商品種を増やせるようにすることを目的とし、まず、既存の商品のコラムを減らして空けることのできるコラムの最大値を求めてから、この最大値まで新規商品を追加（入替）する、という手法をとっている。また、当該他の実施例では、各商品（またはコラム）をＨＯＴ（温）とＣＯＬＤ （冷）とで区別して扱っている。つまり、自動販売機で販売される商品は、ＨＯＴ（温）で販売するものとＣＯＬＤ（冷）で販売するものがあり、１台の自動販売機で両者を混在させて販売する場合もある（商品を加熱するコラムと冷却するコラムがある）。そして、同一商品であってもＨＯＴ（温）かＣＯＬＤ（冷）かで販売状況は異なることから、両者を別物として扱うようにすることで、より現実の状況に沿った商品の入替え等が行なえるようになる。
【０２３４】
図４４は、他の実施例による調整商品決定処理全体の流れを概略的に示す図である。
本例によるロケ毎の調整商品決定処理（商品種変更内容の決定処理）（ステップＳ１６０）は、上記の通り、調整対象ロケ抽出処理部１０−１により求められた各調整対象ロケ毎に、それぞれ実行するものであり、大別して、「現在そのロケに置かれている商品（既存商品）のなかでコラム数を減らせる商品の選定処理」（ステップＳ１６１）と、「現在そのロケに置かれていなくて新規に置くべき商品の選定と割付け処理」（ステップＳ１６４）とから成る。
【０２３５】
ステップＳ１６１の処理は、「既存商品の中で商品削除せずにコラム数を減らせる商品種を、ＨＯＴ／ＣＯＬＤ別に抽出する処理（Ｃ１、Ｈ１抽出）」（ステップＳ１６２）と、「既存商品から削除する商品種を、ＨＯＴ／ＣＯＬＤ別に抽出する処理（Ｃ２、Ｈ２抽出）」（ステップＳ１６３）とから成る。
【０２３６】
ステップＳ１６４の処理は、「商品種（バラエティ）を増やすべき商品系を抽出する処理」（ステップＳ１６５）、「抽出された商品系で、増やす商品種候補を抽出する処理（Ｃ３、Ｈ３抽出）」（ステップＳ１６６）、「“増やす商品種（新しく追加したい商品）”を“減らすコラム”に割付ける処理」（ステップＳ１６７）とから成る。
【０２３７】
以下、ステップＳ１６２、Ｓ１６３、Ｓ１６５、Ｓ１６６、Ｓ１６７の各処理について、詳細に説明していく。
但し、その前に、これら各処理で用いるデータについて説明しておく。
【０２３８】
本例の処理に用いるデータは、基本的には、図１における商品構成バランス情報Ｂ６とロケ／商品売上ＡＢＣマトリクスＢ７（例えば図３８に示すロケ／商品マトリクス表）等であるが、上記の通り、ＨＯＴ／ＣＯＬＤを区別しているので、これらデータは、ＨＯＴ／ＣＯＬＤを区別して作成されているものとする。本例で用いるロケ／商品マトリクス表の一例を図４５に示す。
【０２３９】
図４５に示すロケ／商品マトリクス表３０は商品種をＨＯＴ／ＣＯＬＤで区別して扱っていること以外は、基本的には、図３８に示すロケ／商品マトリクス表と同様のフォーマットである（メーカー、原価等は、本処理では必要ないので、示していないだけである）。商品名３１は、各商品種の商品名であるが、ＨＯＴとＣＯＬＤとで別物として扱っている（例えば、“コーヒー４（Ｈ）”と“コーヒー４（Ｃ）”）。そして、各商品名３１毎に、その商品が属する商品系３２、ＨＯＴ／ＣＯＬＤの区別（Ｈ／Ｃ３３）、各ロケでの売上３４、各グループでの売上集計３５、各グループでの設置ロケでの平均３６、全体での売上集計３７、全体での設置ロケでの平均３８の各データが格納される。
【０２４０】
各グループでの設置ロケでの平均３６は、商品別に各ロケでの売上３４を集計したもの（各グループでの売上集計３５）を、その商品を現在販売しているロケの数で割ったものである。全体での設置ロケでの平均３８についても、グループ別ではなく全てを対象としている点を除いて、同様である。
【０２４１】
尚、図３８に示すロケ／商品マトリクス表における「商品別の総売上」は、上記各グループでの売上集計３５または全体での売上集計３７に相当し（どちらであってもよい）、「商品別の平均売上」は、上記各グループでの設置ロケでの平均３６または全体での設置ロケでの平均３８に相当する（どちらであってもよい）。以上、図４５は売上についてのロケ／商品マトリクス表３０を一例として示しているが、このような例に限らず、例えば粗利や粗利率についてのロケ／商品マトリクス表も同様に考えられる。
以下の説明では、各グループでの設置ロケでの平均３６または全体での設置ロケでの平均３８のことを、“設置ロケでの平均売上（または平均粗利あるいは平均粗利率”と呼ぶものとする。
【０２４２】
以下、まず、ステップＳ１６２の「既存商品の中で商品削除せずにコラム数を減らせる商品種を、ＨＯＴ／ＣＯＬＤ別に抽出する処理」について、図４６を参照して説明する。
【０２４３】
このステップＳ１６１の処理では、基本的には、売上（または粗利や粗利率）の割にはコラム数を多くとっている商品のコラム数を、減らすようにする。
【０２４４】
図４６において、まず、処理対象の調整対象ロケの既存商品毎に、補充指標効率Ｆ１（商品）、Ｆ２（商品）を算出する。Ｆ１（商品）はその商品の現在のコラム数での補充指標効率であり、Ｆ２（商品）はその商品のコラム数を１減らした場合の補充指標効率であり、算出式は以下に示す通りである。
【０２４５】
【数３】

【０２４６】
尚、Ｆ１（商品）、Ｆ２（商品）は、ある“商品”について計算された補充指標効率Ｆ１、Ｆ２の値を意味する。例えば、“コーヒー１”という商品があった場合、この“コーヒー１”について計算された補充指標効率Ｆ１、Ｆ２の値を、Ｆ１（コーヒー１）、Ｆ２（コーヒー１）という形で表す。
【０２４７】
そして、上記既存商品毎に算出した各Ｆ１（商品）中で、最大値をとるＦ１ （商品）を求め、これを補充効率基準Ｆ０とする（Ｆ０＝ max（Ｆ１（商品）））。尚、 max（Ｆ１（商品））となる商品は、売上（または粗利や粗利率）の割にはコラム数が少ない商品ということになる。
【０２４８】
（以上、ステップＳ１７１）
次に、上記既存商品毎に算出した各Ｆ２（商品）中で、最小値をとるＦ２（商品）を求め（ minＦ２（商品）と表す）、これと補充効率基準Ｆ０とを比較して、 minＦ２（商品）＜Ｆ０の条件が満たされる場合には、この minＦ２（商品）の商品のコラム数を１減らす（ステップＳ１７２）。つまり、現在のコラム数から１減らしても、なお、補充指標効率がＦ０より小さい場合には、その商品は、現在、販売数の割にはコラム数が多すぎる状況にあると考えられるので、コラム数を１減らしても構わないと判定する。
【０２４９】
そして、ステップＳ１７２で minＦ２（商品）＜Ｆ０の条件が満たされた場合には、コラム数を減らした商品について、Ｆ２（商品）を算出し直して（ステップＳ１７３）（つまり、現在の状況（まだコラム数を減らしていない）を基準に考えると、実質、「Ｆ２（商品）＝（その商品の販売数）÷（その商品のコラム数−２）」を算出することになる）、再び、ステップＳ１７２の処理に戻る（ステップＳ１７４，ＮＯ）。尚、再び行なったステップＳ１７２の処理において、上記コラム数を減らした商品のＦ２（商品）が、再び minＦ２（商品）となる場合も有り得る。つまり、ある商品について、２コラム以上減らすことになる可能性もある。
【０２５０】
一方、ステップＳ１７２でＦ２（商品）＜Ｆ０の条件が満たされなくなったら、コラム数を減らせる商品は無くなったと判定し（ステップＳ１７４，ＹＥＳ）、ステップＳ１７５の処理に進み、減らすコラム数を、ＨＯＴ／ＣＯＬＤ別に集計して、当該処理を終了する。尚、上記処理により得られた“商品を削除せずに減らせるコラム”は、図ではＨＯＴについては“Ｈ１”、ＣＯＬＤについては “Ｃ１”で示している。
【０２５１】
次に、ステップＳ１６３の「既存商品から削除する商品種を、ＨＯＴ／ＣＯＬＤ別に抽出する処理」について、図４７を参照して説明する。
ここでは、上記ステップＳ１６２の処理により減らされたコラム数に基づいて、以下のステップＳ１８１の処理により、ＨＯＴ／ＣＯＬＤ別に、削除すべき商品の候補を求める。つまり、上記ステップＳ１６２の処理によりコラム数を減らされた結果、１コラムとなった商品についても、以下のステップＳ１８１における“コラム数が１コラムである商品”として扱うようにする。これにより、例えば元々３コラムあった商品が、１コラムに減らされ、更に商品が削除されるような事も起こり得る。
【０２５２】
ステップＳ１８１の処理では、各ロケ毎に、売上（または粗利や粗利率）が少なく（例えば、下位数％〜数十％程度）且つコラム数が１コラムである商品の中で、その商品を削除してもその商品が属する商品系の商品は無くならない（他に同じ商品系の商品がある）ものを、削除すべき商品の候補とする（これらは、図ではＨＯＴは“Ｈ２”、ＣＯＬＤは“Ｃ２”で示している）。このように「商品を削除せずに減らせるコラム」Ｃ１、Ｈ１を計算した後に、「商品削除によって減らすコラム」Ｃ２、Ｈ２を計算するようにしたのは、まず商品削除はできるだけ行なわずに空きコラムＣ１とＨ１を作り、その上で更に空きコラムが必要であれば商品削除を行なって空きコラムＣ２とＨ２を使うためである。こうすることにより、商品削除をできるだけ行なわずに新しい商品を設置し、商品種（バラエティ）を増やす操作を行う上で便利である。
【０２５３】
次に、以下、ステップＳ１６５の「商品種（バラエティ）を増やすべき商品系を抽出する処理」、ステップＳ１６６の「抽出された商品系で、増やす商品種候補を抽出する処理」について、図４８（ａ）、（ｂ）を参照して説明する。この図４８（ａ）及び（ｂ）の処理は、基本的には、そのロケに置けば売れる（または利益が上がる）と期待できる商品を抽出するものである。
【０２５４】
図４８（ａ）は、ステップＳ１６５の「商品種（バラエティ）を増やすべき商品系を抽出する処理」の詳細フローチャート図である。
同図において、各ロケ毎に、全ての商品系についてチェックが終了するまで （ステップＳ１９４）、以下の処理を繰り返し行う。
【０２５５】
まず、その商品系が、Ｇｒ平均が上位３位以内の商品系であるか否かを判定し（ステップＳ１９１）、そうである場合には（ステップＳ１９１，ＹＥＳ）、続いて、その商品系のそのロケでの売上占有率ＳＲと、Ｇｒ平均とを比較して、 「ＳＲ＜Ｇｒ平均」の条件を満たす場合には（ステップＳ１９２，ＹＥＳ）、その商品系をリストに登録する（ステップＳ１９３）。つまり、Ｇｒ平均が上位３位以内の商品系であっても、そのロケでの売上占有率がＧｒ平均以上（ＳＲ≧Ｇｒ平均」である場合には、これ以上商品種を増やしても売上が伸びる可能性が高いとは言えないので、リストに登録しない。尚、ステップＳ１９２の判定方法は、一例であり、これに限るものではなく、別の方法であってもよい。または「Ｇｒ平均の上位３商品系」を選ぶ代わりに「Ｇｒ平均の上位の数％〜数十％を占める商品系」を用いてもよい。これはＧｒ平均の多いほうから積算していって全体の数％〜数十％を占める商品系ということで、商品系についてのＡＢＣ分析をしたときのＡランクの商品系である。
【０２５６】
尚、例えば、図２８の例では、商品系Ｃ、Ｎ、Ｗが、Ｇｒ平均が上位３位内の商品系である。但し、以下の説明では、特に具体的な商品系名は示さずに、Ｇｒ平均が上位３位内の商品系は、仮に１位が商品系Ｘ、２位が商品系Ｙ、３位が商品系Ｚであったものとして説明する（上記処理により、図４８に示すリスト４０における商品系Ｘ、Ｙ、Ｚが登録されたものとする）。
【０２５７】
また、尚、リスト４０に示す例は、上位３位内の商品系全てが上記ステップＳ１９２の「ＳＲ＜Ｇｒ平均」の条件を満たした場合を示すが、例えば商品系Ｙは条件を満たさなかった場合には、商品系Ｘ、Ｚのみが登録される。
【０２５８】
図４８（ｂ）は、ステップＳ１６６の「抽出された商品系で、増やす商品種候補を抽出する処理」の詳細フローチャート図である。
図４８（ｂ）に示す処理は各ロケ毎に行うものであり、各ロケ毎に、上記リストに登録された各商品系毎に、例えばロケ／商品マトリクス表３０を用いて、以下に説明する処理を行う。
【０２５９】
まず、上記リストに登録された中のある商品系（以下、処理対象商品系という）について、そのロケ（以下、処理対象ロケという）に置いていない商品があるか否かを判定し（ステップＳ１９５）、ある場合には（ステップＳ１９５，ＹＥＳ）、その商品系の商品の中で、“設置ロケでの平均売上（または平均粗利や平均粗利率）”が高い商品を、ＨＯＴ／ＣＯＬＤで各々選択する（ステップＳ１９６）。そして、選択した商品種を、リスト４０に登録する（ステップＳ１９７）。
【０２６０】
以上の処理は、例えば、ロケ／商品マトリクス表３０に示す例において、仮に、処理対象ロケが No.１（ロケＩＤ＝1015）のロケであり、処理対象商品系が商品系Ｃであったものとすると、ロケ／商品マトリクス表３０では空欄で示す商品が、処理対象ロケに置いていない商品ということである。この例では、商品名３１が“コーヒー４（Ｃ）”、“コーヒー６（Ｃ）”、“コーヒー２（Ｈ）”等が空欄であるので、ステップＳ１９５の判定はＹＥＳとなる。続いて、これら商品の中で“設置ロケでの平均売上（または平均粗利や平均粗利率）”が高い商品を選択するが、これはロケ／商品マトリクス表３０は既に“設置ロケでの平均売上（または平均粗利や平均粗利率）”が高い順にソートされているので、上から順に検索すると、ＣＯＬＤについては“コーヒー４（Ｃ）”、ＨＯＴについては “コーヒー２（Ｈ）”が選択され、リスト４０に登録される。
【０２６１】
尚、更に、「“設置ロケでの平均売上（または平均粗利や平均粗利率）”が上位数％〜数１０％以内（例えば図３８におけるランクＡ）」であることを条件に加えてもよい。この場合、例えば、ＣＯＬＤについては“コーヒー４（Ｃ）”が選択されるが、ＨＯＴについては何も選択されない場合も起こり得る。このような場合には、リスト４０において対応する箇所は、空欄とする。
【０２６２】
以上の処理を、リスト４０に登録した全ての商品系について行なったら（ステップＳ１９８，ＹＥＳ）、当該処理は終了となり、“コラムを増やすべき商品のリスト”（すなわちリスト４０）が完成することになる（ステップＳ１９９）。
【０２６３】
図示の例では、仮に、商品系Ｘに関して選択／登録された商品が、ＣＯＬＤが“ＸＣ”、ＨＯＴが“ＸＨ”であったものとする。同様に、商品系Ｙ、Ｚに関しても、各々、“ＹＣ”、“ＹＨ”、“ＺＣ”、“ＺＨ”が選択／登録されたものとする。尚、リスト４０は、図４４等ではＨＯＴは“Ｈ３”、ＣＯＬＤは“Ｃ３”で示している。
【０２６４】
そして、最後に、このようにしてリスト４０に登録された商品（Ｃ３、Ｈ３）を、Ｃ１、Ｈ１、Ｃ２、Ｈ２に割付ける処理を行う（上記ステップＳ１６７の処理）。但し、（Ｃ１、Ｈ１、Ｃ２、Ｈ２）は、そのロケで空けることができるコラム数の最大値を意味するのであり、この最大値までは自由に新たな商品を追加できるようにしているのであり、必ず最大値まで新たな商品を追加しなければならないわけではなく、例えば以下にも述べるようにＣ１、Ｈ１の空きコラムだけを使用して商品を追加してもよい（結果的に新たな商品を追加しなかったコラムは、元のままとする）。
【０２６５】
この処理に関しては、２つの方法を提案する。
まず、図４９を参照して、第１の方法について説明する。
図４９（ａ）は、リスト４０に登録された商品、すなわち新しく追加したい商品を、Ｃ１／Ｈ１（コラム減による空きコラム）、Ｃ２／Ｈ２（商品削減による空きコラム）に割付ける様子を、概略的に示す図である。
【０２６６】
図４９（ｂ）は、リスト４０に登録された商品を、ＨＯＴ／ＣＯＬＤ別に、 “設置ロケでの平均売上”が高い順にソートし直してから、Ｃ１／Ｈ１、Ｃ２／Ｈ２に割付ける様子を、概略的に示す図である。つまり、リスト４０では、商品系として平均売上占有率が高い順（Ｘ、Ｙ、Ｚの順）に並べてあるが、各商品毎に見た場合、必ずしも商品系Ｘの商品（ＸＣ、ＸＨ）が最も“設置ロケでの平均売上”が高いとは限らないからである（図では、結局、商品系の順番通りとなっているが、これに限らないということである）。
【０２６７】
図４９（ａ）、（ｂ）は、どちらも、処理内容は同じである。すなわち、Ｃ１、Ｃ２合わせて、空きコラム数が３つ以上であれば、ＸＣ、ＹＣ、ＺＣは全て、各コラムに割付ける（どのコラムに割付けるかは自由である）。同様に、Ｈ１、Ｈ２合わせて、空きコラム数が３つ以上であれば、ＸＨ、ＹＨ、ＺＨは全て、各コラムに割付ける。尚、リスト４０において、上記の通り、必ずしも商品系Ｘ、Ｙ、Ｚ全てに対して商品種が登録されるとは限らないので、例えば２つの商品種しか登録されなかった場合には、上記「空きコラム数が３つ以上であれば」は、「空きコラム数が２つ以上であれば」に変わる。
【０２６８】
一方、空きコラム数が３つ以上ない場合には、上記ソートした順番に従って、最上位の商品種から順番に、空きコラムに割付ける。割付けられなかった商品種は、追加されないことになる。
【０２６９】
ここで、Ｃ１、Ｈ１への割付けは、既存の商品のコラム数が減るだけであり、既存の商品を削除するわけではないが、Ｃ２、Ｈ２への割付けは、既存の商品を削除するのであるから、新しく追加する商品の“設置ロケでの平均売上”が、削除する商品より少ないようでは意味がない（わざわざ、売上が少ない商品を入れることになってしまう）。よって、このような場合には、Ｃ２、Ｈ２への割付けは行なわないようにしてもよい（その商品種は、追加されないことになる）。
【０２７０】
あるいは、そもそも、当該他の実施例の目的は、商品種のバラエティを増やすことであり、Ｃ２、Ｈ２への割付けは、商品種を変更するだけであり、商品種が増えるわけではないので、Ｃ２、Ｈ２の抽出は（当然、割付けも）、行なわないようにしてもよい（Ｃ１、Ｈ１への割付けのみ行う）。
【０２７１】
次に、以下、図５０を参照して、第２の方法について説明する。
まず、第１の方法では、図４９（ｂ）に示すように、ＨＯＴ／ＣＯＬＤ別に “設置ロケでの平均売上”が高い順にソートしていたが、第２の方法では、図５０（ａ）に示すように、ＨＯＴ／ＣＯＬＤを区別することなく、“設置ロケでの平均売上”が高い順にソートした商品リスト５０を作成する。図５０（ａ）の例では、“設置ロケでの平均売上”が高い順に、ＸＣ、ＹＣ、ＸＨ、ＹＨ、ＺＨ、ＺＣの順番にソーティングされたものとする。
【０２７２】
この商品リスト５０を用いて、図５０（ｂ）に示す処理により、空きコラムへの割付けを行う。
図５０（ｂ）に示す処理は、順位が高い商品から順次処理対象にして行う。まず、処理対象商品がＣ１またはＨ１に割付け可能か否か、更に割付けできない場合にはＣ２またはＨ２に割付け可能か否かを判定する（ステップＳ２０１）。最初は、最も順位が高い商品が対象になるので、この例では、商品ＸＣが、Ｃ１に割付け可能か否か判定し、Ｃ１に割付けできない場合には更にＣ２に割付け可能か否かを判定する。Ｃ１、Ｃ２の何れにも割付け可能な空きコラムがない場合には（ステップＳ２０１，ＮＯ）、次に順位が高い商品を対象としてステップＳ２０１の処理を行う。
【０２７３】
Ｃ１／Ｃ２どちらかに割付け可能である場合には（ステップＳ２０１，ＹＥＳ）、この割付け可能と判断された空きコラムに、当該処理対象商品を割付け（ステップＳ２０２）、更にこの処理対象商品と同じ商品系で冷／温が反対の商品を、商品リスト５０から削除する（ステップＳ２０３）。例えば、商品ＸＣがＣ１／Ｃ２どちらかに割付け可能であった場合、商品ＸＨを商品リスト５０から削除する。これより、商品ＸＨは、順位は３番目であるが、当該処理対象から外れるので、そのロケに追加されることはなくなる。以上の処理を、商品リスト５０の全ての商品について行なったら（ステップＳ２０４，ＹＥＳ）（但し、削除された商品については行なわない）、処理を終了する。
【０２７４】
このようにすることで、例えばＣ１、Ｃ２、Ｈ１、Ｈ２全て合わせても空きコラムが３つしかなかった場合でも、商品系Ｘ、Ｙ、Ｚ全てに１つずつ新しい商品を追加することが可能となる。
【０２７５】
尚、第２の方法においても、上記のように、Ｃ２／Ｈ２への割付けが可能な場合であっても、新しく追加する商品の“設置ロケでの平均売上”が、削除する商品より少ない場合には、その商品は追加しないようにしてもよい。同様に、そもそもＣ２、Ｈ２の抽出自体を行なわないようにしてもよい。
【０２７６】
尚、ＨＯＴ、ＣＯＬＤで区別して扱うようにすることは、当該他の実施例に限らず、上述した図１〜図４３の実施例においても行うようにしてもよい。
図５１は、上記自販機の商品構成調整支援システムを実現する情報処理装置のハードウェア構成の一例、及び上述した各種処理／機能を実現させるプログラムを記憶した記憶媒体の一例を示す図である。
【０２７７】
図５１において、情報処理装置２０は、ＣＰＵ２１、記憶部２２（可搬記憶媒体２２ａを含む）、メモリ２３、表示部２４、操作部２５、入出力インタフェース部２６、通信部２７等より構成される。
【０２７８】
ＣＰＵ２１は、情報処理装置２０全体を制御する中央処理装置である。
記憶部２２は、少なくとも、上述した自販機の商品構成調整支援システムの各種処理機能を実現するためのプログラムが記憶されているＨＤＤ等の記憶装置である。または記憶部２２は可搬記憶媒体２２ａとその駆動読み取り装置の組合せ（例えば、フロッピディスク（ＦＤ）とフレキシブルディスクドライブ（ＦＤＤ））であってもよい。可搬記憶媒体２２ａには上記ＦＤの他にＣＤ−ＲＯＭ、メモリカード、ＤＶＤ、ＭＯ等がある。
【０２７９】
メモリ２３は、記憶部２２に格納されているプログラムを一時的に記憶し当該プログラムをＣＰＵ２１に実行させる為のＲＡＭ等である。
表示部２４は、ディスプレイ等であり、例えば上述した自販機の商品構成調整支援システムの各種処理機能により得られたデータ（表等）等を表示する。
【０２８０】
操作部２５は、キーボード、マウス等であり、これを用いてユーザ等は表示部２４に表示させる内容を選択／指定したり、あるいは表示される内容の一部分を変更する等の所望の操作／入力を行う。
【０２８１】
入出力インタフェース部２６は、例えば何等かの通信線を介してデータベース等と接続して、上述した各種情報を入力する為のインタフェースである。
通信部２７は、例えば、公衆回線網、専用の通信網等のネットワーク等を介して、外部の他の情報処理装置と通信する機能を提供する構成であり、一般的に良く知られているものである。
【０２８２】
ここで、本発明は、自販機の商品構成調整支援システムといった装置それ自体に限らず、コンピュータにより使用されたときに、本発明に係わる各種機能を実現させる為のプログラムが格納されたコンピュータ読出し可能な記録媒体（記憶媒体）自体として構成することもできる。
