201118809 六、發明說明： 【發明所屬之技術領域】 本發明係關於現金補充計畫支援系統 支援方法，特別是適用於以從自動地進行 易的複數台ATM (自動櫃員機）中得到 礎’預測各ATM中分別所需的資金量而 計畫制定的系統的現金補充計畫支援系統 支援方法。 【先前技術】 近年來，自動地進行現金的進出款5 了銀行等金融機構以外，還被大量地設置 站內等各種場所。使用者將磁卡或1C卡条 能夠利用使用者自身的操作來進行現金的 存款（進款）等交易。這樣的交易被頻繁 會發生在交易途中無現金的情況，而需要 補充現金。 但是，現金的交易量（出款量和進 ATM的地域、店鋪或工作時域等環境而 何時補充多少數量的現金爲佳的最佳補充 ATM通常都不同。但是，習知技術下工 的經驗制定補充計畫，對於多台ATM正 同的最佳補充計畫是極其困難的。因此， 斷失誤而引起無現金的情況。 及現金補充計畫 現金的進出款交 的現金資訊爲基 支援現金的補充 及現金補充計畫 ζ易的 ATM，除 於便利商店或車 _入ATM之後， 提款（出款）或 地進行，爲了不 在 ATM適當地 款量）係依設置 不同。因此，在 計畫，對於每台 作人員依靠自身 確地判斷分別不 存在工作人員判 201118809 對此，提出了求出各ATM的進款累計量和出款累計 量，並根據該求出的進款累計量與出款累計量的差額來預 測出各ATM各自的所需資金量的技術（例如，參照專利 文獻1〜3 )»在這些專利文獻1~3中，也記載有按幣種計 算各ATM所需的資金量的內容。記載有根據該習知技術 能夠容易地進行複數台ATM整體的所需資金量的預測。 (先前技術文獻） (專利文獻） 專利文獻1 :日本公開公報特開平1 0-91 702號 專利文獻2 :日本公開公報特開平1 0-240843號 專利文獻3 :日本公開公報特開平1 0-24721 4號 【發明內容】 (發明所欲解決之課題） 但是，在上述專利文獻1〜3所記載的習知技術中，根 據進款累計量與出款累計量的差額來預測所需資金量，而 關於具體地藉由怎樣的方法由進款累計量與出款累計量的 差額來進行預測並沒有詳細的記載。也就是說，僅有根據 進款累計量的推移與出款累計量的推移來求出所需資金量 的記載，而藉由怎樣的計算來求出所需資金量則不清楚。 本發明是爲了解決這樣的問題而作成者，其目的在於 提供即使不依靠工作人員的經驗，也能夠容易地生成有關 複數台ATM各自的最佳現金補充計畫用的具體手段。 201118809 (解決課題之手段） 爲了解決上述課題，在本發明中，係 自動櫃員機（ATM )的進款和出款的交易 款與出款的差分的積算値，由該所求出的 求出近似線。接著，根據該近似線和自動 的現有餘款，來預測將來的某一時點的所 (發明之效果） 根據如上所述所構成的本發明，由依 的進款和出款的時間序列資訊中，現金按 進行增減的交易趨勢作爲近似線而求出， 勢中預測出將來的某一時點的ATM中的 此，即使不依靠工作人員的經驗，也能夠 來的何時對 ATM補充或回收多少程度的 現金補充計畫。 【實施方式】 以下’根據附圖對本發明的一實施形 1圖是表示第一實施形態的現金補充計畫 結構例的方塊圖。如第1圖所示，第一實 充計畫支援系統10透過網際網路等通信 台ATM100連接。另外，現金補充計畫支 備有交易資訊取得部11'差分積算部12 在每次取得有關 資訊時，求出進 複數個積算値來 櫃員機（ATM) 需資金量。 次從 ATM取得 多少程度的比率 並從該交易的趨 所需資金量。藉 容易地生成在將 現金爲佳的最佳 態進行說明。第 支援系統的功能 施形態的現金補 網路200與複數 :援系統1 〇係具 ί、近似線運算部 201118809 1 3、所需資金預測部1 4以及補充計畫輸出部1 5作爲其功 能結構而構成。 第1圖所示的現金補充計畫支援系統（1 〇 )的各功能 方塊亦可藉由硬體構成、DSP ( Digital Signal Processor 、數位訊號處理器）和軟體的任一者加以實現。例如，在 藉由軟體來實現時，本實施形態的現金補充計畫支援系統 10實際上具備有計算機的CPU (中央處理器）、或MPU (微處理器）、RAM (隨機存取記憶體）或ROM (唯讀 記憶體）等而構成，並可藉由使記憶在RAM或ROM中的 程式動作而實現。 交易資訊取得部11係透過通信網路200從複數台 ATM 100中依次取得有關各ATM 100的進款和出款的交易 資訊。例如，交易資訊取得部1 1每天在預先決定的時間 取得各ATM 100的交易資訊。交易資訊取得部11所取得 的交易資訊，是表示一萬曰圓的紙幣被提取多少張並被存 入多少張、五千日圓的紙幣被提取多少張並被存入多少張 、一千日圓的紙幣被提取多少張並被存入多少張等分幣種 的進款和出款的資訊。 差分積算部1 2係在每次藉由交易資訊取得部1 1取得 交易資訊時，對每台ATM 100按複數個幣種中的每一幣種 求出進款與出款的差分的積算値。差分積算部1 2例如以 預定的期間（例如90天）爲單位而執行該積算値的計算 。也就是說’在最初的第一天，藉由交易資訊取得部11 所取得的交易資訊所表示的進款與出款的差分値，成爲積 -8- 201118809 算値本身。在藉由交易資訊取得部11取得第二天的交易 資訊時’差分積算部12計算出藉由第二天的交易資訊所 表示的進款與出款的差分，並將該差分値與第一天的差分 値（=截至第一天爲止的積算値）相加，藉此求出截至第 二天爲止的差分的積算値。 同樣地，在藉由交易資訊取得部1 1取得第三天的交 易資訊時，差分積算部（12)計算出藉由第三天的交易資 訊所表示的進款與出款的差分，並將該差分値與截至第二 天爲止的差分的積算値相加，藉此求出截至第三天爲止的 差分的積算値。差分積算部（12)將該動作重複進行90 天。在經過90天後的第一天（第91天），可以將90天 的積算値全部清除而從最初開始進行積算，但是也可以不 這樣進行。在本實施形態中係將截至第一天爲止的積算値 刪除而將第二天視爲積算的起始日，並且，將第91天視 爲積算的終止曰而求出90天的積算値。第92天以後也進 行同樣的處理。藉此，能夠一直得到9 0天的積算値。 其中，在本實施形態中，差分積算部1 2計算進款與 出款的差分，但是按複數個幣種中的每個幣種進行該計算 。具體地說，分別求出一萬日圓的紙幣被提取的張數與被 存入的張數的相差張數、五千日圓的紙幣被提取的張數與 被存入的張數的相差張數、一千日圓的紙幣被提取的張數 與被存入的張數的相差張數，作爲進款與出款的差分。 近似線運算部1 3係由複數個積算値對各自的每台 ATM 100按複數個幣種中的每一幣種求出近似線，其中， -9 - 201118809 複數個積算値是每次藉由交易資訊取得部1 1取得交易資 訊時利用差分積算部1 2所求出的値。也就是說，近似線 運算部13係每天（實際上能夠進行計算的是第90天以後 的每天）由90天份的差分的積算値來求出近似線。具體 地說，藉由線性近似、對數近似、多項式近似、乘冪近似 、指數近似等複數種模式的近似運算，使90天份的積算 値的分佈以各種近似線近似。然後，選擇各種近似線中與 90天份的積算値的相關（correlation )最高的近似線（差 最小的近似線）。 第2圖是表示藉由近似線運算部13所求出的近似線 的一例圖。該第2圖表示對某一台ATM 100的某一幣種所 求出的近似線的例子。在第2圖中，各圓點表示每天的積 算値，直線表示藉由近似線運算部1 3由該積算値所求出 的近似線。一般來說，現金的交易量（出款量和進款量） 係依設置ATM100的地域、店鋪或工作時域等環境的不同 而每台ATM100都不同。但是，在關注一台ATM100時， 由於該AT Μ 100的環境不會大幅地變化，因此，每天的交 易量呈現出大致相同的趨勢的情況較多。因此，90天的 積算値的分佈能夠以如第2圖般往右逐漸上升的直線近似 的情況較多。也就是說，對環境不同的各個ATM 1 00所求 出的近似線，雖然根據進出款的差分的大小而斜率分別不 同，但是均呈直線的情況較多。 所需資金預測部1 4係根據藉由近似線運算部丨3所求 出的近似線與 ATM 100的現有餘款，對各自的每台 -10- 201118809 ATM100預測將來的某一時點的ATM100 在此，ATM 100的現有餘款係藉由從前一 點的ATM 100的餘款，減去從該現金補充 的進出款的差分積算値而求出。另外，將 例如設定爲比ATM 100的現有餘款變爲零 或現金不能完全存入ATM100的現金滿載 間（例如提前一天）的時點。 也就是說，所需資金預測部14係 ATM 100的現有餘款變爲零的無現金時點 存入ATM 100的現金滿載時點，從而預測 點或現金滿載時點提前一天的所需資金量 量）。如上所述，近似線呈往右逐漸上升 多，該直線的斜率表示平均每天的現金增 夠藉由使用近似線的線性函數運算而預測 現金滿載時點及其前一天的餘款。然後， 的餘款預測出該時點的所需資金量。 在此，在本實施形態中，對複數個幣 預測出將來的某一時點的ATM 1 00的所需 一個幣種預測所需資金量時，即使在同一 會根據幣種而在無現金時點產生偏差。因 測部1 4係由近似線對複數個幣種中的: ATM 1 00的現有餘款變爲零的無現金時點 存入ATM 1 00的現金滿載時點，並將比複 測爲ATM 1 0 0的餘款最先變爲零的幣種的 的所需資金量。 次的現金補充時 時點至現在爲止 來的某一時點係 的無現金時點、 時點提前預定時 S近似線預測出 或現金不能完全 出比該無現金時 (補充量或回收 的直線的情況較 減量。因此，能 出無現金時點或 能夠從其前一天 種中的每一幣種 資金量。在對每 ATM 1 00 中，也 此，所需資金預 寒一幣種預測出 或現金不能完全 數個幣種中被預 無現金時點、或 -11 - 201118809 被預測爲最先不能完全存入ATM 100的幣種的現金滿載時 點提前一天的時點作爲將來的某一時點，而預測出每一個 幣種的該時點的所需資金量。 該情況下，不論哪個幣種，都能夠藉由使用 ATM 1 00 的現有餘款和近似線（具體而言爲表示該近似線的近似式 )的線性函數運算而預測出如上所述所決定的將來的某一 時點的所需資金量。此處求出的所需資金量係設爲例如 ATM 100的最適度餘款（在現金補充計畫支援系統10中 預先被設定）、與ATM 100內的現金變爲零或滿載的前一 天的時點的ATM 1 00的餘款（如上所述藉由使用近似線的 線性函數運算而求出）的差額。 補充計畫輸出部1 5係根據藉由所需資金預測部1 4對 各台ATM100按複數個幣種中的每一幣種所求出的將來的 所需資金量，生成在何時對哪台ATM 100補充或回收多少 程度的現金的現金補充計畫並進行輸出。最簡單的方法是 將所需資金預測部1 4中的預測結果直接照原樣作爲現金 補充計畫而輸出的方法。 也就是說，如上所述，所需資金預測部1 4對各個每 台ATM100求出在ATM 100內的現金變爲零或滿載的前一 天所需的資金量。因此，補充計畫輸出部1 5若制定在現 金變爲零或滿載的前一天補充或回收現金的計畫的話，則 能夠生成將藉由所需資金預測部1 4所求出的所需資金量 作爲現金補充量或現金回收量、將現金變爲零或滿載的前 —天作爲補充、回收時機的現金補充計畫。補充計畫輸出 -12- 201118809 部1 5係將該現金補充計畫輸出至未圖示的顯示裝置 '或 輸出至未圖示的印表機。或者，也可以輸出至未圖示的記 錄媒體、或透過通信網路200而輸出至未圖示的其他終端 機。 接著，對如上所述所構成的第一實施形態的現金補充 計畫支援系統10的動作進行說明。第3圖是表示第一實 施形態的現金補充計畫支援系統丨〇的動作例的流程圖。 第3圖所示的流程圖係在交易資訊取得部n從ATM 1 0 0 中取得交易資訊時開始。其中，在此以將過去90天份的 交易資訊已經取得完成且重新取得現在的交易資訊的情況 ，來說明動作。 在第3圖中，首先，交易資訊取得部1 1係判斷是否 已到達作爲取得交易資訊的時機（timing)而決定的規定 時機（步驟S 1 )。在判斷爲已到達規定時機時，交易資 訊取得部1 1從ATM 100中取得有關進款和出款的交易資 訊（步驟S2 )。在此，從已到達規定時機的複數台 ATM100中取得交易資訊。 在交易資訊取得部1 1取得交易資訊時，差分積算部 係將從某台ATM1 00中所取得的交易資訊作爲對象， 求出藉由該交易資訊所表示的進款與出款的差分，並將該 差分相對於過去的積算値進一步進行積算（步驟S3)。 接著，近似線運算部1 3係由90天份的積算値來求出近似 線（步驟S 4 )。 接著，所需資金預測部14係由近似線預測出 -13- 201118809 ATM100的現有餘款變爲零的無現金時點、或現 全存入ATM100的現金滿載時點，從而對複數個 每一幣種預測出比該無現金時點或現金滿載時點 的時點的所需資金量（步驟S5 )。然後，補充 部1 5係根據藉由所需資金預測部1 4按複數個幣 一幣種所求出的將來的所需資金量，生成在何時 象的ATM100補充或回收多少程度的現金的現金 並進行輸出（步驟S6)。 之後，差分積算部1 2進行判斷在藉由交易 部11取得交易資訊的ATM 100中是否還剩下未 補充計畫的ATM 100 (步驟S7 )。在還剩下未生 充計畫的 ATM100時，返回步驟 S3，並以有 ATM100的交易資訊爲對象而執行步驟S3〜S6的 後，在對取得交易資訊的所有ATM 100都生成現 畫後，結束第3圖所示的流程圖的處理。 如以上詳細說明，在第一實施形態中，在每 關ATM 100的進款和出款的交易資訊時，求出進 的差分的積算値，並由該所求出的複數個積算値 似線。然後，根據該近似線和ATM 1 00的現有餘 出將來的某一時點的所需資金量。 根據如上所示所構成的本實施形態的現金補 援系統1 〇，基於ATM 1 00中所進行的進款和出 情況，將現金按多少程度的比率進行增減的交易 爲近似線而被求出，並由該交易的趨勢來預測出 金不能完 幣種中的 提前一天 計畫輸出 種中的每 對作爲對 補充計畫 資訊取得 生成現金 成現金補 關下一台 處理。然 金補充計 次取得有 款與出款 來求出近 款，預測 充計畫支 款的實際 的趨勢作 將來的所 -14- 需資金量。藉此，即使不依靠工作人員的經顆 易地生成在將來的何時對哪台A Τ Μ 1 0 0補充、 度的現金爲佳的最佳現金補充計畫。 201118809 另外，在第一實施形態中，係每天求出 差分積算値的分佈近似的近似線，且每天計 —時點的所需資金量。在同一台ΑΤΜ100中 不會突然大幅地發生變化，每天求出的近似 大幅地發生變化。但是，近似線的角度根據 況一點一點地發生變化。