JP2004323162A - Maintenance business simulator, maintenance service planning support system and maintenance business support system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To support planning of maintenance service of which customer satisfaction is high, and to effectively manage maintenance business. <P>SOLUTION: This maintenance business support system is provided with a maintenance business simulation part for calculating a probability distribution characteristics of an evaluation index regarding the maintenance business. The maintenance business support system is further provided with a business plan making support part for supporting making of a maintenance business plan and a business support part 5 for giving support information with respect to business action with the usage of output of the maintenance business simulation part. The maintenance business support system is furthermore provided with a data collecting system 7 for collecting maintenance work results, apparatus operation results and customer data, and a part life analysis part 8 for analyzing life distribution of apparatus parts by the collected data. Further, the maintenance business support system is provided with a use terminal 16 and a data collecting terminal 17 in a supplier 30 and an own factory 25. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

図１０に、保全作業シミュレータ２４の処理フローを示す。
【０１１３】
上述したように、本実施形態では、モンテカルロ法を用いる。本実施形態のシミュレーションの基本的な処理の流れは、部品寿命を擬似乱数列を用いて発生させ、保全作業の回数をカウントするというものである。
【０１１４】
なお、以下の処理は、メモリに格納された予め定められたプログラムに従って、保全作業シミュレータ２４が行なう処理である。
【０１１５】
保全作業シミュレータ２４は、起動すると、ステップＳ２において、指定された顧客の所有する昇降機の機種、その台数、各昇降機の構成部品、各構成部品の寿命の確率分布関数、各構成部品の予防交換インターバルとを含むデータを、それぞれ、データベースシステム１０の顧客データベース１２、製品仕様データベース１１、保全実績データベース１５などから取得し、メモリに格納する。
【０１１６】
続くステップＳ３において、メモリに格納されている作成した標本関数の数と、入力を受け付けてメモリに格納されている生成標本関数の数とを比較し、所望の本数だけ標本関数を生成したかを判定し、判定結果をメモリに格納する。そして、メモリに格納された判定結果を読み出し、［Ｙｅｓ］であればステップＳ４へ、［Ｎｏ］であればステップＳ１４へ進む。
【０１１７】
ステップＳ４では、シミュレーション時刻ｔ，作業回数カウンタ
_ｉＮＷＰ^ｊ _ｋ（ｔ_ａ）、_ｉＮＷＢ^ｊ _ｋ（ｔ_ａ）を０で初期化し、メモリに格納する。ここで、シミュレーション時刻ｔはある時間刻みを用いることにより，整数値をとるように離散化されているものとする。
【０１１８】
続くステップＳ５では、それぞれの部品_ｉＰ^ｊ _ｋの寿命_ｉτ^ｊ _ｋを確率密度関数_ｉｆ^ｊ _ｋ（_ｉτ^ｊ _ｋ）に従う擬似乱数を用いて、部品寿命を決定し，更に決定された寿命を前記時間刻みを用いて離散化し、メモリに格納する。
【０１１９】
次に、メモリに格納された部品寿命を用いて，部品_ｉＰ^ｊ _ｋが寿命に到達する時刻_ｉｔｌ^ｊ _ｋ＝ｔ＋ _ｉτ^ｊ _ｋを算出し、メモリに格納する。
【０１２０】
続くステップＳ６において，現シミュレーション時刻ｔにおいて交換作業が発生する（ｔ＝_ｉｔｌ^ｊ _ｋ，もしくは，部品取付から予防交換インターバルだけ時間が経過）部品が存在するかどうかを、メモリに格納された部品寿命と、現シミュレーション時刻ｔとを比較することで判定し、判定結果および交換作業が発生した部品を特定する情報をメモリに格納する。
【０１２１】
そして、メモリに格納された判定結果を読み出し、［Ｙｅｓ］の場合にはステップＳ７へ，［Ｎｏ］であればステップＳ１１へ進む。
【０１２２】
ステップＳ７では，メモリに格納された現シミュレーション時刻ｔが評価指標集計期間内かどうかを、メモリに格納された評価指標集計期間の終わりの時刻と比較することで判定し、判定結果をメモリに格納する。メモリに格納された判定結果を読み出し、［Ｙｅｓ］であればステップ８，［Ｎｏ］であればステップＳ１０へ進む。
【０１２３】
ステップＳ８では、ステップＳ６において交換作業が発生すると判定された部品をメモリから読み出し、読み出した部品_ｉＰ^ｊ _ｋが寿命に達する時刻_ｉｔｌ^ｊ _ｋを集計時間刻みｄｔ_ａを用いて離散化し（下記数式２）、メモリに格納し、ステップＳ９へ進む。ここで、［ｘ］⁻は、ガウス記号であり，ｘを超えない最大の整数値を表す。
【０１２４】
【数２】