【０２８３】
この場合、「記録媒体」には、例えば図５１で可搬記憶媒体２２ａの一例として示しているＣＤ−ＲＯＭ２８、フロッピィーディスク２９等の可搬記憶媒体や（勿論、これら一例に限らず、ＭＯ、ＤＶＤ、リムーバブルハードディスク等、「可搬記憶媒体」の範疇に入るものであれば何であってもよい）だけでなく、ネットワークを介して通信可能な外部の任意の情報処理装置（不図示）の「記録媒体」も含まれる。また、当然、上記情報処理装置２０内の記憶装置（ＲＡＭ／ＲＯＭ又はハードディスク等）も含まれる。
【０２８４】
【発明の効果】
以上、詳細に説明したように、本発明の自販機の商品構成調整支援システムによれば、より適切な調整対象自販機の選定、より適切な商品構成の調整が行なえるようになり、以て自販機で販売する商品の総売上(又は総利益)を上げることができる。
【図面の簡単な説明】
【図１】自販機の商品構成調整支援システム全体の概略的な機能構成と処理の流れを示す図である。
【図２】ロケ属性特徴分析部における処理手順を説明する為のフローチャート図である。
【図３】属性からの分類（評価項目；カテゴリ名）の一例を示す図である。
【図４】評価結果の一例（５段階評価）を示す図である。
【図５】因子分析により得られた因子負荷量の一例を示す図である。
【図６】図５を棒グラフ化した因子負荷量棒グラフである。
【図７】因子スコアの一例である。
【図８】（ａ）〜（ｃ）は、図７の因子スコアを棒グラフ化したものである。
【図９】売上特徴分析部における処理手順を説明する為のフローチャート図である。
【図１０】図１の商品分類情報の一例である。
【図１１】ロケ別の商品系の売上占有率一覧表の一例である。
【図１２】因子分析により得られた因子負荷量の一例を示す図である。
【図１３】図１２を棒グラフ化した因子負荷量棒グラフである。
【図１４】因子スコアの一例である。
【図１５】（ａ）〜（ｃ）は、図１４の因子スコアを棒グラフ化したものである。
【図１６】属性特徴からのロケ分類処理部／売上特徴でのロケ分類処理部における処理手順を説明する為のフローチャート図である。
【図１７】属性特徴からのロケ分類処理部による分類結果の一例を示す図である。
【図１８】売上特徴でのロケ分類処理部による分類結果の一例を示す図である。
【図１９】特異ロケ分析部の処理手順を説明するためのフローチャートである。
【図２０】図１９のステップＳ２９のグループ対応付処理の詳細を説明する為のフローチャート図である。
【図２１】ロケ競合分析部による処理手順について説明する為のフローチャート図である。
【図２２】商品構成バランス分析部の処理手順を説明する為のフローチャート図である。
【図２３】図２２のステップＳ５６のバランスビジュアル化処理の詳細フローチャート図である。
【図２４】バランスビジュアル化の一例を示す図である。
【図２５】ロケのコスト／利益バランス分析部の処理手順を説明する為のフローチャート図である。
【図２６】商品系のコスト／利益バランス分析部の処理手順を説明する為のフローチャート図である。
【図２７】調整対象ロケ抽出処理部における処理手順を説明する為のフローチャート図である。
【図２８】グループの商品系平均売上占有率（売上パターン）の一例を棒グラフ化した一例を示す図である。
【図２９】ロケ／商品系売上占有率リスト（各グループの主力ロケ／メイン商品系表）の一例を示す図である。
【図３０】商品系間の相関関係分析部における処理手順を説明する為のフローチャート図である。
【図３１】（ａ）、（ｂ）は商品系相関分析の説明図である。
【図３２】相関関係マトリクスの一例を示す図である。
【図３３】調整商品系決定処理部における処理手順を説明する為のフローチャート図（その１）である。
【図３４】調整商品系決定処理部における処理手順を説明する為のフローチャート図（その２）である。
【図３５】調整商品系決定処理部における処理手順を説明する為のフローチャート図（その３）である。
【図３６】ロケ制約分析部による処理手順を説明する為のフローチャート図である。
【図３７】商品売筋を分析するロケ／商品売上クロスＡＢＣ分析部、商品収益筋を分析するロケ／商品粗利クロスＡＢＣ分析部の処理手順を説明する為のフローチャート図である。
【図３８】ロケ／商品マトリクス表の一例を示す図である。
【図３９】商品売上変動時系列分析部における処理手順を説明する為の図である。
【図４０】商品売上変動時系列分析の設定入力／表示画面の一例である。
【図４１】商品売上／変動を、棒グラフにより表示する一例を示す図である。
【図４２】調整商品決定処理部における処理手順を説明する為のフローチャート図である。
【図４３】調整方法施策案の出力形式の一例（ロケ／商品マトリクス表に調整方法のマークを付けたもの）を示す図である。
【図４４】他の実施例による調整商品決定処理全体の流れを概略的に示す図である。
【図４５】他の実施例において用いるロケ／商品マトリクス表の一例を示す図である。
【図４６】図４４のステップＳ１６２の詳細フローチャート図である。
【図４７】図４４のステップＳ１６３の詳細フローチャート図である。
【図４８】（ａ）は図４４のステップＳ１６５、（ｂ）はステップＳ１６６の詳細フローチャート図である。
【図４９】（ａ）、（ｂ）は、新しく追加したい商品を空きコラムに割付ける処理についての第１の方法を説明する為の図である。
【図５０】（ａ）、（ｂ）は、新しく追加したい商品を空きコラムに割付ける処理についての第２の方法を説明する為の図である。
【図５１】自販機の商品構成調整支援システムを実現する情報処理装置のハードウェア構成の一例、及び記憶（記録）媒体の一例を示す図である。
【符号の説明】
１ 自販機設置場所(ロケ)の属性特徴分析部
２ ロケでの商品／商品系の売上特徴分析部
３−１ 属性特徴からのロケ分類処理部
３−２ 売上特徴でのロケ分類処理部
４ 自販機内の商品(商品系)構成バランス分析部
５ ロケ競合分析部
６ ロケ制約分析部
７ 商品系間の相関関係分析部
８ ロケ／商品売上クロスＡＢＣ分析部
９ 商品売上変動時系列分析部
１０−１ 調整対象ロケ抽出処理部
１０−２ 特異ロケ分析部
１１ 調整商品系決定処理部
１２ 調整商品決定処理部
１３ 調整方法提示（表示）処理部
１４ ロケ／商品粗利クロスＡＢＣ分析部
１５ ロケのコスト／利益バランス分析部
１６ 商品系のコスト／利益バランス分析部
２０ 情報処理装置
２１ ＣＰＵ
２２ 記憶部
２２ａ 可搬記憶媒体
２３ メモリ
２４ 表示部
２５ 操作部
２６ 入出力インタフェース部
２７ 通信部
２８ ＣＤ−ＲＯＭ
２９ フロッピィーディスク
３０ ロケ／商品マトリクス表
３１ 商品名
３２ 商品系３２
３３ Ｈ／Ｃ
３４ 各ロケでの売上３４
３５ 各グループでの売上集計
３６ 各グループでの設置ロケでの平均
３７ 全体での売上集計
３８ 全体での設置ロケでの平均
４０ リスト
５０ 商品リスト[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a system for selecting a vending machine that requires adjustment, and for determining which products should be added to or removed from which vending machine.
[0002]
[Background Art and Problems to be Solved by the Invention]
Currently, various products including beverages, foods, clothing and the like are sold on vending machines.
Vending machines are unattended and non-interactive sales forms, unlike sales forms at regular stores, so sellers (vending machine companies) can dynamically grasp the configuration and usage patterns of users (consumers). However, the product cannot be sold. A mechanism in which the vending machine only presents product information (samples, etc.) unilaterally to the user, and the seller obtains information from the user and reflects it in the product configuration of the vending machine. Not necessarily established.
[0003]
Also, unlike stores, vending machines do not have warehouses to store inventory, and the types and quantities of products that can be held are extremely limited, so sales efficiency is as good as possible (to increase total sales (or gross profit)). A possible product structure is required.
[0004]
In response to such a problem, the applicant of the present invention has already proposed the invention of Japanese Patent No. 2000-287436.
However, there is always a need for a method / system capable of selecting a more appropriate adjustment target vending machine and adjusting a more appropriate product configuration.
[0005]
The problem of the present invention is to analyze the vending machine sales product, select the vending machine that needs adjustment, clearly indicate which product system should be adjusted and how, and specifically which vending machine to which Providing a vending machine product configuration adjustment support system that can appropriately determine how products should be added / removed, and thus increase the total sales (or gross profit) of products sold by vending machines. It is.
[0006]
[Means for Solving the Problems]
The product configuration adjustment support system of the vending machine according to the present invention performs attribute analysis by factor analysis according to the attribute of each location and obtains a factor score, and performs cluster analysis using the factor score obtained by the attribute feature analysis unit, For each group classified by the location classification processing means from the attribute feature for classifying the location into groups having common features in the attribute, and for each group classified by the location classification processing means from the attribute feature, the sales pattern of the location belonging to the group An adjustment target location extraction processing means for obtaining the location to be adjusted for each group based on the sales pattern is obtained.
[0007]
Alternatively, the vending machine product composition adjustment support system according to the present invention includes a sales feature analysis unit that performs factor analysis based on the sales feature of each location to obtain a factor score, and a cluster analysis using the factor score obtained by the sales feature analysis unit. The location classification processing means for sales features for classifying locations into groups having common characteristics in sales features, and the location belonging to the group for each group classified by the location classification processing means for sales features And an adjustment target location extraction processing means for obtaining a location to be adjusted for each group based on the sales pattern.
[0008]
In order to make an appropriate product composition adjustment, the location to be adjusted must first be determined appropriately.
According to the above-mentioned vending machine product composition adjustment support system, factories and cluster analysis can be used to group locations that have common characteristics into the same group based on the attributes or sales characteristics of each location (vending machine installation location) obtained through surveys, etc. Enter and group locations. An appropriate determination can be made by determining the location to be adjusted for each group.
[0009]
Alternatively, the product composition adjustment support system of the vending machine according to the present invention performs attribute analysis on the factor of each location to obtain a factor score by performing factor analysis and cluster analysis using the factor score obtained by the attribute feature analysis unit. A location classification processing means from attribute features for classifying locations into groups having common features in attributes, a sales feature analysis means for performing factor analysis on the sales features of each location to obtain a factor score, and the sales feature analysis means Cluster analysis is performed using the factor scores obtained in step 1, and the location classification processing means for sales features that classify locations into groups having common features in sales features, and the location classification processing means from the attribute features are classified. Between the group and the group classified by the location classification processing means in the sales feature. For each group classified by the specific location analysis means that uses the location belonging to only one of the groups as the specific location and the location other than the specific location as the non-specific location, and the location classification processing means based on the attribute feature Alternatively, for each group classified by the location classification processing means in the sales feature, a sales pattern of a location that is a non-specific location among the locations belonging to the group is obtained, and non-specificity is determined for each group based on the sales pattern. It is configured to include adjustment target location extraction processing means for obtaining a location to be adjusted from the specific location or / and the specific location.
[0010]
In the above-mentioned vending machine product configuration adjustment support system, for each group into which the location is grouped, a heterogeneous location (single location) that does not reflect the characteristics of the group is extracted, and the location to be adjusted is determined from this unique location . Alternatively, the adjustment target location may be determined from the non-specific location according to the sales pattern (for example, in descending order of sales). This makes it possible to select an appropriate adjustment target location.