根據本實施形態， 似線的變化而適當地修正將來的所需資金量 在第一實施形態中，由於是對A Τ Μ 1 0 0 零或滿載的前一天的時點的所需資金量進行 在剛對現金進行補充或回收後緊接著便對稍 所需資金量進行預測。相對於此，在對現金 收後經過數日時，由於接近ATM 100內的現 載的日期，因此對數曰後的最接近的所需資 。此時，由於是使用利用每天的計算而被適 近似線來預測最接近的所需資金量’因此能 精度。 藉此，在準備資金的現金中心，能夠從 握稍微遠的將來的所需資金量而確保無過與 金，並且，能夠根據最接近的將來預測而制 充計畫。以往由於爲了避免資金不足的風險 要的大量的資金，因此’存在相對於所確保 輿 內 U i 金 金 % ，也能夠容 回收多少程 9 0天份的 出將來的某 交易的趨勢 也幾乎不會 天的交易情 夠根據該近 的現金變爲 測，因此， 遠的將來的 行補充、回 變爲零或滿 量進行預測 地修正後的 實現高預測 早階段便掌 足的所需資 最佳現金補 確保超過需 資金產生的 -15 - 201118809 利息成本變高的問題。但是，根據本實施形態，由於能夠 確保無過與不足的最佳的所需資金，因此，能夠在避免資 金不足的風險的同時，也謀求利息成本的降低。 另外，在上述第一實施形態中，對將藉由補充計畫輸 出部15所生成的現金補充計畫輸出至顯示裝置等的例子 進行了說明，但是除此之外，也可以將藉由近似線運算部 1 3所求出的近似線形成爲曲線圖而輸出至顯示裝置等。 若將近似線形成爲曲線圖來進行顯示的話，則現金中心的 工作人員能夠透過視覺直覺地掌握進出款交易的趨勢。 藉由近似線運算部1 3所求出的近似線在大部分的情 況下呈一條直線，但是存在由於ATM 100的周圍環境的變 化而交易趨勢發生變化時，近似線不再是一條直線的情況 。例如，也存在呈折線、對數曲線、指數曲線、或者其他 形狀的曲線的情況。預先將近似線形成爲曲線圖而顯示於 顯示裝置的話，能夠迅速地從近似線的形狀的變化察覺 ATM 100中的交易趨勢的變化並採取恰當的對策。 例如，能夠從近似線的形狀直覺地瞭解ATM 1 00以某 一時點爲分界而從進款型（現金的存款多於提款的類型） 變化爲出款型（現金的提款多於存款的類型）的交易趨勢 的變化。該情況下，對於該交易趨勢發生變化的ATM100 ，藉由有別於現金補充計畫而另外緊急地進行現金的補充 ，能夠防無現金狀態於未然。 另外，在上述第一實施形態中，對近似線以一條直線 表示時，藉由使用該直線的近似式的線性函數運算而預測 -16- 201118809 出ATM 100的將來的所需資金量的例子進行了說明，但是 ’在近似線不是以一條直線表示時，只要藉由使用表示該 近似線的近似式的函數運算而預測出ATM 1 00的將來的所 需資金量即可。或者’僅使用過去9〇天份的差分積算値 中的最近數天份的差分積算値而重新求出近似線，且所重 新求出的近似線呈一條直線的話，也可以藉由使用該直線 的近似式的線性函數運算而預測出ATM 1 00的將來的所需 資金量。 另外，在上述第一實施形態中對下述例子進行了說明 ’即，在對每一個幣種預測有關某台ATM 1 00的將來的所 需資金量時，將比複數個幣種中被預測爲ATM100的餘款 最先變爲零的幣種的無現金時點、或被預測爲最先不能完 全存入ATM 100的幣種的現金滿載時點提前一天的時點作 爲將來的某一時點，並將該時點作爲所有幣種的現金補充 、回收時機的例子，但是本發明並不限定於此。例如，也 可以將比對於複數個幣種而分別被預測爲ATM 100的餘款 變爲零的無現金時點、或被預測爲最先不能完全存入 ATM 100的幣種的現金滿載時點提前一天的時點，作爲每 —個幣種的現金補充、回收時機。 該情況下，由於即使是—台ATM 100 ’也是在不同的 時機對每一個幣種進行現金的補充、回收’因此’能夠在 最佳的時機對每一個幣種進行現金的補充、回收。但是’ 由於送款車輛的運送次數會增加’因此’在重視運送效率 時，以如上述實施形態所示’將比複數個幣種中被預測爲 -17- 201118809 ATM 100的餘款最先變爲零的幣種的無現金時點、或被預 測爲最先不能完全存入ATM 100的幣種的現金滿載時點提 前一天的時點作爲所有幣種的現金補充、回收時機爲佳。 另外，在上述第一實施形態中，對以90天爲單位而 求出近似線的例子進行了說明，但是，90天這一數字僅 爲例示。例如，也可以以一週單位或一個月單位而求出近 似線。但是，由於以長時距（span )求出近似線者能夠得 到更高的相關，因此，從提高預測的精度方面來看，以 90天左右的長時距求出近似線爲佳。 接著，根據附圖對本發明的第二實施形態進行說明。 第4圖是表示第二實施形態的現金補充計畫支援系統的功 能結構例的方塊圖。其中，在該第4圖中，由於標註與第 1圖所示的元件符號相同的元件符號者具有相同的功能， 因此在此省略重複的說明。如第4圖所示，第二實施形態 的現金補充計畫支援系統20係具備有交易資訊取得部11 、差分積算部12、近似線運算部13、所需資金預測部14 、補充計畫輸出部15、調配車輛數計畫部21以及調配車 輛計畫修正部22作爲其功能結構而構成。 調配車輛數計畫部21係根據對複數台ATM 100分別 求出的將來某一時點的所需資金量（也就是說，藉由補充 計畫輸出部1 5所生成的現金補充計畫）、平均每輛送款 車輛的現金裝載量、以及一輛送款車輛平均每天的運送場 所數，而生成送款車輛的日單位的調配車輛數計畫。 例如，如果在將來的某一天必須補充現金的ATM 1 00 -18- 201118809 的數量爲130台、一輛送款車輛的平均每天的運送場所數 爲13處的話，則該將來的某一天所需的送款車輛的車輛 數爲10輛。但是，在應向130台ATM100補充的所需資 金量的總計多於將平均每輛送款車輛的現金裝載量變爲 1 〇倍後的1 0輛份的現金裝載量時，以1 0輛送款車輛並 無法運送所有的所需資金。該情況下，爲了運送超出的所 需資金而需要準備追加車輛數的送款車輛。例如，若超出 的所需資金量少於平均每輛的現金裝載量時，所需的送款 車輛爲共計1 1輛。 或者，調配車輛數計畫部2 1也可以如下所示計算出 送款車輛的所需車輛數。首先，藉由將應向1 3 0台 ATM 100補充的所需資金量的總計除以平均每輛送款車輛 的現金裝載量，暫時求出將來的某一天所需的送款車輛的 車輛數。然後，檢驗將應補充現金的ATM 1 00的台數（ 130台）除以暫時的調配車輛數後所得的値，是否超過一 輛送款車輛平均每天的運送場所數，若超過，即增加送款 車輛的所需車輛數。 調配車輛數計畫部21係藉由上述任意一種方法並以 曰爲單位來進行送款車輛的所需車輛數的計算，藉此生成 送款車輛的日單位的調配車輛數計畫。其中，此處所說明 的送款車輛的所需車輛數的計算是比較簡單的方法，但是 計算的方法並不限定於此。另外，也可以將藉由調配車輛 數計畫部2 1所求出的日單位的調配車輛數作爲最終的調 配車輛數計畫，但是在本實施形態中，係藉由調配車輛計 -19- 201118809 畫修正部22進一步對該計畫進行修正。 調配車輛計畫修正部22係使送款車輛 運送場所數和送款車輛的日單位的調配車輛 預定値，將平均每天的總運送場所數設定爲固 提。在此前提下，調配車輛計畫修正部22 配車輛數計畫部21所生成的調配車輛數計畫 進行從後一天的調配車輛中調撥在前一天的調 滿足總運送場所數的固定値的部分的處理。 第5圖是用於說明利用調配車輛計畫修H 的調配車輛數計畫的修正處理的圖，將每天的 數作成曲線圖而以模式化表示。總運送場所數 送款車輛平均每天的運送場所數（例如13處 調配車輛數計畫部21所求出的日單位的調配 5圖（a )的情況）、或設定爲固定値的曰單 輛數（第5圖（b )〜（d )的情況）後所得的ii 第5圖（a)表示藉由調配車輛數計畫部 調配車輛數計畫本身。也就是說，平均每天的 數未被設定爲固定値，總運送場所數以日爲單 如上所示，在將藉由調配車輛數計畫部21所 位的調配車輛數作爲最終的調配車輛數計畫時 的不同而送款車輛的所需車輛數產生偏差。 相對於此，第5圖（b)~(d)表示將平 運送場所數設定爲固定値，且每天依次進行利 (left justified type )進行的處理的情況，其 f平均每天的 :分別固定爲 定値作爲前 :根據藉由調 ，每天依次 配車輛中未 ί部22進行 總運送場所 ，是對一輛 )乘以藉由 車輛數（第 位的調配車 直。 2 1所求出的 總運送場所 位而不同。 求出的曰單 ，根據曰期 均每天的總 用前補方式 中，前補方 -20- 201118809 式’是指從後一天的調配車輛中調撥在前一天的調 中未滿足總運送場所數的固定値的部分的方式。在 (b) ~(d)的例子中’分別使送款車輛的平均每 送場所數固定爲13處、將送款車輛的日單位的調 數固定爲4輛’從而將平均每天的總運送場所數 52處。 在第5圖（b)中，將在第i天的調配車輛中 總運送場所數（5 2處）的斜線部分，從其一天之 i + Ι天的調配車輛中調撥。藉由該操作，原預定 i+Ι天的一部分有關ATM100的送款計畫，變更爲 一天的第i天的計畫。 在第5圖（c )中，將在第i+1天的調配車輛 利用第5圖（b )的操作而被調撥的部分之外的剩 配車輛）中未滿足總運送場所數的斜線部分，從其 後的第i + 2天的調配車輛和兩天之後的第i + 3天的 輛中調撥。藉由該操作’原預定輪到第i + 2天的全 關ATM100的送款計畫和原預定輪到第i + 3天的一 關ATM100的送款計畫’變更爲輪爲第i+1天的計 在第5圖（d)中’將在第i + 2天的調配車輛 利用第5圖（c )的操作而被調撥的部分之外的剩 配車輛）中未滿足總運送場所數的斜線部分’從其 後的第i + 3天的調配車輛中調撥。藉由該操作’原 到第i + 3天的一部分有關ΑΤΜ1〇〇的送款S十畫’變 爲第i + 2天的計畫。 配車輛 第5圖 天的運 配車輛 固定爲 未滿足 後的第 輪到第 輪爲前 (除去 餘的調 一天之 調配車 部的有 部分有 畫。 (除去 餘的調 一天之 預定輪 更爲輪 -21 - 201118809 藉由進行如上所示之調配車輛數計畫 夠使每天的調配車輛數平均化，從而能夠 。另外，也能夠使送款車輛不會經常處於 地進行工作。也就是說，如第5圖（a) 調配車輛數存在偏差的情況不變時，必須 運送場所數相對應地準備送款車輛，從而 少時會產生不工作的閒置的送款車輛。相 使每天的調配車輛數平均化的話，預先僅 度的車輛數的送款車輛即可，從而能夠最 款車輛。 另外，在第二實施形態中，由於利用 配車輛數計畫的修正，因此，無需極力錯 輸出部15所生成的最佳的現金補充時機 的調配車輛數平均化。201118809 VI. Description of the Invention: [Technical Field of the Invention] The present invention relates to a method for supporting a cash replenishment support system, and is particularly suitable for obtaining a basis for predicting each of a plurality of ATMs (automated tellers) from an automatic basis. The system's cash replenishment plan support system support method for the amount of money required for each of the ATMs. [Prior Art] In recent years, in addition to financial institutions such as banks, a large number of places such as stations have been installed. The user can use the user's own operation to make a cash deposit (incoming) or the like by using the magnetic card or the 1C card. Such transactions are frequently caused by cashless transactions and require cash replenishment. However, the amount of cash trade (the amount of cash that is sent to the ATM, the store, or the working time domain, etc., and the amount of cash that is added to the best supplement ATM is usually different. However, the experience of conventional technology It is extremely difficult to develop a supplementary plan for the best supplement plan for multiple ATMs. Therefore, it is a case of cashless cash flow and cash advances. The supplement and the cash replenishment plan for the easy ATM, except for the convenience store or the car _ into the ATM, the withdrawal (withdrawal) or the ground, in order not to be appropriate for the ATM). Therefore, in the plan, for each person who makes a decision on their own, there is no staff judgment 201118809. In this regard, it is proposed to calculate the cumulative amount of deposits and the cumulative amount of withdrawals of each ATM, and based on the calculation A technique