続くステップＳ９では，ステップＳ６において交換作業が発生すると判定された部品_ｉＰ^ｊ _ｋと現シミュレーション時刻ｔとをメモリから読み出し、両者を比較することで、当該部品に対して行われる交換作業の種別（予防交換もしくは事後交換）を判定し、メモリに格納された該当作業回数のカウンタ（_ｉＮＷＰ^ｊ _ｋ（ｔ_ａ）もしくは_ｉＮＷＢ^ｊ _ｋ（ｔ_ａ））を読み出して、１だけ増加し、メモリに再度格納し、ステップＳ１０へ移行する。
【０１２５】
ステップＳ１０では、ステップＳ６において交換作業が発生すると判定された部品をメモリから読み出し、当該部品_ｉＰ^ｊ _ｋに対して新たな部品寿命_ｉτ^ｊ _ｋを設定し、寿命到達時間_ｉｔｌ^ｊ _ｋを算出し，メモリに格納し、ステップＳ６に移行する。
【０１２６】
ステップＳ１１では、メモリから読み出したシミュレーション時刻ｔを１だけ増加し、再度メモリに格納し、ステップＳ１２に進む。
【０１２７】
続くステップＳ１２では、シミュレーション時刻ｔと利用者により設定されたシミュレーション期間とをメモリから読み出し、両者を比較することで、シミュレーション時刻ｔが利用者により設定されたシミュレーション期間を超えたかどうかを判定し、その結果をメモリに格納する。判定結果をメモリから読み出し、［Ｙｅｓ］の場合にはステップＳ１３へ、［Ｎｏ］の場合にはステップＳ６へ移行する。
【０１２８】
続くステップＳ１３では、それぞれの部品に対する作業回数_ｉＮＷＰ^ｊ _ｋ（ｔ_ａ）、_ｉＮＷＢ^ｊ _ｋ（ｔ_ａ）に応じて下記の数式３をメモリから読み出し、１だけ増加して再度メモリに格納し、ステップＳ３に進む。
【０１２９】
【数３】

ステップＳ１４では保全作業カウンタをメモリから読み出し、同じくメモリから読み出した作成した標本関数の本数の値で割り、保全作業回数_ｉＮＷＰ^ｊ _ｋ、_ｉＮＷＢ^ｊ _ｋの確率関数の下記数式４を算出し、メモリに格納する。
【０１３０】
【数４】

続くステップＳ１５では、必要に応じて、メモリに格納された保全作業回数を読み出し、データベースシステム１０から読み出してメモリに格納された下記の各データを用いて、保全コスト、保全工数の確率分布や、期待値等の評価指標を算出し、メモリに格納する。
【０１３１】
ここで、評価指標算出式の一例を以下に示す。