[0011]
Further, as the adjustment target location, in addition to the one determined as described above, the following description may be added.
For example, for each location, the balance between the product / product sales occupancy rate / sample occupancy rate / variety occupancy rate handled in the location is analyzed, and the unbalanced location is determined for the location to be adjusted. The product / product system composition balance analysis means as an additional candidate is further provided, and the adjustment target location extraction processing means, if necessary, from among the additional candidates obtained by the product / product system composition balance analysis means, Select / determine a location to be added as a location to be adjusted.
[0012]
In addition, for example, the gross profit rate for each location and the total gross profit rate of all locations are calculated, the unbalanced location is obtained based on the ratio of each gross rate and the gross gross rate, and the unbalanced location is adjusted. A location cost / benefit balance analyzing unit as an additional candidate for the target location is further provided, and the adjustment target location extraction processing unit includes, among the additional candidates obtained by the location / cost balance analyzing unit as needed. Then, the location to be added as the location to be adjusted is selected / determined.
[0013]
Alternatively, for example, the gross profit rate for each location and the gross profit rate for each product system for each location are calculated, and the unbalanced product is calculated based on the ratio of the gross profit rate to the product system for each location. A product-system cost / benefit balance analysis unit that uses a location having a system as an additional candidate for the adjustment-target location, and the adjustment-target location extraction processing unit analyzes the cost / benefit balance of the product system as necessary. The location to be added as the location to be adjusted is selected / determined from the addition candidates obtained by the means.
[0014]
Further, when there is an inappropriate adjustment target location determined as described above, it may be excluded.
For example, it further includes a location competition analysis means for obtaining the presence / absence of each location and the degree of competition, and the adjustment target location extraction processing means has a high degree of competition obtained by the location competition analysis means as necessary. The location is excluded from the location to be adjusted.
[0015]
As described above, it is possible to select a more appropriate location for adjustment.
In addition, the method for determining the location to be adjusted is not limited to a method for obtaining a plurality of locations as described above by grouping, but for example, as described below, each location may be analyzed and obtained individually. Good.
[0016]
The vending machine product composition adjustment support system according to the present invention analyzes the balance among the sales occupancy rate / sample occupancy rate / variety occupancy rate of the product / product system handled in each location for each location. Product / commodity system composition balance analysis means for extracting a location that has not been adjusted, and adjustment target location extraction processing means that uses the unbalanced location extracted by the product / product system composition balance analysis means as a location to be adjusted It is comprised so that it may have.
[0017]
The vending machine product composition adjustment support system according to the present invention calculates a gross margin rate for each location and a total gross margin rate for all locations, and obtains an unbalanced location based on a ratio between each gross margin rate and the total gross margin rate. Cost / benefit balance analysis means, and adjustment target location extraction processing means that uses the unbalanced location obtained by the location cost / profit balance analysis means as the location to be adjusted.
[0018]
The vending machine product composition adjustment support system according to the present invention calculates a gross margin rate for each location and a gross margin rate for each product type for each location, and calculates the gross margin for each location and the gross margin rate for each product type. The product cost / benefit balance analysis means for obtaining a location having an unbalanced product system based on the ratio, and the location having the unbalanced product system obtained by the cost / profit balance analysis means for the product system are to be adjusted. It is comprised so that it may have the adjustment target location extraction process means used as the location of this.
[0019]
Then, for each adjustment target location, it is determined which product is to be specifically adjusted and how (adjustment methods such as addition, removal, replacement, sample number division, etc.). And how to adjust them. That is,
The vending machine product composition adjustment support system according to the present invention analyzes, for each location to be adjusted, the balance between the sales occupancy rate / sample occupancy rate / variety occupancy rate of the product system handled in the location, and is set in advance. An adjustment that determines an adjustment method for each adjustment target product system according to an adjustment method that is determined in advance for each unbalance condition, while the product system corresponding to any of the unbalanced conditions is an adjustment target product system A product system determination processing unit is provided.
[0020]
Alternatively, the vending machine product configuration adjustment support system according to the present invention calculates the gross margin of the location and the gross margin of each product system for each location to be adjusted, and the gross margin of the location and the product system An unbalanced product system is obtained based on a ratio to another gross profit rate, and an adjusted product system determination processing means for determining an adjustment method for the unbalanced product system is provided.
[0021]
Furthermore, if there is a product system that is not currently handled at the location, the balance analysis cannot be performed on this product system, but if it is a product system that is highly likely to sell, it should be handled at that location.
[0022]
Therefore, for example, for each group in which each location is classified according to attributes or sales characteristics, a correlation analysis unit that analyzes a correlation between each product system in the group is further provided, and the adjusted product system determination processing unit If there is a product system that is not handled in the location to be adjusted, based on the analysis result by the correlation analysis means, the product system that has a positive correlation with the product system that is not handled is sold. It is determined whether or not there is a line product system, and if there is a market product system, the product system that is not handled is handled at the location to be adjusted.
[0023]
A product system that has a positive correlation with a selling product system is likely to sell, so if you handle it at this location, you can increase the total sales (or gross profit) of the product sold at that location .
[0024]
As described above, when the product system to be adjusted and the adjustment method are determined for each location, the product configuration can be adjusted appropriately for each location by determining the adjusted product based on this. It becomes like this. That is,
For example, when the selling product analysis means for finding a selling product based on the sales performance information for each location / product and the adjustment method of the adjustment target product system are addition or replacement, the same as the adjustment target product system An adjustment product determination processing means for determining a product to be added or replaced from among the merchandise selling products is further provided.
[0025]
In addition, for example, when the profit source product analysis means for obtaining the profit source product based on the product cost / selling price information and the adjustment method of the adjustment target product system are addition or replacement, the same product system as the adjustment target product system And adjusting product determination processing means for determining a product to be added or replaced from among the profit-generating products.
[0026]
Alternatively, for example, when the location vending machine to be adjusted is a vending machine that handles only a product of a specific manufacturer, the adjusted product determination processing means selects the product to be added or replaced from the products of the specific manufacturer. decide.
[0027]
Further, for example, the adjusted product determination processing means may not make a product in a decline period as the product to be added or replaced.
In addition, the vending machine product configuration adjustment support system according to the present invention includes an adjustment target location extraction processing means for obtaining a location to be adjusted, and a column that can be opened by reducing the number of columns of existing products for each location to be adjusted. Empty column extraction means to be calculated based on the replenishment efficiency of each existing product, additional product determination means to determine a new product to be added, and empty columns determined by the empty column extraction means to be determined by the additional product determination means And an allocating means for allocating the added additional product.
[0028]
In the product configuration adjustment support system configured as described above, for each location, first, the maximum number of columns that can be vacated is obtained, and the type / number of additional products can be freely determined up to this maximum value. . Further, by determining the empty columns based on the replenishment efficiency, for example, the number of columns of products whose sales are not large for a large number of columns can be reduced, and new products can be placed there.
[0029]
For example, the empty column extracting means may further determine a column that can be freed by deleting an existing product.
In addition, for example, the additional product determination unit may select a product to be added from a product system having an average sales or gross profit or gross profit rate higher than the sales or gross profit or gross profit rate at the adjustment target location. decide.
[0030]
This makes it possible to determine a product that can be expected to sell if it is newly placed in the target location.
In addition, the empty column extraction unit / additional product determination unit handles the column / product by distinguishing between hot and cold.
[0031]
Actually, even if the products are the same, the trends in sales, gross profit, and the like are different between hot and cold, so that both can be handled as different products to obtain more realistic processing results.
[0032]
Further, for example, the allocating unit may perform the allocation so that the same product is not added both hot and cold.
[0033]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below with reference to the drawings.
In the following description, the term “location” means each vending machine installation location, or the vending machine itself provided at each installation location (if there are multiple units, they may be handled together). . Here, “if there are multiple units, they may be handled together” means that, for example, for a customer, there is a problem whether there is a desired product at a certain installation location. It is considered that there is no customer who thinks that they do not purchase if there is no product desired in the vending machine, so if there are multiple vending machines at a certain installation location, you may handle each vending machine individually It is possible to handle them collectively.
[0034]
FIG. 1 is a diagram showing a schematic functional configuration and a processing flow of the entire product configuration adjustment support system of a vending machine according to this example.
The system shown in the figure includes, as a functional configuration, an attribute feature analysis unit 1 for a vending machine installation location (location), a sales feature analysis unit 2 for a product / product based on location, a location classification processing unit 3-1 based on attribute features, and a sales feature. Location classification processing unit 3-2, product (product type) composition balance analysis unit 4, location competition analysis unit 5, location constraint analysis unit 6, inter-product correlation analysis unit 7, location in the location (vending machine) Location / product sales cross ABC analysis unit 8, product sales fluctuation time series analysis unit 9, adjustment target location extraction processing unit 10-1, specific location analysis unit 10-2, adjustment product type determination The processing unit 11, the adjusted product determination processing unit 12, and the adjustment method presentation (display) processing unit 13 are included. In addition, competition, constraint information A1, location attribute information A2, product classification information A3, sales performance information A4, product configuration, and sample information A5 are stored in the database in advance (for example, created by survey results or analysis / judgment results). Has been.
[0035]
In addition, it has a location / product gross profit cross ABC analysis unit 14 for analyzing location / product profit sources, a location cost / benefit balance analysis unit 15, a product cost / benefit balance analysis unit 16, and a product cost, The selling price and variable cost rate information A6 are stored in advance in the database.
[0036]
With the above configuration, the system of this example performs marketing analysis of vending machine sales products, finds problems that exist in the product configuration, selects vending machines that require adjustment, and how to select which product system Clarify whether adjustments (addition / removal, etc.) should be made, and give specific adjustment methods (draft measures) for which vending machines should be added / removed.
[0037]
First, the location attribute feature analysis unit 1 uses the location attribute information A2 for the vending machine installation location (hereinafter referred to as location) within the specified range, and uses the location attribute information A2 to determine the gender, age, etc. of the user. The factor analysis is performed by using several vending machine installation location evaluation indexes as categories.
[0038]
From the factor analysis results, find some common potential factors (common factors) on the installation site. Each installation location is evaluated using these common factors as a common measure. Moreover, the feature of each installation place can be grasped using the factor score (factor score) obtained by factor analysis.
[0039]
The sales feature analysis unit 2 uses the product classification information A3 in which the products handled in advance are classified into several product types (or product groups, the same applies below) and the sales performance information A4 to the installation location within the specified range. The sales occupancy rate of the product / product system for evaluating the sales characteristics of the product / product system at each installation location is calculated. Then, factor analysis is performed with the sales occupancy ratio of the product / product system as a category and each installation location as a sample, and some common potential factors (common factors) based on the sales performance of the product / product system are found.
[0040]
Then, each installation location is evaluated using these common factors as a common measure. In addition, using the factor score (factor score) obtained by factor analysis, the characteristics of each installation location are grasped. By doing this, it is possible to grasp the product / product sales characteristics at each installation location.
[0041]
The location classification processing unit 3-1 from the attribute feature performs cluster analysis using the processing result (factor score obtained by the attribute analysis of the installation location) by the location attribute feature analysis unit 1, and determines some vending machine installation locations according to the attributes. Into the group (hereinafter abbreviated as GrLi) and grasp the characteristics of the group GrLi. Further, this classification result is output as location classification information B3 by attribute to the adjustment target location extraction processing unit 10-1, the specific location analysis unit 10-2, the correlation analysis unit 7 between product systems, and the like.
[0042]
Similarly, the location classification processing unit 3-2 by sales feature performs cluster analysis using the factor score obtained by the product / product sales feature analysis at the installation location by the sales feature analysis unit 2, and installs the vending machine. The locations are classified into several groups (hereinafter abbreviated as GrSi) according to product / product sales characteristics, and the characteristics of the group GrSi are grasped. Further, this classification result is output to the adjustment target location extraction processing unit 10-1, the specific location analysis unit 10-2, the inter-product correlation analysis unit 7 and the like as location classification information B4 in sales.
[0043]
The number of groups of the group GrSi is the same as (or close to) the number of groups of GrLi. Then, the specific location analysis unit 10-2 later compares and analyzes the configuration and characteristics of the group classified based on the installation location attribute and the like and the group classified based on the sales results, and grasps the correspondence relationship between the groups.
[0044]
The specific location analysis unit 10-2 associates the group (GrLi) classified by the location classification processing unit 3-1 with the group (GrSi) classified by the location classification processing unit 3-2. Based on the correspondence, among the locations that are members of the group GrLi, those that are also members of the group GrSi corresponding to the group GrLi are classified as non-specific locations, and those that are not are classified as specific locations.
[0045]
The adjustment target location extraction processing unit 10-1 outputs the location classification information B3 with the attributes output by the location classification processing unit 3-1 from the attribute feature and / or the location classification processing unit 3-2 with the sales feature. The location classification information B4 (or the classification result by the specific location analysis unit 10-2) and the sales result information A4 stored in the database are input, and the location to be adjusted (location having a problem) is input. Extract.
[0046]
For example, when the classification result by the specific location analysis unit 10-2 is also used as an example, first, the average sales share of each product system in the non-specific location is calculated, and the product system of the group GrLi or / and the group GrSi Reveal the average sales share pattern. (Furthermore, the average sales share of each product system is calculated from several locations with the highest total sales in non-specific locations, and the product system average of Group GrLi and / or Group GrSi sales top locations (main location) Reveal sales share patterns). Then, comparing the sales occupancy pattern of a specific location with the average sales occupancy pattern of the non-specific location of the group GrLi and / or group GrSi that the location should belong to, there is a problem in the sales performance of which location Can be revealed.
[0047]
Alternatively, in the non-specific location of the group GrLi and / or the group GrSi, the location with the lowest sales for each group can be used as the adjustment location location.
Further, the adjustment target location extraction processing unit 10-1 further includes a location competition analysis unit 5, a product composition balance analysis unit 4, a location cost / benefit balance analysis unit 15, and a product type cost / benefit balance analysis unit 16. At least one processing result is input, and, for example, a vending machine (location) having a problem such as lack of product configuration, sample distribution, or necessary product system is selected for the location to be adjusted (location having a problem). You may make it add to a list.
[0048]
Hereinafter, the location competition analysis unit 5 and the product composition balance analysis unit 4 will be described. The location cost / benefit balance analysis unit 15 and the product cost / benefit balance analysis unit 16 will be described later.
[0049]
Product (product type) composition balance analysis unit 4 determines the sales occupancy rate of each product (product type), product type occupancy rate, product type from the product (product type) installed in the vending machine in the vending machine installation location. Sample occupancy rate is calculated, and the balance of these occupancy rates is analyzed, and there is a shortage of necessary varieties, wasteful use of samples, unsold products, lack of required products, etc. (balance) Location) is extracted, and this is output as product (product type) composition balance information B6. As described above, this information is used by the adjustment target location extraction processing unit 10-1 as necessary.
[0050]
For each location, the location competition analysis unit 5 uses other vendors' vending machines or other types of vending machines (Note: If there are other types of vending machines of the same type, they will not be competing because they are processed at the same installation location), or other competitors. Analyze the presence / absence of the same type of product (stores, convenience stores, restaurants, etc.) and the degree of competition (eg strong, medium, weak). Then, the analysis result is output as location competition information B2 (particularly, a list of locations of strong competition).
[0051]
As a result, it becomes possible to distinguish and grasp the impact on sales due to competition and the impact on sales due to product composition, which will affect the subsequent determination of the adjustment target (vending machine / installation location).
[0052]
Next, the processing result of the adjustment target location extraction processing unit 10-1, the processing result by the product composition balance analysis unit 4 or the location cost / benefit balance analysis unit 15 or the product cost / benefit balance analysis unit 16, Further, the adjusted product system determination processing unit 11 performs processing using the processing result of the product / product system correlation analysis unit 7 as necessary. Here, first, the product system correlation analysis unit 7 will be described.
[0053]
The product / product system correlation analysis unit 7 performs correlation analysis using the sales occupancy rate record of the product / product system based on the sales performance information A4 (or the analysis result by the sales feature analysis unit 2) and the location classification information B3 or B4. By grasping the correlation between the product / product system, the subsequent product / product system adjustment is supported. In other words, in the location within the group or within the specified range, products that have a positive correlation with the main product that can be sold should also sell, so if there are not enough product types, the number of products must be increased or installed. You can make additional adjustments.
[0054]
Based on the balance analysis performed by the product (product system) composition balance analysis unit 4 and the like, the adjusted product system determination processing unit 11 is a product system with a shortage of varieties, a product system with a lot of sample waste, and sales. An undiscovered product system is found, and a product system that is lacking from the result of the correlation analysis and an adjustment product system and an adjustment method (addition, removal, replacement, sample redistribution) are determined.
[0055]
Alternatively, the adjusted product system may be determined by comparing the product sales share of the adjustment-target vending machine (location) with the group average sales pattern and finding a product system lower than the group average product sales share. .
[0056]
The processing result of the adjusted product system determination processing unit 11, the location constraint analysis unit 6, the product sales fluctuation time series analysis unit 9, and the location / product sales cross ABC analysis unit 8 or the location / product gross profit cross ABC analysis unit 14 Using the processing result, the adjusted product determination processing unit 12 determines an adjusted product and an adjustment method for each location (vending machine).
[0057]
First, each of the location constraint analysis unit 6, the product sales fluctuation time series analysis unit 9, the location / product sales cross ABC analysis unit 8, and the location / product gross profit cross ABC analysis unit 14 will be described.
[0058]
The location constraint analysis unit 6 checks whether the vending machine installed at each installation site has a constraint such as being dedicated to a specific product manufacturer, and whether the constraint is all vending machines in the installation site, Check if it is a vending machine or a copy. Then, the processing result is output as location constraint information B1.
[0059]
In this way, by clarifying which of the restricted vending machines is, and which manufacturer is restricted to the more restricted vending machines, in the “addition or replacement” process in the adjusted product determination processing unit 12 later, Gives a condition for searching for additional / replacement products.
[0060]
The product sales fluctuation time series analysis unit 9 analyzes the product sales time series for the specific product (specified by the user), calculates the sales change rate of the product, and outputs this (product sales fluctuation rate information B8). And display again.