for predicting the amount of money required for each ATM by the difference between the accumulated amount and the accumulated amount of the money (for example, refer to Patent Documents 1 to 3)» In these Patent Documents 1 to 3, the calculation by currency is also described. The amount of money required for each ATM. It is described that the amount of money required for the entire ATM unit can be easily predicted based on the conventional technique. (PRIOR ART DOCUMENT) Patent Document 1: Japanese Laid-Open Patent Publication No. Hei No. Hei No. Hei No. Hei No. Hei. In the prior art described in the above-mentioned Patent Documents 1 to 3, the amount of money required is predicted based on the difference between the accumulated amount of money collected and the accumulated amount of money to be dispensed. The prediction by the difference between the accumulated amount of the deposit and the accumulated amount of the withdrawal is not described in detail. In other words, it is unclear what kind of calculation is needed to determine the amount of funds required based on the change in the cumulative amount of deposits and the change in the cumulative amount of withdrawals. The present invention has been made to solve such problems, and an object of the present invention is to provide a specific means for easily generating an optimum cash replenishment plan for each of a plurality of ATMs without relying on the experience of a worker. 201118809 (Means for Solving the Problem) In order to solve the above problem, in the present invention, the accumulation of the difference between the transaction amount of the automatic teller machine (ATM) and the payment and the payment of the withdrawal is obtained by the obtained method. line. Then, based on the approximate line and the automatic existing balance, the future time point is predicted (the effect of the invention). According to the invention constructed as described above, in the time series information of the deposit and withdrawal according to the Cash is calculated by approximating the trend of the increase or decrease of the trade. It is predicted in the ATM at a certain point in the future, even if it is not dependent on the experience of the staff, how much can be added or recovered to the ATM. Cash replenishment plan. [Embodiment] The following is a block diagram showing an example of a cash replenishment plan according to a first embodiment of the present invention. As shown in Fig. 1, the first solid plan support system 10 is connected via a communication station ATM 100 such as the Internet. Further, the cash replenishment plan is provided with the transaction information acquisition unit 11'. The difference integration unit 12 obtains the amount of funds required for the number of ATMs to be added to the ATM each time the information is acquired. How many times the ratio is obtained from the ATM and the amount of funds required from the transaction. It is easy to generate an explanation of the best state in which cash is better. The cash compensation network 200 and the plural of the function configuration of the first support system: the aid system 1 〇 ί, the approximate line calculation unit 201118809 1 3, the required fund prediction unit 1 4, and the supplementary plan output unit 1 5 as their functions Structured. Each function block of the cash replenishment support system (1 〇) shown in Fig. 1 can also be realized by any of a hardware configuration, a DSP (Digital Signal Processor, a digital signal processor), and a software. For example, when implemented by software, the cash replenishment support system 10 of the present embodiment actually includes a CPU (central processing unit) of a computer, an MPU (microprocessor), and a RAM (random access memory). It is constituted by a ROM (Read Only Memory) or the like, and can be realized by causing a program stored in a RAM or a ROM to operate. The transaction information acquisition unit 11 sequentially acquires transaction information on the payment and the withdrawal of each ATM 100 from the plurality of ATMs 100 via the communication network 200. For example, the transaction information acquisition unit 1 1 acquires transaction information of each ATM 100 every day at a predetermined time. The transaction information obtained by the transaction information acquisition unit 11 is how many sheets of 10,000-yuan banknotes are extracted and how many sheets are deposited, and how many sheets are extracted and stored in a thousand yen. How many sheets are withdrawn and how many deposits and withdrawals are deposited into the currency. Each time the transaction information acquisition unit 1 1 acquires the transaction information, the difference totalizer 1 2 calculates the difference between the difference between the payment and the withdrawal for each of the plurality of currencies in each ATM 100. . The difference integrating unit 1 2 performs the calculation of the integrated 例如 in units of, for example, a predetermined period (e.g., 90 days). In other words, on the first day of the first day, the difference between the deposit and the withdrawal indicated by the transaction information obtained by the transaction information acquisition unit 11 becomes the product -8-201118809. When the transaction information acquisition unit 11 acquires the transaction information of the next day, the difference integration unit 12 calculates the difference between the payment and the withdrawal indicated by the transaction information of the next day, and compares the difference with the first The difference 天 of days (= integrated 截至 up to the first day) is added to obtain the integrated 値 of the difference up to the next day. Similarly, when the transaction information acquisition unit 1 1 acquires the transaction information of the third day, the difference accumulation unit (12) calculates the difference between the payment and the withdrawal indicated by the transaction information of the third day, and This difference 値 is added to the integrated 値 of the difference up to the next day, thereby obtaining the integrated 値 of the difference up to the third day. The difference accumulation unit (12) repeats the operation for 90 days. On the first day after 90 days (day 91), the 90-day total accumulation can be cleared and integrated from the beginning, but this may not be done. In the present embodiment, the accumulation 値 as of the first day is deleted, and the second day is regarded as the start date of the integration, and the 91st day is regarded as the cumulative termination 曰 to obtain the 90-day accumulation 値. The same treatment was carried out after the 92nd day. By this, it is possible to always obtain the accumulated calculation of 90 days. In the present embodiment, the difference totalizer 1 2 calculates the difference between the deposit and the withdrawal, but performs the calculation for each of the plurality of currencies. Specifically, the number of sheets of the number of sheets of the ten thousand yen bills to be extracted and the number of sheets to be deposited, the number of sheets of the five thousand yen bills, and the number of sheets to be deposited are respectively determined. The difference between the number of sheets of one thousand yen bills to be extracted and the number of sheets to be deposited is the difference between the deposit and the withdrawal. The approximate line calculation unit 1 3 obtains an approximate line for each of the plurality of currencies for each ATM 100 by a plurality of integrated calculations, wherein -9 - 201118809 is a plurality of integrated calculations each time When the transaction information acquisition unit 1 1 acquires the transaction information, the defect obtained by the difference totalizer 1 2 is used. In other words, the approximate line calculation unit 13 obtains an approximate line from the accumulated product of the difference of 90 days per day (actually, the day after the 90th day can be calculated). Specifically, the approximation of a plurality of modes, such as linear approximation, logarithmic approximation, polynomial approximation, power approximation, and exponential approximation, approximates the distribution of the accumulated 値 of 90 days by various approximation lines. Then, the highest approximation line (the approximate line with the smallest difference) of the correlations of the 90-day totals among the various approximation lines is selected. Fig. 2 is a view showing an example of an approximate line obtained by the approximate line calculation unit 13. This Fig. 2 shows an example of an approximate line obtained for a certain currency of a certain ATM 100. In Fig. 2, each dot indicates the integrated 値 per day, and the straight line indicates the approximate line obtained by the approximate line calculating unit 13 from the integrated 値. In general, the amount of cash transactions (delivery amount and amount of incoming money) varies depending on the environment in which the ATM 100 is installed, the store, or the working time domain. Each ATM 100 is different. However, when one ATM 100 is focused on, since the environment of the AT Μ 100 does not largely change, the daily transaction volume tends to be substantially the same. Therefore, the distribution of the 90-day total enthalpy can be approximated by a straight line that gradually rises to the right as shown in Fig. 2. In other words, the approximate line obtained for each ATM 100 having a different environment has a different slope depending on the size of the difference between the incoming and outgoing, but it is often a straight line. The required fund prediction unit 14 predicts the ATM 100 at a certain point in time for each of each of the-10-201118809 ATM 100 based on the approximate line obtained by the approximate line calculation unit 丨3 and the existing balance of the ATM 100. Thus, the existing balance of the ATM 100 is obtained by subtracting the difference accumulation of the incoming and outgoing money from the cash supplement by the balance of the ATM 100 from the previous point. In addition, for example, it is set to a time when the existing balance of the ATM 100 becomes zero or the cash cannot be completely stored in the cash full load of the ATM 100 (for example, one day earlier). That is to say, the required fund prediction unit 14 is a cash-free point at which the existing balance of the ATM 100 becomes zero, and is stored in the cash-filled time point of the ATM 100, thereby predicting the amount of money required for one day in advance when the cash is fully loaded. As mentioned above, the approximation line gradually rises more to the right, and the slope of the line indicates that the average daily cash increase is predicted by the linear function operation using the approximate line to predict the cash full load point and the balance of the previous day. Then, the balance of the forecast predicts the amount of funds required at that point in time. Here, in the present embodiment, when a plurality of coins are predicted to predict the amount of money required for one currency type of ATM 100 at a certain point in the future, even if the same money is generated at the time of no cash according to the currency type deviation. The measurement unit 1 4 is in the approximate number of pairs of currencies: the cashless point when the existing balance of ATM 100 becomes zero is stored in the cash full load point of ATM 100, and the ratio is retested to ATM 1 0 The balance of 0 is the amount of money required to change to the currency of zero first. The cash replenishment time point is the cash-free time point at a certain point in time up to the present time, and the time is predicted in advance when the S-approximate line is predicted or the cash cannot be completely out of the cashless amount (the amount of the replenishment or the recovered straight line is reduced) Therefore, the amount of money that can be cashed out or can be from the previous day of the previous day. In the case of each ATM 100, the required funds are predicted to be one currency or the cash cannot be fully counted. When there is no cash in the currency, or -11 - 201118809 is predicted to be the first cash that cannot be fully stored in the ATM 100. The time of the full load is one day in advance, and each coin is predicted. The amount