[0061]
As a result, it is possible to grasp the stage in which the product is in the growth period, maturity period (stable period), decline period (for a specific place), etc., and support the determination of the adjusted product. That is, in the processing of the adjusted product determination processing unit 12 later, when a product is added or replaced, a product in a growth period or a mature period (stable period) is set as an additional product or replacement product, and a product in a decline period is added or replaced. By excluding from replacement product candidates, more appropriate adjustment products can be determined.
[0062]
The location / product sales cross ABC analysis unit 8 uses the cross ABC analysis method (a publicly known method) to calculate the total sales of products placed within the designated location range, the average sales at the installation location, and the The sales are calculated, and a table (location / sales merchandise matrix B7) sorted in the order of higher average sales at the installation location is created and output / displayed. Thereby, the adjusted product determination processing unit 12 uses this matrix B7 to search for the adjusted product and determine the added / removed product.
[0063]
The location / product gross profit cross ABC analysis unit 14 is basically the same as the location / product sales cross ABC analysis unit 8 described above, except that it uses revenue / profit instead of sales. That is, calculate the total revenue / profit of products placed in the designated location, the average revenue / profit at the installation location, and the revenue / profit at each installation location, A table (location / product gross profit ABC matrix B12) sorted in the above is created and output / displayed.
[0064]
The adjusted product determination processing unit 12 is determined by the adjusted product system determination processing unit 11 using the matrix B7 or B12 obtained by the location / product sales cross ABC analysis unit 8 or the location / product gross profit cross ABC analysis unit 14. If the adjustment method is product addition, product removal, product replacement, or sample redistribution, the following processing is performed.
[0065]
In the case of adding a product, among the products of the adjustment target product system that are not placed in the adjustment target location (vending machine) (is insufficient), the higher-order products in the matrix B7 or B12 are selected. , And additional product candidates. However, if there is a restriction that the adjustment target vending machine is dedicated to a specific manufacturer, an additional product candidate is found from the products of this specific manufacturer. Furthermore, products that are determined to be products in a decline period by time series analysis of products using the processing results of the product sales fluctuation time series analysis unit 9 may be excluded from additional product candidates.
[0066]
In the case of product removal, products are searched for in the lower order in the adjustment target product system in the adjustment target location (vending machine), and product candidates to be removed are found.
In the case of product replacement, the product removal operation and the product addition operation are sequentially performed.
[0067]
As described above, the determined adjustment product information and the adjustment method (product addition, removal, replacement) information are notified to the adjustment method presentation (display) processing unit 13.
In the case of sample redistribution, sample redistribution information for the product is delivered to the adjustment method presentation (display) processing unit 13.
[0068]
The adjustment method presentation (display) processing unit 13 displays the notified adjustment product information and adjustment method (addition, removal, replacement, sample redistribution) of the adjusted product as an adjustment method (proposed measure) for the adjustment target vending machine. Etc.
[0069]
In the above analysis and processing, it is possible to use the product sales profit share instead of the product sales share.
Each functional configuration of FIG. 1 described above will be described in more detail below.
[0070]
First, the location attribute feature analysis unit 1 will be described with reference to FIGS.
FIG. 2 is a flowchart for explaining a processing procedure in the location attribute feature analysis unit 1.
[0071]
First, location attribute information A2 created based on a survey or the like conducted in advance and stored in the database is acquired (step S11). The location attribute information A2 indicates each vending machine installation location (location) within the specified range, for example, evaluation items as shown in FIG. 3 (user gender, age group, composition, occupation group, usage time zone, etc .; (V1 to V10 are assigned), and the result is 5 levels as shown in FIG. 4 (not limited to this, any of 4 levels,..., 2 levels (0/1) may be used) A table quantified by numerical values is created (for example, created by a user). As shown in the figure, each location is assigned a unique location ID.
[0072]
Next, factor analysis processing is performed using the data in the table shown in FIG. 4 (step S12). Factor analysis, principal component analysis, cluster analysis, and other methods are well known as one method of multivariate analysis, and there are programs that implement these methods on computers. There is no particular description in the description (for example, there are edited by Masanori Kodama; Toshio Sugihara and Wataru Fujita; “Economic Information and Mathematical 13, Multivariate Analysis”, etc.).
[0073]
In the process of step S12, the factor analysis process is executed with the location number / location ID in FIG. 4 as a sample term (specimen) and the evaluation value (location evaluation factor) for the location evaluation item as a category term.
[0074]
Then, when the factor load as shown in FIG. 5 is obtained by the above factor analysis (in addition, by creating a factor load bar graph as shown in FIG. The user interprets each factor (C1) (step S13). As is well known, there is no fixed interpretation of factor interpretation. For example, taking factor 1 shown in FIGS. 5 and 6 as an example, the evaluation item V8, V6, V7, V5 is “small single, family Since it can be determined that “multiple, many couples, 70s and many”, factor 1 can be interpreted as, for example, “(location location is a public service facility, hospital, etc.)” (of course, such as It is not limited to interpretation, but depends on the user's (analyst) mindset).
[0075]
Similarly, factor 2 can be determined as “brain labor system, many women” from the evaluation items V9, V1, and therefore can be interpreted as, for example, “company with many women (office work system), hospital”.
Factor 3 can be interpreted as “10 to 20 generations high, 50 to 60 generations low” from the evaluation items V4 and V2, and can be interpreted as, for example, “young people”.
[0076]
Next, the above-mentioned sample terms (ie, each location) are re-acquired by the above factors 1 to 3 (ie, the factor score C2), and each location is indicated by this factor score (eg, as shown in FIG. 7). The attribute feature (attribute characteristic) is grasped (step S14). At this time, by creating a bar graph of this factor score (FIG. 8), it becomes easier to visually grasp the attribute features of each location. FIG. 8 (a) shows each location evaluated by factor 1, FIG. 8 (b) shows each location evaluated by factor 2, and FIG. 8 (c) shows each location evaluated by factor 3. is there.
[0077]
Note that principal component analysis may be used instead of (or in addition to) factor analysis.
Next, the sales feature analysis unit 2 will be described in detail with reference to FIGS.
[0078]
FIG. 9 is a flowchart for explaining the processing procedure in the sales feature analysis unit 2.
In this process, basically, factor analysis processing is performed in the same manner as the location attribute feature analysis unit 1 on the basis of the product classification information A3 and the sales performance information A4 stored in the database, and each location is classified as a sales feature. In this example, the following process is performed before the factor score to be evaluated is obtained.
[0079]
First, the sales record information A4 stored in the database is acquired (step S16). Then, the total sales for each location is obtained, and ABC analysis is performed based on the total sales (step S17).
[0080]
ABC analysis is a well-known classification technique, but will be briefly described here.
In the ABC analysis in step S17, the locations are first sorted in descending order of total sales. Next, the cumulative total sales occupation ratio is calculated in order from the top location. Locations that are within α% of the cumulative total sales share are ranked A, and locations that are within β% of the cumulative total sales share are ranked B, and the remaining locations Each location is ranked according to the total sales, such as the location within γ% of the cumulative total sales share is C rank (note that α <β <γ, the value is User (analyzer, etc.) may arbitrarily set).
[0081]
Then, for example, data of a location where sales are extremely low (for example, the location of the C rank or a rank lower than the C rank) is not used (excluded) in the processing after step S19 described below (step). S18).
[0082]
The processing after step S19 will be described below. First, data used in this processing will be described.
FIG. 10 is an example of the product classification information A3 in FIG. As shown in the figure, each product is classified in advance into several product systems (also referred to as a product group and a product group) and stored in the database as product classification information A3. For example, it is classified into several product systems such as carbonated drink system, fruit drink system, and coffee drink system, and each product system name (S, F, C,... X) is attached.
[0083]
The sales performance information A4 is basically the sales performance of each product for each location (or each vending machine), but further includes the sales for each product system (refer to the product classification information A3). Or the sales for each product system may be calculated by referring to the product classification information A3 in this process. In the process of step S19, the sales occupancy rate of each product system in each location is calculated for each location based on the sales result data for each product system. For example, the product by location as shown in FIG. A system sales occupation ratio list C3 is created (therefore, the unit of the numerical values shown in the figure is% (percent)). In the example shown in the figure, for example, in a location with a location ID of 1001, sales of products belonging to the product system W (mineral water) account for nearly half (49.26%) of the total sales in the location.
[0084]
These are values of an index for evaluating sales at each location (product system sales occupancy rate evaluation factor).
Next, factor analysis is performed using the table shown in FIG. The location analysis (location ID or location No.) shown in FIG. 11 is used as a sample term, and the sales occupancy rate evaluation factor (S, F, C,... X) of the product system is used as a category term. It performs (step S20).
[0085]
Then, when the factor load as shown in FIG. 12, for example, is obtained by the factor analysis (further, a factor load bar graph as shown in FIG. 13 may be created to make it easier to grasp visually. The user interprets each factor (C4) (step S21). In the example shown in FIGS. 12 and 13, factor 1 can be interpreted as, for example, “soft, feminine” from evaluation factors F (fruit beverage system), K (lactic acid bacteria beverage), and S (carbonated beverage system).
[0086]
Similarly, factor 2, factor 3 and factor 4 can be interpreted as shown in FIGS. 12 and 13 as an example.
Next, the above-described sample terms (ie, each location) are re-acquired (ie, factor score C5) using the above-described factors 1 to 4, and each location is determined based on this factor score (eg, as shown in FIG. 14). Grasping sales characteristics at (step S22). At this time, by creating a bar graph of this factor score (FIG. 15), it becomes easier to visually grasp the sales characteristics at each location. FIG. 15 (a) shows an evaluation of each location by factor 1, FIG. 15 (b) shows an evaluation of each location by factor 2, FIG. 15 (c) shows an evaluation of each location by factor 3, FIG. 15D shows an evaluation of each location by factor 4.
[0087]
Note that principal component analysis may be used instead of (or in addition to) factor analysis.
Next, the location classification processing unit 3-1 from the attribute feature and the location classification processing unit 3-2 from the sales feature will be described with reference to FIGS.
[0088]
The location classification processing units 3-1 and 3-2 have the same basic process flow (shown in FIG. 16). Hereinafter, with reference to FIG. 16, the location classification processing unit 3-1 from the attribute features will be described first.
[0089]
The location classification processing unit 3-1 from the attribute features first acquires the factor score (FIG. 7) obtained by the location attribute feature analysis unit 1 (step S24).
Each factor (factor 1 to factor 3) in this factor score (FIG. 7) is a category, each location (location No / location ID) is a sample term, and each numerical value shown in FIG. Using the evaluation result as sample data, cluster analysis processing for the location is executed (step S25).
[0090]
As a result of the cluster analysis, the process and result of the cluster analysis are displayed as a tree diagram (dendrogram) which is a graph of a tree structure based on the classification distance between members (FIG. 17) (step S26).
[0091]
The user (analyst) refers to the displayed tree diagram (dendrogram), determines the number of groups to be divided, and designates / inputs this. This designation / input method is well known. For example, as shown in FIG. 17, the position where the dendrogram is divided is designated (vertical dotted line; divided into three groups in the illustrated example).
[0092]
Thus, on the system side, the location is grouped by the designated number of groups, and the result (“location ID table with group number” in which each location is given its group number (GrL1, GrL2, GrL3 in the example shown). ) Is output (B3) (step S27). For example, as an example, the group numbers GrL 3 are associated with the locations 4, 15, and 28 in FIG.
[0093]
Note that the “matching rate” shown in the figure is determined in comparison with the groups GrS1, GrS2, GrS3, and GrS4 in FIG. 18 in the group association process in the specific location analysis unit 10-2 later. It is not used in this process.
[0094]
The location classification processing unit 3-2 in the sales feature basically performs the above-described processing shown in FIG. 16, and the data input in step S24 is different, and only the processing result is different.
[0095]
That is, the location classification processing unit 3-2 for sales features first acquires the factor score (FIG. 14) obtained by the sales feature analysis unit 2 (step S24).
Each factor (factor 1 to factor 4) in this factor score (FIG. 14) is a category, each location (location No / location ID) is a sample term, and each numerical value shown in FIG. Using the evaluation result as sample data, cluster analysis processing for the location is executed (step S25).
[0096]
As a result of the cluster analysis, the process and result of the cluster analysis are displayed as a tree diagram (dendrogram) which is a graph of a tree structure based on the classification distance between members (FIG. 18) (step S26).
[0097]
The user (analyst) refers to the displayed tree diagram (dendrogram), determines the number of groups to be divided, and designates / inputs in the same manner as described above. At that time, the number of groups divided in FIG. It is desirable that the number of groups be the same as or close to that.
[0098]
Thus, on the system side, the location is grouped by the designated number of groups, and the result (the location ID with a group number is formed by attaching each group to its location with its group number (GrS1, GrS2, GrS3, GrS4 in the illustrated example). Table ") is output (B4) (step S27). The “match rate” shown in the figure is as described with reference to FIG.
[0099]
Next, the specific location analysis unit 10-2 will be described with reference to FIGS.
FIG. 19 is a flowchart for explaining processing of the specific location analysis unit 10-2. FIG. 20 is a flowchart for explaining details of the group association processing in step S29.
[0100]
In FIG. 19, first, information B3 of a group (hereinafter referred to as a GrL group) classified by the attribute feature analysis unit 1 and the location classification processing unit 3-1 from the attribute feature, a sales feature analysis unit 2, The information B4 of the group (hereinafter referred to as GrS group) classified by the processing of the location classification processing unit 3-2 in the sales feature is input. Then, group association processing is performed (step S29).
[0101]
This group association processing will be described with reference to FIG.
The group association here is performed by the analyst through the GUI based on the structure of the groups based on the respective classification results and the analysis of the characteristics based on the factor interpretation (that is, based on the contents shown in FIGS. 17 and 18). Alternatively, the system will do it automatically. The analyst first designates the number of groups (step S35). Next, the characteristics of the group structure and factor interpretation are analyzed (step S36). Then, the GrL group and the GrS group are associated with each other and the result is input.
[0102]
The above process is a process for obtaining the “match rate” shown in FIGS. 17 and 18 and determining a group to be associated with the “match rate”. That is, for example, when the group GrL1 is designated, the GrL1 includes 20 locations from the location No. 29 to 27 shown in FIG. 17, but when searching for locations included in the GrS1, a total of 16 locations are included. There are places. Therefore, the “match rate” of GrL1 with respect to GrS1 is 16/20, which is 80%. Although not specifically shown, it is obvious that the “match rate” between GrL1 and the other groups GrS2 and GrS3 is considerably lower than 80%. As a result, GrL1 and GrS1 are (Conversely, 19 locations are included in GrS1, and 16 locations are included in GrL1 as well, so the “match rate” is 84/16. %, GrL1 and GrS1 may be associated with each other). GrS4 is a unique location because it is not associated with GrL1.
[0103]
In the example shown in FIGS. 17 and 18, similarly, GrL2 and GrS2 are associated with each other (GrS2 side coincidence rate is 3/4, which is high as 75%), and GrL3 and GrS3 are associated with each other (GrL3 side). The coincidence rate of 3/3 is 100%). This is stored as group correspondence information C8.
[0104]
Returning to the description of FIG. Based on the association result in step S29, the member configurations of the corresponding groups are compared to find uniqueness (location that is not a member belonging to the corresponding group). For example, in the example shown in FIGS. 17 and 18, when GrL1 and GrS1 are associated with each other as described above, for example, the locations 29, 30, 12, and 13 among the locations belonging to GrL1 Since 29 and 30 are not in GrS1, these are specific locations (step S30, NO), and registration processing for adding these location IDs (= 29, 30) to the specific location list C9 is performed (step S3). 1). On the other hand, since the locations 12 and 13 are also members of GrS1 (step S30, YES), they are not specific locations (hereinafter referred to as non-specific locations).
[0105]
As described above, for each group, the processing for extracting the specific location from the locations belonging to the group is performed. When the processing is completed for all locations (step S32, YES), the completed specific location list C9 is completed. And a sales record information A4 input from the database, a list C10 indicating a sales pattern of specific locations is created (this is a part of FIG. 29 described later (however, at this stage, “adjustment”) , “Contest” data is not included), and this is displayed (step S33).
[0106]
As described with reference to FIG. 1, the adjustment target location extraction processing unit 10-1 basically extracts a location to be adjusted (adjustment of the product configuration is required) based on the processing result described above. Furthermore, since the location competition analysis unit 5, the product composition balance analysis unit 4, the location cost / benefit balance analysis unit 15, and the product cost / benefit balance analysis unit 16 may be added, First, these will be described.
[0107]
First, with reference to FIG. 21, a processing procedure by the location competition analysis unit 5 will be described.
The location competition analysis unit 5 analyzes the presence / absence and the degree of competition in each location (vending machine installation location).
[0108]
First, location installation information A1 (including competition definition information) based on a survey or the like is stored in advance in the database.
The acquired location information A1 (including competitiveness definition information) from the acquired location information A1 (including competitiveness definition information) indicates that the location of this vending machine or another type of vending machine of your own company Analyzing whether or not there is any other competition (sellers, convenience stores, restaurants, etc. that handle similar products) and the degree of competition (for example, strong, medium, weak) Yes (step S40).
[0109]
Therefore, based on these, in the location ID table, as location conflict information B2, an identification symbol indicating the degree of competition (for example, strong, medium, weak) is attached to the location with conflict (step S41).
[0110]
Further, a location with strong competition is extracted (step S42), and a location list with strong competition is created (step S43). This is for providing information to the adjustment location selection exclusion process in the subsequent adjustment target location extraction processing unit 10-1.
[0111]
The above process is repeated until the analysis is completed for all locations (step S44).
The processing result of the location competition analysis unit 5 is used to distinguish and grasp the influence of the competition on the sales and the influence of the product composition on the sales later, and determine an appropriate adjustment target (vending machine / installation location). Useful.
[0112]
Next, the product configuration balance analysis unit 4 will be described with reference to FIGS.
FIG. 22 is a flowchart for explaining the processing procedure of the product composition balance analysis unit 4.
[0113]
The product composition balance analysis unit 4 first obtains, from the database, sales performance information A4 of products (product series) sold by each vending machine installed in each location, product composition, and sample information A5 ( Step S51).