of money required at that point in time. In this case, regardless of the currency, a linear function can be used by using the existing balance of ATM 100 and the approximation line (specifically, an approximation representing the approximation) The calculation predicts the amount of funds required at a certain point in the future determined as described above. The amount of funds required here is set to the optimum balance of, for example, ATM 100 (in the cash supplement plan) The balance of the ATM 100 that was previously set in the aid system 10 and the time when the cash in the ATM 100 became zero or fully loaded (as determined by the linear function calculation using the approximate line as described above) The supplementary plan output unit 1 5 is based on the amount of future required funds obtained by the required fund predicting unit 14 for each of the plurality of currencies of each ATM 100, and when is the pair generated Which ATM 100 replenishes or recovers the cash replenishment plan of the cash amount and outputs it. The easiest method is to output the prediction result in the required fund prediction unit 14 as a cash supplement plan as it is. That is, as described above, the required funds predicting unit 14 finds the amount of money required for each of the ATMs 100 to be zero or full load on the ATM 100. Therefore, the supplementary plan output unit 1 (5) If a plan is made to replenish or recycle cash one day before the cash becomes zero or full, it is possible to generate the amount of money required by the required fund forecasting unit 14 as a cash replenishment amount or cash recovery. Volume, change cash Zero- or full-load pre-day as a cash replenishment plan for replenishment and recovery timing. Supplementary plan output -12- 201118809 Part 1 5 Output the cash replenishment plan to a display device not shown' or output to The printer shown in the figure may be output to a recording medium (not shown) or output to another terminal (not shown) via the communication network 200. Next, the first embodiment configured as described above The operation of the cash replenishment support system 10 will be described. Fig. 3 is a flowchart showing an operation example of the cash replenishment support system 第一 according to the first embodiment. The flowchart shown in Fig. 3 is based on transaction information. The acquisition unit n starts when the transaction information is acquired from the ATM 100. Here, the operation will be described here in the case where the transaction information of the past 90 days has been obtained and the current transaction information is re-acquired. In Fig. 3, first, the transaction information acquisition unit 1 1 determines whether or not a predetermined timing has been reached as a timing for acquiring transaction information (step S1). When it is determined that the predetermined timing has been reached, the transaction information acquisition unit 1 1 acquires the transaction information regarding the deposit and the withdrawal from the ATM 100 (step S2). Here, the transaction information is obtained from a plurality of ATMs 100 that have reached the predetermined timing. When the transaction information acquisition unit 1 1 acquires the transaction information, the difference integration unit obtains the difference between the payment and the withdrawal indicated by the transaction information from the transaction information acquired in one of the ATMs 100, and This difference is further integrated with respect to the past integration (step S3). Next, the approximate line calculation unit 13 obtains an approximate line from the accumulated product of 90 days (step S4). Next, the required fund prediction unit 14 predicts from the approximate line that the existing balance of the ATM 100 becomes zero, or the cash-free time point that is currently stored in the ATM 100, and thus for each of the plurality of currencies. The amount of funds required at the time point when the cashless hour or cash is fully loaded is predicted (step S5). Then, the replenishing unit 15 generates cash in accordance with how much the amount of cash is replenished or recovered by the ATM 100 at the time of the required amount of funds obtained by the required fund predicting unit 14 in a plurality of coins per currency. And output is performed (step S6). Thereafter, the difference totaling unit 1 2 determines whether or not the ATM 100 having the unreplenished plan remains in the ATM 100 that has acquired the transaction information by the transaction unit 11 (step S7). When there is still an ATM 100 that has not been recharged, the process returns to step S3, and after steps S3 to S6 are performed for the transaction information of the ATM 100, after all the ATMs 100 that have acquired the transaction information are generated, The processing of the flowchart shown in FIG. 3 is ended. As described in detail above, in the first embodiment, when the transaction information of the ATM 100 is paid and the transaction information of the withdrawal is made, the integrated calculation of the difference is obtained, and the complex number of the calculated lines is obtained. . Then, according to the approximation line and the existing amount of ATM 100 existing surplus at a certain point in the future. According to the cash supplement system 1 of the present embodiment configured as described above, based on the amount of money received and discharged in the ATM 100, the transaction in which the amount of cash is increased or decreased is approximated. Out, and the trend of the transaction is used to predict each pair of the one-day plan output in the unfinished currency, as the next step is to generate cash for the supplementary plan information. In addition, the funds are replenished to obtain the funds and funds to determine the actual trend of the expenditures for the future. In this way, even if you don't rely on the staff, you can easily generate the best cash replenishment plan for which A Τ 0 1 0 0 supplement, the amount of cash is good in the future. In addition, in the first embodiment, the approximate line of the distribution approximation of the difference product 値 is obtained every day, and the amount of funds required at the time point is calculated every day. In the same ΑΤΜ100, there is no sudden and significant change, and the approximation obtained every day changes greatly. However, the angle of the approximate line changes little by little. According to the present embodiment, the amount of money required in the future is appropriately corrected in accordance with the change in the line. In the first embodiment, the amount of money required for the time of the previous day of A Τ Μ 1 0 0 or full load is performed. Just after replenishing or recycling the cash, the amount of money required is predicted. On the other hand, when the cash is collected for several days, since it is close to the date of the load in the ATM 100, the closest requirement is after the logarithm. At this time, since it is the use of the daily calculation to be approximated by the line to predict the closest required amount of funds', it is possible to accurately. In this way, in the cash center where the funds are prepared, it is possible to ensure that there is no excess of money from the amount of money required in the future, and that the plan can be prepared based on the closest future prediction. In the past, due to the large amount of funds required to avoid the risk of insufficient funds, there is almost no trend in the future of a certain transaction that is able to recover 90% of the total amount of U i gold in the guarantee. The trading situation of Huitian is enough to be measured according to the recent cash. Therefore, the far-term future line is supplemented, the return is zero or the full amount is predicted, and the required amount of the high-predicting early stage is achieved. Good cash make up to ensure that the cost of interest is higher than the -15 - 201118809 interest generated. However, according to the present embodiment, since it is possible to secure the optimum required funds without excessive or insufficient, it is possible to avoid the risk of insufficient funds and to reduce the interest cost. Further, in the above-described first embodiment, an example in which the cash replenishment plan generated by the supplementary plan output unit 15 is output to the display device or the like has been described. However, it may be approximated by The approximate line obtained by the line calculation unit 13 is formed into a graph and output to a display device or the like. If the approximate line is formed as a graph for display, the cash center staff can intuitively grasp the trend of the incoming and outgoing transactions. The approximate line obtained by the approximate line calculation unit 13 is a straight line in most cases, but there is a case where the approximate line is no longer a straight line when the transaction tendency changes due to the change of the surrounding environment of the ATM 100. . For example, there are also cases where a curve is a broken line, a logarithmic curve, an exponential curve, or another shape. When the approximate line is formed as a graph and displayed on the display device in advance, it is possible to quickly detect the change in the transaction tendency in the ATM 100 from the change in the shape of the approximate line and take appropriate countermeasures. For example, it is possible to intuitively understand from the shape of the approximate line that ATM 100 changes from the incoming type (cash deposits more than the type of withdrawal) to the withdrawal type (the cash withdrawal is more than the deposit). Type) changes in trading trends. In this case, the ATM 100 whose transaction tendency has changed can be additionally supplemented with cash by being different from the cash replenishment plan, thereby preventing the cashless state from happening. Further, in the first embodiment described above, when the approximate line is represented by a straight line, an example of predicting the future required amount of money of the ATM 100 by using the linear function calculation of the approximate expression of the straight line is performed. In addition, when the approximate line is not represented by a straight line, it is only necessary to predict the future required amount of money of the ATM 100 by using a function calculation indicating an approximate expression of the approximate line. Or by using only the difference product of the nearest days in the difference product of the past 9 days, and re-determining the approximation line, and if the re-determined approximation line is in a straight line, you can also use the line. The approximation of the linear function operation predicts the future amount of money required for ATM 100. Further, in the first embodiment described above, the following examples have been described. That is, when the amount of money required for the future of a certain ATM 100 is predicted for each currency, it is predicted in a plurality of currencies. The cash-free time point of the currency in which the balance of the ATM 100 first becomes zero, or the time when the cash is predicted to be the first currency that cannot be completely stored in the ATM 100, is a point in time in advance, and will be a certain point in the future. This time point is an example of the cash replenishment and recovery timing of all the currencies, but the present invention is not limited thereto. For example, it is also possible to introduce a cash-free time point that is predicted to be the balance of the ATM 100 for a plurality of currencies, or a cash-free time point that is predicted to be the first currency that cannot be completely stored in the ATM 100, one day earlier. At the time, as a cash replenishment and recycling opportunity for each currency. In this case, even if it is the ATM 100', cash is replenished and collected for each currency at a different timing. Therefore, it is possible to replenish and recycle cash for each currency at the optimum timing. However, 'the number of shipments of the delivery vehicle will increase'. Therefore, when the transportation efficiency is emphasized, as shown in the above embodiment, the balance of the ATM 100 predicted to be -17-201118809 in the first