[0114]
Then, for each location, the sales occupancy rate SR, the product type occupancy rate VR, the product type occupancy rate MR of each product (product type) are calculated for each location, and the SR / VR / MR data table is calculated. C14 is created (step S52). Alternatively, for the sales occupation rate SR, the sales occupation rate list C3 of the merchandise system by location obtained by the sales feature analysis unit 2 may be used.
[0115]
Here, the type occupation rate VR and the sample number occupation rate MR are calculated by the following equations.
Variety occupancy rate VR = number of product types in the target product system / total number of product types handled at this location
Sample occupancy rate MR = total number of samples used for products of the target product system / total number of products handled at this location
For example, in a location, there are two types of product type C (coffee beverage type) product types (assuming milk coffee and black coffee), and the total number of product types including these is 10 types. In addition, it is assumed that there are 4 samples of milk coffee and 2 samples of black coffee, and there are 20 samples of all products including these (such information includes product configuration, It is stored in advance in the database as sample information A5). In this case, the type occupation rate VR and the sample number occupation rate MR are
Variety occupation rate VR = 2/10 = 20 (%)
Sample number occupation ratio MR = 6/20 = 30 (%)
It becomes.
[0116]
The processing in steps S51 and S52 is repeated for all locations, and when the processing for all locations is completed (YES in step S53), the balance of the product composition is analyzed by comparing these occupancy rates.
[0117]
That is, based on the SR, MR, and VR values obtained in step S52, for example, if SR >> MR >> VR in a certain merchandise system with a certain location, the merchandise system has few varieties and is not balanced. I understand. Besides these, there are MR >> SR >> VR, MR >> VR >> SR, VR >> MR >> SR, VR >> SR >> MR, SR >> VR >> MR, etc. Will be described later with reference to FIGS. Note that the size of “>>” is determined / set in advance by an analyst or the like based on experience or the like.
[0118]
From the analysis result, an unbalanced installation location (location) is extracted, and an unbalanced location list (and their SR / VR / MR data table) C15 is created (step S54).
[0119]
Next, it is determined whether or not the balance visualization is performed (the user determines and gives an instruction) (step S55).
If step S55 is YES, a radar chart is created using the SR, MR, and VR values, and the balance of the structure is visually observed to analyze it more intuitively. “Additional information to target location list” C16 can be created (step S56).
[0120]
When step S55 is NO, the unbalanced location list C15 may be used as the additional information C16 as it is.
FIG. 23 is a detailed flowchart of the balance visualization process in step S56, and FIG. 24 is a diagram showing an example of visualization.
[0121]
First, an analyst or the like designates a location range to be visualized (whether only unbalanced location or all locations) (step S58).
In response to this, when the visualization target is only the unbalanced location (step S59, YES), the data of the data table C15 is acquired (step S60), and when the visualization target is all location ( (Step S59, NO) Data in the data table C14 is acquired (Step S61).
[0122]
Then, based on the data acquired in any of the above steps S60 and S61, a radar chart creation process is performed. At this time, the analyst designates one location to be visualized (step S62, YES), a radar chart is created and displayed for the designated location (step S63).
[0123]
The radar chart created / displayed is, for example, as shown in FIG. 24. As can be seen from FIG. 24, it is easy to visually grasp the balance of the product structure, and unbalanced locations can be obtained. It becomes easy to judge. “The product composition is not balanced” means, for example, a lack of necessary product system in addition to the above, and the system or analyst should make such a determination as an adjustment target after making such a determination. If it is determined whether or not to be adjusted (step S64, YES), the “additional information to the adjustment target location list” C16 is created by instructing it. Is performed (step S65).
[0124]
By doing in this way, the location where the product composition is not balanced is added to the adjustment target location list in the subsequent processing.
Next, processing contents of the location cost / profit balance analysis unit 15 will be described with reference to FIG.
[0125]
First, sales performance information A4, product classification information A3, product sales price, and cost information A6 are acquired from the database (step S67).
Next, based on the acquired data, the gross margin Ra for each location and the total gross margin Rall for all locations are calculated, and the ratio between the gross margin Ra and the total gross margin Rall is calculated for each location (step S68). ).
[0126]
Here, the gross margin Ra and the total gross margin Rall for each location are obtained by the following equations.
[0127]
[Expression 1]

[0128]
When the above processing is completed for all locations (step S69, YES), unbalanced locations are extracted based on the ratio between the gross margin Ra and the gross gross margin Rall, and this is output (or displayed). (Step S70). That is, the location where the cost / benefit balance is not extremely compared with the whole is extracted / output as the adjustment target candidate.
[0129]
Next, with reference to FIG. 26, processing contents of the product cost / profit balance analysis unit 16 will be described.
First, sales performance information A4, product classification information A3, product sales price, and cost information A6 are acquired from the database (step S67).
[0130]
Next, based on the acquired data, for each location, the above-mentioned gross profit rate Ra for each location, and the gross profit rate Rk for each location and each product type are calculated, and the ratio of Ra and Rk is obtained (step S73). ).
[0131]
Here, the gross profit rate Rk for each product type for each location is obtained by the following equation.
[0132]
[Expression 2]

[0133]
When the above processing is completed for all locations (step S74, YES), an unbalanced commodity system is extracted based on the ratio of the gross profit rate Ra and Rk, and this is output (or displayed) ( Step S75). That is, according to the processing result up to step S73 (display this), for each location, a product system (unbalanced product) in which the balance of cost / benefit is not extremely compared with the gross profit rate Ra of the entire location. Analysts, etc. can determine whether such an unbalanced product system exists, and the location where such an unbalanced product system exists is subject to adjustment in later processing. Can be an additional candidate for location.
[0134]
It should be noted that analyzing the cost / benefit balance as described above is effective regardless of whether the vending machine is an operator company or a manufacturer company. An operator company is a company that purchases (rents) vending machines, purchases the contents from the manufacturer, and sells the products with the vending machine. The feature is that the products of a plurality of product manufacturers are placed in the same vending machine, and the product cost varies depending on the manufacturer / variety (there is a product with a large difference). Therefore, it is desirable for operator companies to sell many products with low cost (high gross profit per product) if the products have the same selling price, but products with low cost are usually unpopular products. Since there are many cases and the number of sales is low, after all, "gross profit x number of sales" is low (the profit is thin). On the other hand, merchandise with a large number of sales (popular merchandise) usually tends to have high costs (low gross profit per merchandise). Therefore, even if it sells a lot, there are cases where the profit is thin (however, the above is a general trend, and if you look at each product, for example, even if it is a product with a low cost, you can sell a lot There may be products).
[0135]
Because of the circumstances described above, it is important for the operator company to balance the cost / benefit, and the location where the unbalanced product system exists is considered as an additional candidate for the location to be adjusted. That is very meaningful.
[0136]
On the other hand, in the case of a manufacturer company, the products handled by vending machines are only their own products. That is, only a single product manufacturer's product is placed in one vending machine. However, since the cost of goods varies depending on the product type, it is still effective to analyze the cost / benefit balance.
[0137]
The adjustment target location extraction processing unit 10-1 includes the above-described specific location analysis unit 10-2, location competition analysis unit 5, product composition balance analysis unit 4, location cost / benefit balance analysis unit 15, product type cost / The processing result of the profit balance analysis unit 16 and the location classification information B3 by attribute, the location classification information B4 by sales, and the sales performance information A4 (however, not all of them are used) are shown in FIG. Execute processing to select location candidates to be adjusted.
[0138]
Hereinafter, the adjustment target location extraction processing unit 10-1 will be described with reference to FIGS. 27 to 29.
FIG. 27 is a flowchart for explaining a processing procedure in the adjustment target location extraction processing unit 10-1.
[0139]
First, the location classification information B3 by attribute and the sales performance information A4 stored in the database, which are created by the location classification processing unit 3-1 from the attribute feature, are acquired (step S77). However, this is only an example, and the location classification information B4 in sales may be used.
[0140]
Next, the location / product sales share list C20 of non-specific location (main location / main product system table of each group) is created for each group (GrL) by the processing of step S78 and step S79 (FIG. 5). 29 shows an example of GrL1, but GrL2 and GrL3 are created in the same manner; FIG. 29 also shows a list of specific locations C10).
[0141]
In accordance with the location classification information B3 (GrL1, GrL2, GrL3 in the above example), for each group (GrL), a merchandise sales occupation rate ABC analysis by location is performed for all locations belonging to that group (step S78). Further, since the non-specific location is known from the specific location list C9, the average of the sales occupancy rate of each product system of the non-specific location is calculated. For example, in the example of FIG. 29, the average sales share of each product system of non-specific location (location No = 11, 8, 6,... 22) is 49.04 for product system C, 14.02 for product system N, etc. FIG. 28 is a bar graph for visual understanding. Each product system is sorted in descending order of average sales share (in FIG. 29, the product systems C, N, W, D... From the left are the results of sorting) ( Step S79).
[0142]
In this way, for each group, the average of the product sales share of the location excluding the specific location that is unique in expressing the characteristics of the group (that is, non-specific location) is calculated, and the product Sorting the systems in order from the largest to the smallest sales occupancy rate reveals the sales pattern of the group and makes it easier to grasp the sales characteristics such as what the group's main product system is (FIG. 29). In the example, it is understood that the main product system of GrL1 is the product system C, N, W, etc.).
[0143]
Further, in FIG. 29, non-specific locations are sorted in order from the top of the total sales, and in the example shown, they are sorted in the order of location numbers = 1, 8, 6,. As shown in the figure, the specific location may be sorted in order from the top of the total sales. However, in the case of the specific location, the location at the bottom of the total sales is not always a candidate for adjustment. May be.
[0144]
In addition, the average sales share of each product system is calculated only for locations with the highest total sales in non-specific locations, and the sales pattern / sales characteristics of the group GrLi's top sales (main location) are calculated. It can also be revealed).
[0145]
By creating the non-specific location / occupation sales ratio list C20 (main location / main product table of each group) for each group by the above-described processing, the location to be adjusted based on this is created. The candidate (primary candidate) is selected.
[0146]
This selection may be made for non-specific locations in order from the bottom of the total sales for each group (in the example shown, only the lowest location (location No = 22) is selected for adjustment). (A circle is shown in the leftmost “adjustment” item in the figure)), or a location that has been ranked C by ABC analysis based on total sales is selected as a primary candidate (step S81).
[0147]
In addition, the specific location is comprehensively determined by, for example, an analyst. That is, as described also in FIG. 1, the sales share pattern of a specific location is expressed as the average of the product-based sales share by the non-specific location of the group GrLi (or / and group GrSi) to which the location belongs ( By comparing with the sales pattern of the group; the processing results of steps S78 and S79), an analyst or the like can determine which location has a problem in sales performance.
[0148]
From this, for example, in the example shown in the figure, the analyst or the like determines that the location No. 16 and 19 should be adjusted, and instructs / inputs (circled in the “adjustment” item on the left side in the figure). (Step S80).
[0149]
In addition, it is good also considering the thing with low total sales in a specific location as an adjustment object location. In addition, both of the processes of steps S80 and S81 may be performed, or one of them may be performed.
[0150]
When the above-described processing is performed for all the groups and the primary candidate location list for the adjustment target location is created, the above-described unbalanced location is added to the primary candidate list, and the secondary candidates, tertiary candidates,...・ Create a candidate list. Note that not all the processes of steps S82 to S85 described below are performed. Any one or more processes may be performed (by an analyst selecting / instructing etc.), or the primary candidate may be used as the final candidate list as it is.
[0151]
The processing in steps SS82 to S84 has no particular processing order. However, in this case, the primary candidate using the additional information C16 to the adjustment target location list created by the product composition balance analysis unit 4 will be described with reference to the drawings. A location candidate “not balanced (extremely)” is added to the location list to create a secondary candidate location list (step S82).
[0152]
A location where the cost / benefit balance is not (extremely) compared to the average of all locations obtained by the location cost / profit balance analysis unit 15 is added to the secondary candidate location list as a candidate for adjustment. A tertiary candidate location list is created (step S83).
[0153]
There is a product system (unbalanced product system) in which the cost / profit balance is not (extremely) out of comparison with the gross margin Ra of the entire location determined by the cost / profit balance analysis unit 16 of the product system. The location is added to the tertiary candidate location list as an adjustment target candidate, and a quaternary candidate list is created (step S84).
[0154]
Then, the quaternary candidate list is created by excluding the location with strong competition obtained by the location competition analysis unit 5 (step S85).
The location list obtained by the above processing is output as the final candidate location list C21, and the adjustment location list is output and notified to the adjusted product system determination processing unit 11 (step S86).
[0155]
The process shown in FIG. 27 is an example, and the present invention is not limited to this. For example, the same processing as in steps S78, S79, and S81 is performed for each group classified by the location classification information B3 by attribute or the location classification information B4 by sales without classifying into specific location / non-specific location. It may be.
[0156]
Alternatively, the location classification information B3 and B4 is not limited to determining a location to be adjusted by analyzing a plurality of locations, but should each location be processed independently, and that location be the location to be adjusted? It may be determined whether or not. That is, the unbalanced location may be set as the location to be adjusted using only the analysis result by the product composition balance analysis unit 4 or the location cost / benefit balance analysis unit 15 or the product cost / benefit balance analysis unit 16. Good.
[0157]
Next, the adjusted product system determination processing unit 11 processes the processing result of the adjustment target location extraction processing unit 10-1, the product composition balance analysis unit 4, and the processing of the correlation analysis unit 7 between product systems described below. Using the result, a processing unit for determining the adjusted product system is performed.
[0158]
Hereinafter, the correlation analysis unit 7 between product systems will be described with reference to FIGS. 30 to 32.
FIG. 30 is a flowchart for explaining the processing procedure in the correlation analysis unit 7 between product systems.
[0159]
Note that this processing itself will be described simply because a correlation coefficient matrix (correlation coefficient and significance level based on significance probability) as shown in FIG. 32, for example, is obtained using an existing analysis tool.
[0160]
First, the sales result information A4 (or “product type sales occupancy rate evaluation table C3” obtained by the processing of the sales feature analysis unit 2) and the location classification information B3 or B4 stored in the database are acquired. (Step S89).
[0161]
Next, for each group, a correlation coefficient calculation between each product system in the group, a significant analysis, and a creation process of a correlation matrix are performed (step S90). This obtains a correlation coefficient with the sales occupancy rate of the product system for all product systems handled in the location belonging to each group.
[0162]
31 (a) and 31 (b) are explanatory diagrams of product system correlation analysis.
The formula shown in FIG. 11A is an example of a calculation formula for the correlation coefficient r, which is a calculation formula for the correlation coefficient ru between the product system U and the product system T shown in FIG. is there.
[0163]
The correlation coefficient is always in the range of −1 to 1. That is, when the positive correlation is strong, the value is close to 1, when the negative correlation is strong, the value is close to −1, and when the correlation is weak, the value is close to 0.
[0164]
The process of step S90 described above is executed for all combinations of product systems, and these are sorted together (sorted for a certain product system by the value of the correlation coefficient with other product systems), as shown in FIG. A correlation matrix C17 is created and displayed (step S91). Further, in the significance calculation process, a product system having a significant correlation coefficient is extracted, and a marker is displayed on the matrix (in FIG. 32, * and ** are added to the right side of the numerical value and displayed. , * Means significant at the significance probability level of 5% (both sides), ** means significance at the significance probability level of 1% (both sides)) (step S92).
[0165]
By referring to the correlation matrix as shown in FIG. 32, the analyst can know which product system and which product system have a certain degree of correlation and the order of the correlation degree. Furthermore, the significance of the correlation between commodity systems can be grasped.
[0166]
Here, the grasp of significance will be described. The significance probability (or P value) is “the probability that a statistic is calculated from data obtained by a single survey and a statistic greater than that value is obtained”.
[0167]
For example, if the significance probability is smaller than a certain value, it becomes suspicious that “H0 is correct” (rejection), and thus the validity of H1 is supported (adopted). The criterion of how small the significance probability should be to reject H0 (adopt H1) is called the “significance level”, which is represented by α. The significance level of 5% means α = 0.05, which means that there is a risk of erroneous determination once in 20 times when testing is performed under the same conditions. Alternatively, the probability of rejecting H0 by mistake even though H0 is correct in one determination is 5 (%).
[0168]
Thus, the significance can be grasped by referring to the correlation matrix as shown in FIG.
When the correlation analysis within the group is completed by the above-described processing (step S93, YES), the correlation analysis between the product systems for all locations is performed in the same manner as the above-described correlation analysis within the group ( Step S94). And this analysis result is output as correlation information C19 between each product system in all locations. Corresponding to the fact that the actual situation of the product system installed in the group may not reflect the actual characteristics of the installation location. The above deficiency is compensated by performing correlation analysis.
[0169]
By grasping the correlation between the product / product system by the above processing, the adjustment of the product / product system in the adjusted product system determination processing unit 11 is supported later. In other words, a product system that has a positive correlation with a very popular product system (main product system) should be sold within the group or within the entire range, so if there are not enough product types It is possible to obtain an appropriate product configuration by making adjustments such as adding varieties or adding products if the product of the corresponding product system is not handled (sold) in that location.
[0170]
In the above processing, the sales occupancy rate of the product may be used instead of the sales occupancy rate of the product system.
Hereinafter, with reference to FIGS. 33 to 35, processing in the adjusted product system determination processing unit 11 will be described.
[0171]
FIGS. 33 to 35 are flowcharts (No. 1), (No. 2) and (No. 3) for explaining the processing procedure in the adjusted product system determination processing unit 11. FIG.
First, the analyzer or the like designates the adjustment target location by displaying the adjustment candidate location list C21 (final candidate location list) obtained by the adjustment target location extraction processing unit 10-1 (step S96).
[0172]
Then, the processing of step S97 to step S118 is executed for the designated location, and when the target product system / adjustment method list C22 (product system to be adjusted and its adjustment method list) is created, the next adjustment target If there is a location, specify it and repeat the same process. Instead of designating by an analyst or the like, the processing may be performed in order from the top of the adjustment candidate location list.
[0173]
First, the location / commodity sales occupancy ratio lists C10 and C20 obtained by the adjustment target location extraction processing unit 10-1 (a list of groups to which the adjustment target location belongs; for example, if the adjustment target location belongs to the group GrL1, FIG. Information) is acquired (step S97).