few currencies will be changed first. The cash-free time of the zero currency, or the cash that is predicted to be the first currency that cannot be fully stored in the ATM 100 is fully loaded at the time of the day as the cash replenishment and recovery timing of all currencies. Further, in the first embodiment described above, an example in which the approximate line is obtained in units of 90 days has been described. However, the number of 90 days is merely shown as an example. For example, the approximate line can also be obtained in units of one week or one unit. However, since a higher correlation can be obtained by finding the approximate line with a long time span, it is preferable to obtain an approximate line with a long time interval of about 90 days from the viewpoint of improving the accuracy of prediction. Next, a second embodiment of the present invention will be described with reference to the drawings. Fig. 4 is a block diagram showing an example of the functional configuration of the cash replenishment plan support system of the second embodiment. In the fourth embodiment, the same reference numerals are given to the same reference numerals as those in the first embodiment, and thus the overlapping description will be omitted. As shown in FIG. 4, the cash replenishment plan support system 20 of the second embodiment includes a transaction information acquisition unit 11, a difference integration unit 12, an approximate line calculation unit 13, a required fund prediction unit 14, and a supplementary plan output. The unit 15, the planned vehicle number planning unit 21, and the deployment vehicle plan correction unit 22 are configured as functional configurations. The deployment vehicle number planning unit 21 is based on the amount of funds required at a certain point in time for each of the plurality of ATMs 100 (that is, the cash replenishment plan generated by the supplementary plan output unit 15), The average cash load per vehicle and the average number of shipping locations per day for a sending vehicle, and the number of vehicles allocated for the day unit of the sending vehicle. For example, if the number of ATMs 1 00 -18- 201118809 that must be replenished with cash on a certain day in the future is 130, and the average number of delivery places per day for a delivery vehicle is 13, then the future day will be required. The number of vehicles for the delivery vehicle is 10. However, when the total amount of funds required to be replenished to 130 ATMs is more than the cash load of 10 vehicles after the average cash load per vehicle is changed to 1 ,, 10 vehicles are delivered. The vehicle is not able to ship all the required funds. In this case, it is necessary to prepare a delivery vehicle for the number of additional vehicles in order to transport the excess funds required. For example, if the required amount of funds exceeded is less than the average cash load per vehicle, the total number of vehicles required to be delivered is 11. Alternatively, the number of required vehicles of the delivery vehicle may be calculated as shown below. First, by dividing the total amount of funds required to be replenished to 130 ATMs by the average cash load per vehicle, temporarily find the number of vehicles required to deliver the vehicle on a certain day in the future. . Then, it is checked whether the number of ATMs that should be replenished with cash (130 units) divided by the number of temporary vehicles allocated is more than the average number of places per day for a sending vehicle. If it exceeds, it will be increased. The number of vehicles required for the vehicle. The dispatching vehicle number planning unit 21 calculates the number of required vehicles of the paying vehicle by the above-described method and generates the number of planned vehicles in the daily unit of the paying vehicle. Among them, the calculation of the required number of vehicles of the delivery vehicle described here is a relatively simple method, but the calculation method is not limited thereto. In addition, the number of dispatched vehicles in the daily unit obtained by the dispatching vehicle number calculation unit 21 may be used as the final planned vehicle number. However, in the present embodiment, the vehicle is allocated by the vehicle -19- The 201118809 drawing correction unit 22 further corrects the plan. The deployment vehicle plan correction unit 22 sets the number of the delivery vehicle transportation destinations and the daily vehicle allocation unit of the delivery vehicle, and sets the average number of total transportation destinations per day as the default. On the premise of this, the deployment vehicle plan correction unit 22 allocates the number of planned vehicles generated by the vehicle number planning unit 21 to transfer the fixed number of vehicles that have been transferred from the next day's deployment vehicle to the previous day. Partial processing. Fig. 5 is a view for explaining a correction process of the planned vehicle number plan using the deployment vehicle plan repair H, and graphically shows the daily number as a graph. The number of total transportation places The number of transportation places per day of the delivery vehicle (for example, the allocation of the daily unit obtained by the 13-numbered vehicle number planning unit 21 (a)) or the fixed unit The number (the case of Fig. 5(b) to (d)) is obtained by ii. Fig. 5(a) shows the vehicle number plan itself by the number of vehicles to be allocated. In other words, the average number of days per day is not set to a fixed number, and the total number of places to be transported is as shown in the above table, and the number of vehicles to be allocated by the number of vehicles in the planing unit 21 is used as the final number of vehicles. There is a difference in the number of vehicles required to deliver the vehicle when the plan is different. On the other hand, (b) to (d) of FIG. 5 show a case where the number of flat transport places is set to be fixed and the processing is performed by the left justified type every day, and the f average daily is fixed to As for the front: According to the adjustment, the total transportation location is carried out in the vehicle in the middle of the day, and the vehicle is transported by the number of vehicles. The location is different. The obtained stipulations are based on the total daily use of the squatting period. The former remedy -20-201118809 type refers to the transfer from the day after the deployment of the vehicle in the previous day. In the case of (b) to (d), the number of the average number of places to be delivered to the delivery vehicle is fixed at 13 and the daily unit of the delivery vehicle is adjusted. The number is fixed at 4 vehicles' so that the average number of total transportation places per day is 52. In Figure 5 (b), the number of total transportation places (52 points) in the deployment vehicle on the i-th day will be In the day of the i + Ι day of the allocation of vehicles in the allocation. With this operation, the original I plan to transfer the ATM 100's payment plan to the i-day of the day. In Figure 5 (c), we will use the fifth map on the i+1th day. b) The portion of the remaining vehicle that is not transferred to the operation) is not satisfied with the number of the total number of places to be transported, from the next i + 2 days of the deployment of the vehicle and the i + 3 days after the second day Transfer in the middle of the car. With this operation, the original plan to turn to the i-th + 2 day of the full-off ATM 100 and the original scheduled turn to the i + 3 day of the ATM 100's delivery plan 'change to the wheel for the i+ In the case of the first day, in the figure (d), the total transportation place is not satisfied in the remaining vehicle other than the part allocated by the operation of the fifth (c) in the i-th + 2nd day. The slashed portion of the number is transferred from the next i + 3 day blending vehicle. By the operation 'original to the part of the i-th + 3rd day, the payment of the S-picture of the ΑΤΜ1〇〇' becomes the i-th + 2 day plan. The delivery vehicle of the fifth day of the vehicle is fixed to the front of the first round after the unsatisfied (there is a part of the redemption of the car part after the adjustment of the remaining one day. Wheel-21 - 201118809 It is possible to average the number of deployed vehicles per day by performing the number of planned vehicles as shown above. In addition, it is also possible to make the paying vehicle not always work on the ground. If the number of vehicles to be distributed does not change as shown in Fig. 5(a), the number of transport locations must be correspondingly prepared for the delivery vehicle, so that there is a small idle transport vehicle that does not work. If the number is averaged, the vehicle can be transported only in advance with a minimum number of vehicles, and the vehicle can be the most expensive. In addition, in the second embodiment, since the correction is performed by the number of vehicles, there is no need for an error output unit. The number of deployed vehicles with the best cash replenishment generated by 15 is averaged.
其中，在上述第一實施形態和第二實 易資訊取得部11透過通信網路200從複 直接取得交易資訊的例子進行了說明，但 定於此。例如，也可以形成爲使其他的終 介於ATM 100與交易資訊取得部1 1之間 部1 1間接地取得複數台ATM 100的交易S 此外，上述實施形態均只是表示實施 化的一例的方式，並不能據此而限定性地 術範圍。也就是說，本發明在不脫離其精 的情況下能夠以各種方式進行實施。 的修正處理，能 實現有效的送款 閒置狀態而有效 所示保持每天的 預先與峰値的總 在總運送場所數 對於此，若能夠 準備所需最小限 大限度地活用送 前補方式進行調 開藉由補充計畫 ，而能夠使每天 施形態中，對交 數台 ATM100中 是本發明並不限 端機或記錄媒體 ，交易資訊取得 ί訊。 本發明時的具體 解釋本發明的技 神或其主要特徵 -22- 201118809 【圖式簡單說明】 第1圖是表示第一實施形態的現金補充計畫支援系統 的功能結構例的方塊圖。 第2圖是表示藉由本實施形態的近似線運算部所求出 的近似線的一例圖。 第3圖是表示第一實施形態的現金補充計畫支援系統 的動作例的流程圖。 第4圖是表示第二實施形態的現金補充計畫支援系統 的功能結構例的方塊圖。 第5圖是用於說明第二實施形態的調配車輛計畫修正 部進行的調配車輛數計畫的修正處理的圖。 【主要元件符號說明】 1 0 ' 20 :現金補充計畫支援系統 1 1 :交易資訊取得部 12 :差分積算部 1 3 :近似線運算部 1 4 :所需資金預測部 1 5 :補充計畫輸出部In the above, the first embodiment and the second fact information acquisition unit 11 have described the example in which the transaction information is directly obtained from the communication network 200, but it is to be noted. For example, it is also possible to form a transaction S in which a plurality of ATMs 100 are indirectly obtained between the ATM 100 and the transaction information acquisition unit 1 1 in addition, and the above embodiments are merely examples of implementation examples. It is not possible to limit the scope of the procedure accordingly. That is, the present invention can be implemented in various ways without departing from the essence thereof. The correction process can realize the effective payment idle state and effectively display the total number of pre- and peak-to-peak total transportation sites per day. For this, if the minimum required limit can be prepared, the pre-replenishment mode can be used. By replenishing the plan, it is possible to make the daily transaction form, and the transaction ATM information is the invention of the present invention and is not limited to the machine or the recording medium, and the transaction information is obtained. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT OF THE INVENTION The present invention is a block diagram of a functional configuration of a cash replenishment support system according to a first embodiment. Fig. 2 is a view showing an example of an approximate line obtained by the approximate line calculation unit of the embodiment. Fig. 3 is a flowchart showing an operation example of the cash replenishment plan support system of the first embodiment. Fig. 4 is a block diagram showing an example of the functional configuration of a cash replenishment plan support system according to the second embodiment. Fig. 5 is a view for explaining a correction process of the planned vehicle number plan performed by the deployment vehicle plan correction unit of the second embodiment. [Description of main component symbols] 1 0 ' 20 : Cash replenishment plan support system 1 1 : Transaction information acquisition unit 12 : Difference integration unit 1 3 : Approximation line calculation unit 1 4 : Required fund prediction unit 1 5 : Supplementary plan Output department
2 1 :調配車輛數計畫部 22 :調配車輛計畫修正部 100 ： ATM 200 :通信網路 -23-2 1 : Dispatch vehicle number planning department 22 : Dispatch vehicle plan correction unit 100 : ATM 200 : Communication network -23-