[0174]
Furthermore, the balance information B6 obtained by the product composition balance analysis unit 4 is acquired (step S98).
First, based on the location / commodity sales occupancy rate list (for example, FIG. 29), for each product system, the product sales occupancy rate (SR) of the adjustment target location and the average product of the group to which the adjustment target location belongs. The system sales occupancy rate (Gr average) is compared, and each product system is classified into two depending on whether SR is higher or lower than the Gr average (step S100).
[0175]
However, if there is a product system with SR = 0 before that (step S99, YES), the processing of steps S113 to S115 described later is performed for this product system.
[0176]
Next, based on the balance information acquired in step S98, a product system that is not balanced is determined as an adjustment target product system, and an adjustment method is determined (steps S101 to S112).
[0177]
As shown in the figure, the conditions shown in step S101 (conditions for making an unbalanced product system; SR >> MR >> VR, MR >> SR >> VR) and the conditions shown in step S107 are the same. The adjustment method differs depending on the determination result. The same applies to step S103 and step S109. The same is true for step S105 and step S111.
[0178]
First, when there is a product system that satisfies SR <Gr average (YES in step S100) and satisfies the condition (SR >> MR >> VR or MR >> SR >> VR) shown in step S101 (step S100). (S101, YES), the product system is determined to be the adjustment target product system, and the adjustment method is determined. If the condition shown in step S101 is satisfied, it is considered that “the existing product type (currently sold at the location) is insufficient in the product system”, and therefore, “add another product of the same system by dividing the number of samples”. The adjustment method is determined (step S102).
[0179]
Here, “with sample number division” means, for example, that the product system is a coffee beverage system C, the existing product type is “black coffee”, and the sample number is “5”. In this case, this means that the sample number “5” is divided into divisions (for example, 3 and 2). And, for example, when “milk coffee” is added as another product of the same series (however, this is decided in the later product determination process, here “provisional” story), for example, this “milk coffee” If the sample number “2” is assigned, the sample number of “black coffee” is “3”. Usually, since one column is allocated to one sample, the maximum stock number of “black coffee” is reduced. In this way, the adjustment method is, for example, when “Black Coffee” currently on sale does not need to be completely replaced with other products, but it is better to reduce the handling quantity and replace it with other products. This is the adjustment method.
[0180]
It should be noted that the number of samples to be divided into what number may be set in advance, or may be designated by an analyst or the like each time.
On the other hand, if there is a product system that satisfies SR ≧ Gr average (NO in step S100) and satisfies the condition shown in step S107 (same as step S101) (YES in step S107), the product system is adjusted. The product type is determined and the adjustment method is determined. In this case, it is considered that “the merchandise system has a high sales share, but there are few types of merchandise”, and therefore the adjustment method is “add other related products” (step S108).
[0181]
If there is a product system that satisfies SR <Gr average (step S100, YES) and satisfies the condition (MR >> VR >> SR) shown in step S103 (step S103, YES), the product system Is determined as an adjustment target product system, and the adjustment method is determined. In this case, it is thought that “there are many unpopular products in the product system, or the products in the product system do not meet the taste of the user group of the location”. “Add other product” or “Replace with other related products” (step S104).
[0182]
Similarly, if SR <Gr average (step S100, YES) and there is a product system that satisfies the condition shown in step S105 (step S105, YES), the method of adjusting “Replace with other related products” is adjusted. (Step S106). If there is a product system that satisfies SR ≧ Gr average (step S100, NO) and satisfies the condition shown in step S109 (step S109, YES), “Add another product system product by dividing the number of samples. "Is an adjustment method (step S110). If SR ≧ Gr average (step S100, NO), and there is a product system that satisfies the condition shown in step S111 (step S111, YES), “other product system products shall be added, “Reduce product type for this purpose” is set as the adjustment method (step S112).
[0183]
Furthermore, when SR <Gr average (step S100, YES), and when there is a product system in which SR is somewhat smaller than the Gr average (step S201, YES) (in the example of step S201 illustrated, “SR <1/2 Gr average ”is required, but the coefficient by which such Gr average is multiplied is not limited to 1/2, but any value (but naturally less than 1) may be used) The system is determined to be the adjustment target product system, and the adjustment method is determined. In this case, “add other similar products” is set (step S202).
[0184]
Although not particularly shown in FIG. 34, if the determination in steps S101, S103, S105, S107, S109, and S111 is “NO”, other conditions are determined and all the conditions are determined. If the product condition is “NO” (for all of the product systems determined to be SR <Gr average (step S100, YES), for example, all the conditions of steps S101, S103, and S105), the product system is subject to adjustment. Judge that it is not a product type.
[0185]
As described above, a product system that is not balanced is determined as a product system to be adjusted, and the reason why the product system is unbalanced (a product system with insufficient varieties, a product system with a lot of sample waste, a product system that cannot be sold, etc.) Thus, the adjustment method can be determined in accordance with this, and the appropriate adjustment target product system and adjustment method can be determined for each location.
[0186]
On the other hand, the following processing is performed for the merchandise item that is YES (SR = 0) in the determination in step S99.
Here, since any product system that cannot be sold at all (SR = 0) is almost unthinkable, SR = 0 means that that product system is not handled at all in that location. (It is not sold), but the response is completely different depending on whether it is not popular because it is unpopular at that location, or whether it is not handled despite being highly likely to sell if sold Become.
[0187]
Therefore, the information of the correlation matrix (intra-group C17 or / and the entire range C19) obtained by the inter-product correlation analysis unit 7 is acquired (step S113), and the product system is within the group or It is determined whether or not there is a (strong) positive correlation with any one or more of the merchandise systems that are the selling merchandise system in all locations (step S114). As described above, since a product system having a positive correlation with a product system that is selling (main product system) should sell, if it is determined that there is a positive correlation (step S114, YES) ), This product line should sell (highly likely to sell), but it was determined that it was not sold at that location (the necessary product was not placed). The addition is defined as an adjustment method (step S115). If not (NO in step S114), it is determined that the cause is simply unpopular in the location, and is left as it is.
[0188]
Further, the number of adjustments is determined for each adjustment target product system and the adjustment method determined by the above-described processing (step S116). For example, when it is determined that “add other similar products”, the number of product types or the handling quantity of each product type (for example, in units of columns such as 1 column or 2 columns) is determined.
[0189]
When it is determined that a certain product system is an adjustment target product system by the above processing, all the adjustment target product systems and adjustment methods (addition / removal of products, A process of adding the product system name and its adjustment method to the target product system / adjustment method list, which is a list in which the replacement and sample redistribution are combined (step S117).
[0190]
The above-described processing is performed until processing for all product systems is completed for a certain adjustment target location (step S118, YES), and then, when there is an unprocessed adjustment target location (step S119, NO). The same processing is performed for the next adjustment target location.
[0191]
When processing for all the adjustment target locations is completed (step S119, YES), a completed product system / adjustment method list (adjustment target product system for each adjustment target location, its adjustment method, and number of adjustments) is displayed. Output (notify the adjusted product determination processing unit 12) (step S120).
[0192]
The adjusted product determination processing unit 12 analyzes the output of step S120, the location constraint analysis unit 6, the location / product sales cross ABC analysis unit 8 for analyzing location / product sales sources, and / or the location / product revenue sources. / Adjusted products are determined based on the processing results by the product gross profit cross ABC analyzing unit 14 and the product sales fluctuation time series analyzing unit 9.
[0193]
First, the process performed by the location constraint analysis unit 6 will be described with reference to FIG.
First, in the location setting information stored in the database, whether each location (and for each vending machine belonging to the location) is a vending machine that exclusively handles products of a specific manufacturer. Information indicating whether or not (in this example, this is regarded as a kind of “constraint” and is referred to as constraint presence / absence information) is included.
[0194]
The location constraint analysis unit 6 acquires the constraint presence / absence information included in the location setting information A1 (step S122), and performs processing for all locations (step S125). It is determined whether or not there is the “restriction” (step S123), and the restricted location list information C11 is created. That is, for a location with “constraint” (step S123, YES), the location ID of the location and constraint information (information indicating which specific manufacturer's product vending machine is dedicated) are sequentially added to the list B1. (Step S124), the list B1 is completed when the above processing for all locations is completed (YES in Step S125).
[0195]
Further, for restrictions, it may be checked whether all the vending machines in the installation place or a part of the vending machines.
Next, referring to FIG. 37, the processing procedure of the location / product sales cross ABC analysis unit 8 that analyzes location / product sales sources, and the location / product gross profit cross ABC analysis unit 14 that analyzes location / product revenue sources. explain. The cross ABC analysis units 8 and 14 have the same basic processing flow, and will be described with reference to FIG.
[0196]
In general, the processing is performed using a cross ABC analysis technique (a known technique). Here, to explain briefly, the cross ABC analysis method is a method in which the already described ABC analysis method and the cross analysis method are combined. The cross analysis method is a method of tabulating and analyzing data by crossing the breakdown of results according to factors, rather than simply summing up the results when the data is aggregated or represented in a table (of course, well-known methods ). For example, it is possible to create a table (cross tabulation table) in which the aggregated data is arranged in the vertical axis direction / horizontal axis direction according to factors, and to grasp the overall tendency and the nature of the problem. A cross ABC analysis is an application of ABC analysis (A, B, C rank for each factor) in the vertical axis / horizontal axis direction.
[0197]
In this example, location and product are used as the above “by factor”, for example, location is in the horizontal axis direction and product (product name) is in the vertical axis direction, and analysis is performed by sales (or gross profit).
First, the location / product sales cross ABC analysis unit 8 will be described.
[0198]
First, the product ID of the product to be processed, the analysis range (a range such as the entire range, only the group GrL1, etc.), and analysis period data are acquired (step S127). This may be specified by an analyst or the like each time.
[0199]
Next, the sales result information A4 is acquired, and the sales for each location within the analysis range and the cumulative total (total sales) are calculated for the product to be processed. Furthermore, calculate the value obtained by dividing the total sales by the number of locations handled (the number of locations selling the product), and use this as the average sales of the product (ie, even if the total sales are large, it is the number of locations handled) This may be because there is a large number (of course, the reverse is also possible), so it is necessary to use the average sales as an index to determine whether or not the product is a selling product).
[0200]
The above-described processing is performed for all products to be processed. Then, the products are sorted in the order of higher average sales. Further, for each location, the total sales of all the products handled in the location are obtained, and the locations are sorted in the order of the highest sales.
[0201]
Further, using the above-described ABC analysis method, the cumulative occupation ratio is obtained in the order of higher average sales, and ABC ranking is performed. Similarly, the cumulative occupancy ratio is obtained in the order of the total sales and ABC ranking is performed.
[0202]
Thus, for example, a location / product matrix table as shown in FIG. 38 is created (step S128).
Then, the location / product matrix table and the ABC analysis result (location / product sales cross ABC matrix B7) are output / displayed (step S129).
[0203]
As a result, the main location / main product (determined from the sales source) within the specified range is clarified and used for the adjustment product search and the addition / removal product determination processing in the subsequent adjustment product determination processing unit 12.
[0204]
Next, the location / product gross profit cross ABC analysis unit 14 will be described.
First, the product ID of the product to be processed, the analysis range (a range such as the entire range, only the group GrL1, etc.), and analysis period data are acquired (step S127). This may be specified by an analyst or the like each time.
[0205]
Next, the product cost and selling price information A6 are acquired, and the gross profit (revenue / profit) for each location within the scope of analysis and the cumulative total (total gross profit (revenue / profit)) Is calculated. Further, a value obtained by dividing the gross profit (revenue / profit) by the number of locations handled (the number of locations selling the product) is calculated as the average gross profit (revenue / profit) of the product.
[0206]
The above-described processing is performed for all products to be processed. Then, the products are sorted in the order of higher average gross profit (revenue / profit). Further, for each location, the total of the gross profit (revenue / profit) of all products handled in the location is obtained, and the locations are sorted in the order of the highest gross profit (revenue / profit).
[0207]
Further, using the above-described ABC analysis method, the cumulative occupancy is calculated in the order of higher average gross profit (revenue / profit), and ABC ranking is performed. Similarly, the cumulative occupancy is calculated in the order of the gross profit (revenue / profit) total, and the ABC ranking is performed.
[0208]
Thus, for example, a location / product matrix table as shown in FIG. 38 is created (step S128).
Then, the location / product matrix table and the ABC analysis result (location / product gross profit ABC matrix) are output / displayed (step S129). Although not specifically shown, these can be obtained by displaying the ABC analysis result for each location and product in FIG.
[0209]
As a result, the location / product (determined from the profit source) within the specified range is clarified and used for the adjustment product determination process and the addition / removal product determination process in the adjusted product determination processing unit 12 later.
[0210]
FIG. 39 is a diagram for explaining a processing procedure in the merchandise sales fluctuation time series analysis unit 9.
First, the product ID, the analysis range (the entire range, a range such as only the group GrL1), and the analysis period data of the product to be processed are acquired (step S131). This may be specified by an analyst or the like.
[0211]
From this, the sales range during the analysis range and the analysis period of the product to be processed is calculated, and the change rate (variation rate) is calculated based on this (step S132). The sales aggregation unit may be anything, but for example, as shown in FIG. 40, aggregation is performed in units of “month”, and the rate of change in which the sales in each month are compared with the sales in the previous month is calculated.
[0212]
Then, the product sales and change rate results (product sales / change rate table C23) obtained in step S132 are displayed as shown in FIG. 40, for example. If necessary, a bar graph (or a line graph or the like) as shown in FIG. 41 is displayed (step S133).
[0213]
The above processing is repeatedly executed for, for example, all the products handled at the location within the analysis range or any (a plurality of) products specified by the user.
With the above processing, for example, the analyst can grasp the stage in which each product is in a growth period, a maturity period (stable period), a decline period (for a specific place), etc., by referring to the display content. This is utilized in the processing of the adjusted product determination processing unit 12 later. In other words, when adding or replacing products, products that are in the growth period and maturity (stable period) are regarded as additional products or replacement products, and products that are in decline are excluded from additional products or replacement product candidates. Can add / replace.
[0214]
FIG. 42 is a flowchart for explaining the processing procedure in the adjusted product determination processing unit 12.
First, data of the product system / adjustment method list C22 (adjustment product system by adjustment target location, its adjustment method, and number of adjustments) obtained by the process of the adjusted product system determination processing unit 11 is acquired. . Further, the location / product ABC matrix table (by sales or profit) obtained by the location / product sales cross ABC analysis unit 8 or the location / product gross profit cross ABC analysis unit 14 is acquired (step S136). .
[0215]
Then, although not shown in the figure, for example, in the order from the product system / adjustment method list, the processing of steps S137 to S155 is executed with a certain adjustment target location and the adjustment target product system as processing targets. .
[0216]
First, regarding the adjustment target product system having a certain adjustment target location that is the processing target, the adjustment method of the adjustment target product system is determined by the processing of step S137 and step S150 (the adjustment method is the product system / adjustment method). On the method list). When the adjustment method is product replacement, product removal processing and product addition processing are sequentially performed.
[0217]
First, the process when the adjustment method is product addition (step S137, YES) will be described.
In this case, first, data of the list B1 obtained by the location constraint analysis unit 6 is acquired (step S138), and it is determined whether or not there is a “constraint” in the adjustment target location that is the current processing target. (Step S139).
[0218]
If there is a “restriction” (when only a product of a specific manufacturer is handled) (step S139, YES), the product belonging to the product system to be processed from the top of the location / product ABC matrix table Then, search for a product of a specific manufacturer that is not placed in the location to be adjusted, and the necessary additional product candidate (for example, if the number of adjustments is '2', the top and second products) Is extracted (step S141).
[0219]
Next, using the product sales / change rate table C23 obtained by the product sales fluctuation time series analysis unit 9, it is determined whether or not the extracted additional product candidate is a product in decline (a product to be excluded). Determination is made (step S142). For example, if “sales change rate is negative and the absolute value is large”, it is determined that the product is in a decline period and is excluded from the additional product candidates. At the same time, the number of additional product candidates does not satisfy the adjustment number (eg, “2”), so the process returns to step S140 (step S142, YES) to search for the next additional product candidate.
[0220]
If the determination in step S142 is NO, “adjusted product information” (the additional product candidate) is created (step S143), and after confirming again whether the number of adjustments is satisfied (step S144), the above “Adjusted product (additional) information” is put together in a list, and this is notified to the adjustment method presentation (display) processing unit 13. In addition, a list may be displayed at this stage so as to be confirmed by an analyst or the like, and corrected if necessary (step S155).
[0221]
On the other hand, when there is no “restriction” (step S139, NO), the product belongs to the product system that is the object of processing from the top of the location / product ABC matrix table and is not placed in the adjustment target location. A product is searched and necessary additional product candidates are extracted (step S145).
[0222]
Next, using the product sales / change rate table C23 obtained by the product sales fluctuation time series analysis unit 9, it is determined whether or not the extracted additional product candidate is a product in decline (a product to be excluded). Determination is made (step S146). For example, if “sales change rate is negative and the absolute value is large”, it is determined that the product is in a decline period and is excluded from the additional product candidates. At the same time, since the number of additional product candidates does not satisfy the adjustment number, the process returns to step S145 (step S147, YES), and the next additional product candidate is searched.
[0223]
If the determination in step S147 is NO, “adjusted product information” (the additional product candidate) is created (step S148), and after confirming again whether the number of adjustments is satisfied (step S149), the above “Adjusted product (additional) information” is put together in a list, and this is notified to the adjustment method presentation (display) processing unit 13. In addition, a list may be displayed at this stage so as to be confirmed by an analyst or the like, and corrected if necessary (step S155).
[0224]
On the other hand, if the determination in step S137 is NO, it is next determined whether or not it is “removal of product” (step S150). If it is not “removal of product” (step S150, NO), it is assumed that it is “re-segmentation of the number of product sample cans”, and the information on the re-division of the number of product sample cans is the above-mentioned product system / adjustment method Obtained from the list (step S154), an analyst etc. refers to the sample number information of each product of the current system, determines a product to be subdivided (usually a product with a large number of samples), and presents an adjustment method ( Display) Notify the processing unit 13. In addition, it may be displayed at this stage, checked by an analyst or the like, and corrected if necessary (step S155).
[0225]
In the case of “removal of product” (step S150, YES), the product of the target product system is searched from the lower level of the product placed in the target vending machine (location) column in the location / product ABC matrix table and removed. Candidates are found (step S151).
[0226]
Then, when the necessary number of product candidates to be removed are extracted (YES in step S152), information on the adjusted product (removal) is created (step S153), and the adjustment method presentation (display) processing unit 13 in step S155. Notify and display.
[0227]
The adjustment method presentation (display) processing unit 13 adds the product addition / removal list notified as described above (adjusted products and adjustment methods (addition, removal, replacement, sample redistribution) for each adjustment target location). Information) is output to a display screen, a printer, etc. as a proposal for an adjustment method for the adjustment target vending machine.
[0228]
In addition, using the product sales / change rate table obtained by the product sales fluctuation time series analysis unit 9, not only the products in the decline period are excluded as described above, but also in the growth period and maturity period (stable period). The product may be an additional product or a replacement product candidate.
[0229]
For example, as shown in FIG. 43, the adjustment method measure output format includes an adjustment method mark (addition, removal, replacement, sample reproduction) in the field corresponding to the target location-target product in the location / product matrix table. (Of course, it may be another independent table or the like on which information on adjustment location, adjustment product, and adjustment method (addition, removal, replacement, sample redistribution) of information is placed).
[0230]
In FIG. 43, in the location / product ABC matrix table, in the field corresponding to the target location-target product, as an adjustment method mark, 1) in the case of adding a product, the product to be added in the case of replacement. 2) / X indicates a product to be deleted, X indicates a downward arrow (↓) if the sample is reduced, and X indicates that the product is deleted.
[0231]
For example, location No. Taking the location of = n as an example, it can be seen that “product 38” is an additional product in the adjustment target product system C, and “product 34” is an additional product in the adjustment target product system W. It can also be seen that “product 20” and “product 5” are interchanged, and product 5 is removed and “product 20” is added. “Product 17” is removed.
[0232]
Next, another embodiment relating to the adjusted product / adjusted product determination process will be described below. The processing according to the other embodiment is basically performed in place of the processing by the adjusted product system determination processing unit 11 and the adjusted product determination processing unit 12 in FIG.
[0233]
In another embodiment, the purpose is to increase the number of product types as much as possible for each adjustment target location obtained by the adjustment target location extraction processing unit 10-1, and first, the column of existing products is reduced. After finding the maximum value of the column that can be vacated, a new product is added (replaced) up to this maximum value. Further, in the other embodiment, each product (or column) is handled by distinguishing between HOT (warm) and COLD (cold). In other words, the products sold at the vending machine include those sold at HOT (hot) and those sold at COLD (cold), and both may be sold together in one vending machine ( There are columns for heating and cooling the products). And even if it is the same product, the sales situation differs depending on whether it is HOT (warm) or COLD (cold). By treating both of them as separate items, it is possible to replace the product according to the actual situation. Become so.
[0234]
FIG. 44 is a diagram schematically showing the flow of the entire adjusted product determination process according to another embodiment.
The adjustment product determination process for each location according to this example (product type change content determination process) (step S160) is, as described above, for each adjustment target location obtained by the adjustment target location extraction processing unit 10-1. In general, "Product selection process that can reduce the number of columns in the product (existing product) currently placed in the location" (step S161) and "Currently placed in the location And “selection and allocation process of new product to be placed” (step S164).
[0235]
The processing in step S161 includes “processing for extracting product types that can reduce the number of columns without deleting products among existing products by HOT / COLD (C1, H1 extraction)” (step S162), and “from existing products The process includes a process of extracting the product type to be deleted by HOT / COLD (C2, H2 extraction) ”(step S163).
[0236]
The process of step S164 includes "a process for extracting a product type that should increase the product type (variety)" (step S165) and "a process for extracting a product type candidate to be increased in the extracted product type (C3, H3 extraction)". (Step S166), “Process of assigning“ Increase product type (product to be newly added) ”to“ Decrease column ”” (Step S167).
[0237]
Hereinafter, each process of steps S162, S163, S165, S166, and S167 will be described in detail.
However, before that, data used in each of these processes will be described.
[0238]
The data used for the processing in this example is basically the product composition balance information B6 and the location / product sales ABC matrix B7 (for example, the location / product matrix table shown in FIG. 38) in FIG. , HOT / COLD is distinguished from each other, and it is assumed that these data are created by distinguishing HOT / COLD. An example of the location / product matrix table used in this example is shown in FIG.
[0239]
The location / product matrix table 30 shown in FIG. 45 is basically the same format as the location / product matrix table shown in FIG. 38, except that product types are distinguished and handled by HOT / COLD (manufacturer, Costs etc. are not shown because they are not needed in this process). The product name 31 is a product name of each product type, but HOT and COLD are handled separately (for example, “coffee 4 (H)” and “coffee 4 (C)”). For each product name 31, the product system 32 to which the product belongs, HOT / COLD distinction (H / C33), sales 34 in each location, sales summary 35 in each group, installation location in each group The average 36, the total sales summary 37, and the average 38 in the entire installation location are stored.
[0240]
The average 36 in the location setting in each group is the total of sales 34 in each location by product (sales total 35 in each group) divided by the number of locations currently selling that product. It is. The same applies to the average 38 of the installation locations as a whole, except that not all groups but all.
[0241]
Note that the “total sales by product” in the location / product matrix table shown in FIG. 38 corresponds to the sales summary 35 in each group or the overall sales summary 37 (whichever is appropriate). “Another average sales” corresponds to an average 36 in the installation location in each group or an average 38 in the installation location in the whole (whichever may be used). As described above, FIG. 45 shows the location / product matrix table 30 for sales as an example. However, the present invention is not limited to such an example, and for example, a location / product matrix table for gross profit and gross profit rate is also conceivable.
In the following explanation, the average 36 in the installation location in each group or the average 38 in the installation location in the entire group is referred to as “average sales in the installation location (or average gross profit or average gross profit rate)”. To do.
[0242]
Hereinafter, first, a process of extracting product types that can reduce the number of columns without deleting products among existing products by HOT / COLD in step S162 will be described with reference to FIG.
[0243]
In the process of step S161, basically, the number of columns of a product having a large number of columns for sales (or gross profit or gross profit rate) is reduced.
[0244]
In FIG. 46, first, replenishment index efficiency F1 (product) and F2 (product) are calculated for each existing product of the adjustment target location to be processed. F1 (product) is the replenishment index efficiency at the current number of columns of the product, F2 (product) is the replenishment index efficiency when the number of columns of the product is reduced by 1, and the calculation formula is as follows: is there.
[0245]
[Equation 3]

[0246]
Note that F1 (product) and F2 (product) mean values of replenishment index efficiencies F1 and F2 calculated for a certain “product”. For example, when there is a product “coffee 1”, the values of the replenishment index efficiencies F1 and F2 calculated for “coffee 1” are expressed in the form of F1 (coffee 1) and F2 (coffee 1).
[0247]
Then, F1 (product) having the maximum value is obtained from each F1 (product) calculated for each existing product, and this is set as the replenishment efficiency standard F0 (F0 = max (F1 (product))). Note that a product with max (F1 (product)) is a product with a small number of columns for sales (or gross profit or gross profit rate).
[0248]
(End of step S171)
Next, in each F2 (product) calculated for each existing product, F2 (product) that takes the minimum value is obtained (represented as minF2 (product)), and this is compared with the replenishment efficiency standard F0, and minF2 If the condition of (product) <F0 is satisfied, the number of product columns of this minF2 (product) is reduced by 1 (step S172). In other words, if the replenishment index efficiency is smaller than F0 even if the current number of columns is reduced by 1, the product is considered to have too many columns for the number of sales. It is determined that the number of columns can be reduced by one.
[0249]
If the condition of minF2 (product) <F0 is satisfied in step S172, F2 (product) is recalculated for the product with the reduced number of columns (step S173) (that is, the current status (still If we consider that the number of columns is not reduced), in effect, "F2 (product) = (number of sales of that product) / (number of columns of that product-2)" will be calculated), again, step The process returns to S172 (step S174, NO). In the process of step S172 performed again, the product F2 (product) with the reduced number of columns may become minF2 (product) again. In other words, there is a possibility of reducing two or more columns for a certain product.
[0250]
On the other hand, if the condition of F2 (product) <F0 is not satisfied in step S172, it is determined that there is no product that can reduce the number of columns (step S174, YES), and the process proceeds to step S175. Totaling by / COLD, the process ends. The “column that can be reduced without deleting products” obtained by the above processing is indicated by “H1” for HOT and “C1” for COLD in the figure.
[0251]
Next, the “process for extracting product types to be deleted from existing products by HOT / COLD” in step S163 will be described with reference to FIG.
Here, based on the number of columns reduced by the process of step S162, candidates for products to be deleted are obtained for each HOT / COLD by the process of step S181 below. That is, as a result of reducing the number of columns by the process of step S162, a product that has one column is also treated as “a product having one column” in the following step S181. As a result, for example, a product that originally had three columns may be reduced to one column, and a product may be deleted.
[0252]
In the process of step S181, for each location, the product (or the gross margin or the gross margin) is small (for example, about several percent to several tens percent) and the number of columns is one column. Even if the product is deleted, the product product to which the product belongs does not disappear (there are other products of the same product system) as candidates for the product to be deleted (these are “H2” in the figure, COLD Is indicated by “C2”). After calculating “columns that can be reduced without deleting products” C1 and H1, as described above, “columns to be reduced by deleting products” C2 and H2 are calculated without first deleting products as much as possible. This is because the columns C1 and H1 are created, and if more empty columns are necessary, the product is deleted and the empty columns C2 and H2 are used. By doing so, it is convenient to perform an operation of installing a new product and increasing the product type (variety) without deleting the product as much as possible.
[0253]
Next, with respect to “processing for extracting product types for which product types (variety) should be increased” in step S165 and “processing for extracting product type candidates to be increased in extracted product types” in step S166, FIG. A description will be given with reference to a) and (b). The processes of FIGS. 48A and 48B basically extract products that can be expected to sell (or increase profits) if placed in the location.
[0254]
FIG. 48A is a detailed flowchart of the “processing for extracting a product system whose product type (variety) should be increased” in step S165.
In the figure, for each location, the following processing is repeated until the check is completed for all product systems (step S194).
[0255]
First, it is determined whether or not the product system is a product system whose Gr average is within the top three (step S191). If so (step S191, YES), then the product system When the sales occupation ratio SR at the location is compared with the Gr average and the condition of “SR <Gr average” is satisfied (step S192, YES), the product system is registered in the list (step S193). . In other words, even if the Gr average is in the top three product categories, if the sales share at that location is greater than or equal to the Gr average (SR ≧ Gr average), sales will increase even if the product type is increased further. Since it cannot be said that there is a high possibility of growth, it is not registered in the list.The determination method in step S192 is an example, and is not limited to this, and may be another method. Instead of selecting “Top 3 product series”, “Product series occupying several percent to several tens of percent of Gr average” may be used. Since it is a product system occupying several tens of percent, it is a product system of A rank when ABC analysis is performed on the product system.
[0256]
For example, in the example of FIG. 28, the product systems C, N, and W are product systems with the Gr average within the top three. However, in the following explanation, a specific product name is not shown, and a product system whose Gr average is in the top three is temporarily the product system X, the second is the product system Y, and the third is the product Description will be made assuming that the system Z is the system Z (assuming that the product systems X, Y, and Z in the list 40 shown in FIG. 48 are registered by the above processing).
[0257]
In addition, the example shown in the list 40 shows a case where all the product systems in the top three satisfy the condition of “SR <Gr average” in step S192. For example, the product system Y does not satisfy the condition. In this case, only the product systems X and Z are registered.
[0258]
FIG. 48B is a detailed flowchart of the “process for extracting product type candidates to be increased in the extracted product system” in step S166.
The process shown in FIG. 48 (b) is performed for each location, and will be described below for each location, for each product system registered in the list, using, for example, the location / product matrix table 30. Process.
[0259]
First, it is determined whether there is a product that is not placed in the location (hereinafter referred to as a processing target location) for a certain product type (hereinafter referred to as a processing target location) registered in the list (step S195). ), In some cases (step S195, YES), among the products of the merchandise, products having a high “average sales (or average gross profit or average gross profit rate) at the installation location” are respectively HOT / COLD. Select (step S196). Then, the selected product type is registered in the list 40 (step S197).
[0260]
For example, in the example shown in the location / product matrix table 30, the above processing is assumed that the processing target location is No. 1 (location ID = 1015) and the processing target product type is the product type C. Then, in the location / product matrix table 30, the product indicated by a blank is a product that is not placed in the processing target location. In this example, since the product name 31 is “Coffee 4 (C)”, “Coffee 6 (C)”, “Coffee 2 (H)”, etc., the determination in Step S195 is YES. Subsequently, among these products, a product having a high “average sales (or average gross profit or average gross profit margin) at the installation location” is selected. This is because the location / product matrix table 30 is already “average at the installation location”. “Sales (or average gross margin or average gross margin)” is sorted in descending order, so searching from the top selects “Coffee 4 (C)” for COLD and “Coffee 2 (H)” for HOT And registered in the list 40.
[0261]
Furthermore, even if it is added to the condition that “the average sales at the installation location (or average gross profit or average gross profit rate)” is within the top few percent to several tens of percent (for example, rank A in FIG. 38). Good. In this case, for example, “Coffee 4 (C)” may be selected for COLD, but nothing may be selected for HOT. In such a case, the corresponding part in the list 40 is blank.
[0262]
When the above processing is performed for all the products registered in the list 40 (step S198, YES), the processing ends, and the “list of products for which columns are to be increased” (that is, the list 40) is completed. (Step S199).
[0263]
In the example shown in the figure, it is assumed that the product selected / registered with respect to the product system X has COLD “XC” and HOT “XH”. Similarly, it is assumed that “YC”, “YH”, “ZC”, and “ZH” are selected / registered for the product systems Y and Z, respectively. The list 40 is indicated by “H3” for HOT and “C3” for COLD in FIG.
[0264]
Finally, a process of assigning the products (C3, H3) registered in the list 40 to C1, H1, C2, H2 in this way is performed (the process of step S167). However, (C1, H1, C2, H2) means the maximum value of the number of columns that can be vacated at that location, and new products can be freely added up to this maximum value. However, it is not always necessary to add new products up to the maximum value. For example, as described below, products may be added using only the empty columns C1 and H1 (resulting in new products). Columns that have not been added will remain unchanged.)
[0265]
Two methods are proposed for this process.
First, the first method will be described with reference to FIG.
FIG. 49A schematically shows how products registered in the list 40, that is, products to be newly added are allocated to C1 / H1 (empty column due to column reduction) and C2 / H2 (empty column due to product reduction). FIG.
[0266]
FIG. 49 (b) shows how the products registered in the list 40 are sorted into C1 / H1 and C2 / H2 after sorting them in descending order of “average sales at installation location” by HOT / COLD. FIG. That is, in the list 40, the products are arranged in the order of the highest average sales occupancy rate (in the order of X, Y, Z), but when viewed for each product, the products of the product system X (XC, XH) are not necessarily included. This is because the “average sales at the installation location” is not always the highest (in the figure, it is the order of the merchandise system, but it is not limited to this).
[0267]
49A and 49B, the processing contents are the same. That is, if C1 and C2 are combined and the number of empty columns is three or more, XC, YC and ZC are all assigned to each column (which column can be assigned freely). Similarly, if the number of empty columns is 3 or more in total including H1 and H2, XH, YH and ZH are all assigned to each column. In the list 40, as described above, the product types are not necessarily registered for all the product types X, Y, and Z. For example, when only two product types are registered, the above “ “If the number of empty columns is 3 or more” changes to “If the number of empty columns is 2 or more”.
[0268]
On the other hand, if the number of empty columns is not three or more, the empty columns are assigned in order from the highest product type according to the sorted order. Product types that have not been assigned will not be added.
[0269]
Here, allocation to C1 and H1 only reduces the number of columns of existing products and does not delete existing products, but allocation to C2 and H2 deletes existing products. Therefore, it would be meaningless if the “average sales at the installation location” of newly added products is less than the products to be deleted (it will bother inserting products with low sales). Therefore, in such a case, allocation to C2 and H2 may not be performed (the product type is not added).
[0270]
Alternatively, in the first place, the purpose of the other embodiment is to increase the variety of product types, and the assignment to C2 and H2 only changes the product type, and does not increase the product type. , H2 may be extracted (of course, allocation), but may not be performed (only allocation to C1 and H1 is performed).
[0271]
Next, the second method will be described below with reference to FIG.
First, in the first method, as shown in FIG. 49 (b), the “average sales at the installation location” is sorted in descending order by HOT / COLD, but in the second method, FIG. 50 (a) As shown in FIG. 6, the product list 50 sorted in descending order of “average sales at the installation location” is created without distinguishing HOT / COLD. In the example of FIG. 50 (a), it is assumed that sorting is performed in the order of XC, YC, XH, YH, ZH, and ZC in descending order of “average sales at installation location”.
[0272]
Using this product list 50, allocation to an empty column is performed by the processing shown in FIG.
The processing shown in FIG. 50B is performed on the processing targets in order from the product with the highest ranking. First, it is determined whether or not the product to be processed can be assigned to C1 or H1, and if it cannot be further assigned, it is determined whether or not it can be assigned to C2 or H2 (step S201). At first, since the product with the highest ranking is targeted, in this example, it is determined whether or not the product XC can be assigned to C1, and if it cannot be assigned to C1, it is further determined whether or not it can be assigned to C2. . If there is no empty column that can be assigned to either C1 or C2 (NO in step S201), the process of step S201 is performed for the next highest product.
[0273]
If it can be assigned to either C1 / C2 (step S201, YES), the processing target product is allocated to the empty column determined to be assignable (step S202), and the same product as the processing target product The product having the opposite cold / warm in the system is deleted from the product list 50 (step S203). For example, if the product XC can be assigned to either C1 / C2, the product XH is deleted from the product list 50. As a result, although the product XH is ranked third, it is not added to the location because it is excluded from the processing target. When the above process is performed for all the products in the product list 50 (step S204, YES) (however, the deleted product is not performed), the process is terminated.
[0274]
In this way, for example, even if all of C1, C2, H1, and H2 are combined and there are only three empty columns, it is possible to add new products one by one to all product systems X, Y, and Z It becomes.
[0275]
Even in the second method, as described above, even when allocation to C2 / H2 is possible, the “average sales at the installation location” of newly added products is smaller than the products to be deleted. The product may not be added. Similarly, C2 and H2 may not be extracted in the first place.
[0276]
It should be noted that handling by distinguishing between HOT and COLD is not limited to the other embodiments, and may be performed in the above-described embodiments of FIGS.
FIG. 51 is a diagram illustrating an example of a hardware configuration of an information processing apparatus that implements the above-described vending machine product configuration adjustment support system, and an example of a storage medium that stores a program that implements the various processes / functions described above.
[0277]
51, the information processing apparatus 20 includes a CPU 21, a storage unit 22 (including a portable storage medium 22a), a memory 23, a display unit 24, an operation unit 25, an input / output interface unit 26, a communication unit 27, and the like. .
[0278]
The CPU 21 is a central processing unit that controls the entire information processing apparatus 20.
The storage unit 22 is a storage device such as an HDD in which programs for realizing various processing functions of at least the above-described vending machine product configuration adjustment support system are stored. Alternatively, the storage unit 22 may be a combination of a portable storage medium 22a and its drive reading device (for example, a floppy disk (FD) and a flexible disk drive (FDD)). In addition to the FD, the portable storage medium 22a includes a CD-ROM, a memory card, a DVD, an MO, and the like.
[0279]
The memory 23 is a RAM or the like for temporarily storing the program stored in the storage unit 22 and causing the CPU 21 to execute the program.
The display unit 24 is a display or the like and displays, for example, data (table or the like) obtained by various processing functions of the above-described vending machine product configuration adjustment support system.
[0280]
The operation unit 25 is a keyboard, a mouse, or the like, and a user or the like uses the operation unit 25 to select / designate contents to be displayed on the display unit 24 or to change a part of the displayed contents. I do.
[0281]
The input / output interface unit 26 is an interface for connecting to a database or the like via, for example, any communication line and inputting the various information described above.
The communication unit 27 is configured to provide a function of communicating with other external information processing apparatuses via a network such as a public line network or a dedicated communication network, and is generally well known. It is.
[0282]
Here, the present invention is not limited to the device itself, such as the product configuration adjustment support system of the vending machine, but can be read by a computer in which programs for realizing various functions according to the present invention are stored when used by a computer. It can also be configured as a recording medium (storage medium) itself.
[0283]
In this case, the “recording medium” includes, for example, a portable storage medium such as the CD-ROM 28 and the floppy disk 29 shown as an example of the portable storage medium 22 a in FIG. 51 (not limited to these examples, MO, (Anything that falls within the category of “portable storage media” such as a DVD, a removable hard disk, etc.), as well as any external information processing device (not shown) that can communicate via a network. Recording media "are also included. Of course, a storage device (RAM / ROM, hard disk or the like) in the information processing apparatus 20 is also included.
[0284]
【The invention's effect】
As described above in detail, according to the vending machine product configuration adjustment support system of the present invention, more appropriate adjustment target vending machines can be selected and more appropriate product configuration can be adjusted. You can increase the total sales (or gross profit) of the products you sell.
[Brief description of the drawings]
FIG. 1 is a diagram illustrating a schematic functional configuration and a processing flow of an entire product configuration adjustment support system of a vending machine.
FIG. 2 is a flowchart for explaining a processing procedure in a location attribute feature analysis unit;
FIG. 3 is a diagram illustrating an example of classification from attributes (evaluation items; category names).
FIG. 4 is a diagram showing an example of evaluation results (five-step evaluation).
FIG. 5 is a diagram showing an example of factor loading obtained by factor analysis.
FIG. 6 is a factor loading bar graph obtained by converting FIG. 5 into a bar graph.
FIG. 7 is an example of a factor score.
8A to 8C are bar graphs of the factor scores shown in FIG.
FIG. 9 is a flowchart for explaining a processing procedure in a sales feature analysis unit.
10 is an example of the product classification information in FIG.
FIG. 11 is an example of a sales occupancy rate list for merchandise by location.
FIG. 12 is a diagram showing an example of factor loading obtained by factor analysis.
13 is a factor loading bar graph obtained by converting FIG. 12 into a bar graph.
FIG. 14 is an example of a factor score.
15A to 15C are bar graphs of the factor scores shown in FIG.
FIG. 16 is a flowchart for explaining a processing procedure in a location classification processing unit from attribute features / location classification processing unit in sales features;
FIG. 17 is a diagram illustrating an example of a classification result by a location classification processing unit based on attribute features;
FIG. 18 is a diagram illustrating an example of a classification result by a location classification processing unit for sales characteristics;
FIG. 19 is a flowchart for explaining a processing procedure of a specific location analysis unit;
FIG. 20 is a flowchart for explaining details of group association processing in step S29 of FIG. 19;
FIG. 21 is a flowchart for explaining a processing procedure by a location competition analysis unit;
FIG. 22 is a flowchart for explaining a processing procedure of a product composition balance analysis unit.
FIG. 23 is a detailed flowchart of the balance visualization process in step S56 of FIG.
FIG. 24 is a diagram showing an example of balance visualization.
FIG. 25 is a flowchart for explaining a processing procedure of a location cost / profit balance analysis unit;
FIG. 26 is a flowchart for explaining a processing procedure of a product cost / profit balance analysis unit.
FIG. 27 is a flowchart for explaining a processing procedure in an adjustment target location extraction processing unit;
FIG. 28 is a diagram showing an example of a bar graph showing an example of a group product average sales occupancy rate (sales pattern).
FIG. 29 is a diagram showing an example of a location / product sales occupancy rate list (main location / main product table of each group);
FIG. 30 is a flowchart for explaining a processing procedure in a correlation analysis unit between product systems.
FIGS. 31A and 31B are explanatory diagrams of product system correlation analysis. FIGS.
FIG. 32 is a diagram illustrating an example of a correlation matrix.
FIG. 33 is a flowchart (No. 1) for explaining the processing procedure in the adjusted product system determination processing unit.
FIG. 34 is a flowchart (No. 2) for explaining the processing procedure in the adjusted product system determination processing unit.
FIG. 35 is a flowchart (No. 3) for explaining the processing procedure in the adjusted product system determination processing unit.
FIG. 36 is a flowchart for explaining a processing procedure by a location constraint analysis unit;
FIG. 37 is a flowchart for explaining processing procedures of a location / product sales cross ABC analysis unit that analyzes product sales sources and a location / product gross profit cross ABC analysis unit that analyzes product revenue sources;
FIG. 38 is a diagram showing an example of a location / product matrix table;
FIG. 39 is a diagram for explaining a processing procedure in a product sales fluctuation time series analysis unit;
FIG. 40 is an example of a setting input / display screen for merchandise sales fluctuation time series analysis;
FIG. 41 is a diagram illustrating an example of displaying product sales / variation by a bar graph;
FIG. 42 is a flowchart for explaining a processing procedure in an adjusted product determination processing unit.
FIG. 43 is a diagram illustrating an example of an output format of an adjustment method measure plan (a location / product matrix table with an adjustment method mark);
FIG. 44 is a diagram schematically showing the flow of the entire adjusted product determination process according to another embodiment.
FIG. 45 is a diagram showing an example of a location / product matrix table used in another embodiment.
FIG. 46 is a detailed flowchart of step S162 in FIG. 44.
47 is a detailed flowchart of step S163 of FIG. 44. FIG.
48A is a detailed flowchart of step S165 of FIG. 44, and FIG. 48B is a detailed flowchart of step S166.
FIGS. 49A and 49B are diagrams for explaining a first method of processing for assigning a product to be newly added to an empty column. FIGS.
FIGS. 50A and 50B are diagrams for explaining a second method for allocating a product to be newly added to an empty column. FIGS.
FIG. 51 is a diagram illustrating an example of a hardware configuration of an information processing apparatus that implements a product configuration adjustment support system for a vending machine, and an example of a storage (recording) medium.
[Explanation of symbols]
1 Vending machine location (location) attribute feature analysis department
2. Sales feature analysis of products / products related to location
3-1 Location classification processing unit from attribute features
3-2 Location classification processing section by sales characteristics
4 Product (product) composition balance analysis department in vending machine
5 Location Competition Analysis Department
6 Location constraint analysis department
7 Correlation analysis section between products
8 Location / Product Sales Cross ABC Analysis Department
9 Product Sales Fluctuation Time Series Analysis Department
10-1 Location extraction processing unit for adjustment
10-2 Specific location analysis section
11 Adjustment product system decision processing section
12 Adjustment product decision processing section
13 Adjustment method presentation (display) processing unit
14 Location / Commodity Gross Cross ABC Analysis Department
15 Location Cost / Profit Balance Analysis Department
16 Product-related Cost / Profit Balance Analysis Department
20 Information processing device
21 CPU
22 Memory unit
22a Portable storage medium
23 memory
24 display
25 Operation unit
26 I / O interface section
27 Communication Department
28 CD-ROM
29 floppy disks
30 Location / Product Matrix
31 Product name
32 product series 32
33 H / C
34 Sales at each location 34
35 Total sales in each group
36 Average of installation locations in each group
37 Total sales summary
38 Average for all locations
40 lists
50 product list

Claims (24)

Attribute feature analysis means for performing factor analysis according to the attribute of each location and obtaining a factor score;
Clustering using the factor score obtained by the attribute feature analysis means, and location classification processing means from attribute features that classify locations into groups having common features in attributes,
For each group classified by the location classification processing means from the attribute feature, the sales pattern of the location belonging to the group is obtained, and the adjustment target location extraction processing means for obtaining the location to be adjusted for each group based on the sales pattern When,
Vending machine product composition adjustment support system characterized by comprising: Sales feature analysis means for performing factor analysis according to the sales feature of each location and obtaining a factor score;
Cluster analysis using the factor score obtained by the sales feature analysis means, and the location classification processing means with the sales feature for classifying the location into a group having a common feature in the sales feature;
Adjustment target location extraction processing means for obtaining a sales pattern of a location belonging to the group for each group classified by the location classification processing means for the sales feature, and obtaining a location to be adjusted for each group based on the sales pattern When,
Vending machine product composition adjustment support system characterized by comprising: Attribute feature analysis means for performing factor analysis according to the attribute of each location and obtaining a factor score;
Clustering using the factor score obtained by the attribute feature analysis means, and location classification processing means from attribute features that classify locations into groups having common features in attributes,
Sales feature analysis means for performing factor analysis according to the sales feature of each location and obtaining a factor score;
Cluster analysis using the factor score obtained by the sales feature analysis means, and the location classification processing means with the sales feature for classifying the location into a group having a common feature in the sales feature;
The group classified by the location classification processing means from the attribute feature is associated with the group classified by the location classification processing means in the sales feature, and the locations belonging to only one of the associated groups Specific location analysis means that uses a location other than the specific location as a non-specific location, location classification processing means for each group classified by the location classification processing means from the attribute features and the sales feature For each group categorized according to the above, the sales pattern of the location that is non-specific among the locations belonging to the group is obtained, and adjustment is made from among the non-specific location and / or specific location for each group based on the sales pattern Adjustment target location extraction processing means for obtaining the target location;
Vending machine product composition adjustment support system characterized by comprising: It said to each location, to analyze the balance of sales occupancy rate and sample occupancy rate and varieties occupancy rate of the products / goods system is dealing with the location, the location where the balance is not achieved, the adjustment target location of additional candidates The product / product system composition balance analysis means
The adjustment target location extraction processing unit selects / determines a location to be added as a location to be adjusted from among additional candidates obtained by the product / commodity configuration balance analysis unit, as necessary. The product configuration adjustment support system for a vending machine according to any one of claims 1 to 3. Calculate the gross margin for each location and the total gross margin for all locations, determine the unbalanced location based on the ratio between each gross rate and the gross gross rate, and add the unbalanced location to the location to be adjusted It further includes a cost / benefit balance analysis means for the location as a candidate,
The adjustment target location extraction processing unit selects / determines a location to be added as the location to be adjusted from the additional candidates obtained by the cost / benefit balance analysis unit of the location as necessary. The product configuration adjustment support system for a vending machine according to any one of claims 1 to 3. A location having an unbalanced product system is calculated based on a ratio of the gross profit rate for each location and the gross rate for each product system. Further comprising a product cost / benefit balance analysis means for making the adjustment target location as an additional candidate,
The adjustment target location extraction processing means selects / determines a location to be added as the adjustment target location from among additional candidates obtained by the cost / profit balance analysis means of the product system, as necessary. The product configuration adjustment support system for a vending machine according to any one of claims 1 to 3. A location conflict analysis means for determining the presence or absence of each location and the degree of competition;
The adjustment target location extraction processing unit excludes a location having a high degree of competition obtained by the location competition analysis unit from the location to be adjusted as necessary. Vending machine product composition adjustment support system. A location cost / benefit balance analysis means for calculating a gross margin rate for each location and a total gross rate for all locations, and obtaining an unbalance location based on a ratio between each gross rate and the total gross rate,
An adjustment target location extraction processing means that uses the unbalance location determined by the location cost / profit balance analysis means as the location to be adjusted;
Vending machine product composition adjustment support system characterized by comprising: Calculate the gross margin rate for each location and the gross margin rate for each product type for each location, and find locations that have an unbalanced product type for each location based on the ratio between the gross rate and the gross rate for each product type. The cost / benefit balance analysis means for the desired product system,
The product configuration adjustment of the vending machine, characterized by having adjustment target location extraction processing means that uses the location having the unbalanced product system, which is obtained by the product cost / profit balance analysis means, as the location to be adjusted Support system.   For each location to be adjusted, analyze the balance between the sales occupancy rate / sample occupancy rate / variety occupancy rate of the product system handled in that location, and products that meet one of the preset unbalance conditions The system is an adjustment target product system, and has adjustment product system determination processing means for determining an adjustment method for each adjustment target product system by an adjustment method determined in advance for each unbalance condition. Vending machine product composition adjustment support system.   For each location to be adjusted, calculate the gross margin of the location and the gross margin of each product system in that location, and based on the ratio between the gross margin of the location and the gross margin for each product system, And a product configuration adjustment support system for a vending machine, characterized by comprising adjustment product system determination processing means for determining an adjustment method for the unbalanced product system. Correspondence analysis means for analyzing the correlation between each product system in the group for each group in which each location is classified by attribute or sales feature,
When there is a product system that is not handled in the location to be adjusted, the adjusted product system determination processing unit determines a positive correlation with the product system that is not handled based on the analysis result by the correlation analysis unit. It is determined whether or not there is a selling product system in the product system that has, and if there is a selling product system, the product system that is not handled is handled at the location to be adjusted The product configuration adjustment support system for a vending machine according to claim 10 or 11 . The vending machine product configuration adjustment support system according to any one of claims 10 to 12 , wherein the location to be adjusted is determined according to any one of claims 1 to 9 . Based on the sales result information for each location / product, a sales product analysis means for finding a sales product,
When the adjustment method of the adjustment target product system is addition or replacement, an adjustment product determination processing means for determining a product to be added or replaced from the selling products of the same product system as the adjustment target product system;
The vending machine product configuration adjustment support system according to claim 10 , further comprising: Revenue source product analysis means for finding profitable products based on product cost / selling price information,
When the adjustment method of the adjustment target product system is addition or replacement, an adjusted product determination processing means for determining a product to be added or replaced from among the profitable products of the same product system as the adjustment target product system;
The vending machine product configuration adjustment support system according to claim 10 , further comprising: When the location vending machine to be adjusted is a vending machine that handles only products of a specific manufacturer, the adjusted product determination processing means determines the product to be added or replaced from the products of the specific manufacturer. 16. The vending machine product configuration adjustment support system according to claim 14 or 15 . The vending machine product configuration adjustment support system according to any one of claims 14 to 16 , wherein the adjusted product determination processing unit does not set a product in a decline period as the product to be added or replaced. Adjustment target location extraction processing means for obtaining a location to be adjusted;
Empty column extraction means for obtaining a column that can be vacated by reducing the number of columns of existing products for each location to be adjusted, based on the replenishment efficiency of each existing product;
An additional product determination means for determining a new product to be added;
Allocating means for allocating the additional product determined by the additional product determining means to the empty column obtained by the empty column extracting means;
Vending machine product composition adjustment support system characterized by comprising: 19. The vending machine product configuration adjustment support system according to claim 18, wherein the empty column extraction means further obtains a column that can be opened by deleting an existing product. The additional merchandise determining means determines a merchandise to be added from a merchandise system in which the average of sales or gross profit or gross margin is higher and is larger than the sales or gross profit or gross margin at the location to be adjusted. 20. The vending machine product configuration adjustment support system according to claim 18 or 19 . The vending machine product configuration adjustment support system according to any one of claims 18 to 20 , wherein the empty column extraction unit / additional product determination unit treats the column / product separately from hot and cold. The vending machine product configuration adjustment support system according to claim 21 , wherein the allocation unit performs allocation so that the same product is not added both hot and cold. Divide the location into multiple groups,
Within each group, select the location to be adjusted,
For each location to be adjusted, a balance analysis of product structure or cost / benefit is performed, and an unbalanced product system is adjusted according to the type of unbalance together with the adjustment target product system in the adjustment target location. Decide how,
A vending machine characterized in that among the products belonging to each of the adjustment target product systems, a product to be added or replaced adjustment product is determined from the selling or profitable products in the group to which the adjustment target location belongs. Product composition adjustment method. When used in a computer
The ability to divide a location into multiple groups,
Within each group, a function for selecting a location to be adjusted;
For each location to be adjusted, a balance analysis of product structure or cost / benefit is performed, and an unbalanced product system is adjusted according to the type of unbalance together with the adjustment target product system in the adjustment target location. The ability to decide how,
A function to determine a product to be added or replaced adjustment product among the products belonging to each of the adjustment target product systems from the selling or profitable products in the group to which the adjustment target location belongs;
The computer-readable recording medium storing a program for realizing the above.

Images