続くステップＳ１６において、ステップＳ１５で算出した結果をメモリから読み出し、シミュレーション結果ファイル４に出力し、また、必要に応じて、営業支援部５、事業計画立案支援部３に受け渡す、ＧＵＩ部１６−２の表示画面に出力するなどして終了する。
【０１３２】
このように、本実施形態の保全事業シミュレーション部２は、この保全作業シミュレータ２４の処理により様々な評価指標の出力が可能である。これらの出力を利用して、営業活動や事業計画の立案を支援することができる。
【０１３３】
なお、上述の保全事業シミュレーション部２は、保全作業に関する評価指標（部品交換回数、コスト、作業量等）を算出する保全作業シミュレータ２４から構成される比較的簡易なモデルを用いたものである。
【０１３４】
保全事業シミュレーション部２はこれに限られない。その他の例を図２（ｂ）および図２（ｃ）に示す。
【０１３５】
図２（ｂ）に示した保全事業シミュレーション部２は、図２（ａ）のものを発展させたものであり、保全作業シミュレータ２４に加え、ロジスティクス・シミュレータ２５および間接作業シミュレータ２６を有する。
【０１３６】
ロジスティクス・シミュレータ２５、間接作業シミュレータ２６は、それぞれ、入力データ２７に加えて保全作業シミュレータ２４の結果を入力とし、部品物流に関する評価指標（在庫量、配送量等）、間接業務に関する評価指標（報告書作成工数等）を算出するものであり、保全作業シミュレータ２４からの一方向の連成のみを考慮したモデルである。
【０１３７】
ここで、本実施形態の保全事業シミュレーション部２における「連成」とは、それぞれのシミュレータによるシミュレーション結果が互いに影響を及ぼすことを言う。例えば、欠品などにより部品が配送されなかったり、部品発注業務などの間接業務をこなす人員が不足したりすると、予定通り作業が行なわれなくなり、この結果、故障が発生し、部品の緊急手配、輸送が必要となる、といった例などである。
【０１３８】
図２（ｃ）に示した保全事業シミュレーション部２は、図２（ｂ）のものをさらに発展させたものであり、保全作業シミュレータ２４、ロジスティクス・シミュレータ２５、間接作業シミュレータ２６の相互の連成を考慮したモデルを用いたものである。
【０１３９】
図２（ｂ）、（ｃ）で用いられるロジスティクス・シミュレータ２５、間接作業シミュレータ２６は、保全作業シミュレータ２４で算出した保全作業発生回数の確率分布を入力として用いたモンテカルロ法等により実現することができる。
【０１４０】
図２（ｂ）、図２（ｃ）や、保全作業シミュレータ２４、ロジスティクス・シミュレータ２５、間接作業シミュレータ２６等を別個に利用する場合も同様に考えられる。
【０１４１】
次に部品寿命解析部９の詳細について説明する。
【０１４２】
図１１は部品寿命解析部９の構成図である。本図に示すように、部品寿命解析部９は、部品解析エンジン３２と部品寿命解析ナレッジデータベース３３とを備える。
【０１４３】
部品寿命解析エンジン３２は、与えられたデータを基に部品寿命分布を算出する解析エンジンであり、部品寿命解析ナレッジＤＢ３３は、解析時に用いる部品寿命解析に関するナレッジを格納するものである。
【０１４４】
次に、部品寿命解析エンジンの処理動作を説明する。
【０１４５】
図１２は、部品寿命解析エンジン３２の処理フロー図である。
【０１４６】
なお、下記の処理は、予め定められたプログラムに従って、記憶部に記憶されたデータを用いて、ＣＰＵが行なうものである。
【０１４７】
部品寿命解析エンジン３２は、実行の指示を受け付けると、ステップＳ２１において、部品寿命解析に関するナレッジを部品寿命解析ナレッジＤＢ３３より取得し、記憶部に記憶する。
【０１４８】
ここで、部品寿命解析に関するナレッジとは、「温泉の近くで使用される部品と、通常の環境で使用される部品は分けて解析」、「部品Ｘは累積走行時間ではなく、使用年数で寿命が決まる」、「部品Ｙの寿命はワイブル分布に従う」といった類のものであり、部品寿命解析時に利用される。
【０１４９】
次のステップＳ２２では、部品寿命試験データ（製品開発時試験データ、回収した部品の余寿命試験データ等）や、故障実績データ等の解析元データをデータベースシステム１０より取得し、記憶部に記憶する。
【０１５０】
次のステップＳ２３では、ステップＳ２２で取得した解析元データとステップＳ２１で取得した部品寿命解析に関するナレッジとを読み出し、解析元データをナレッジに従い取捨選択を行い、その結果を記憶部に記憶する。この時、部品寿命ではなく顧客起因の要因（いたずら等）により故障となった故障データも取り除く。
【０１５１】
ステップＳ２４では、ステップＳ２３で選別された元データを記憶部から読み出し、読み出したデータを用いて、部品寿命解析を行い、その結果を記憶部に記憶する。
【０１５２】
なお、部品寿命解析については、ワイブル・ハザード解析等数多くの手法が知られ、これらの手法を用いて行なう。
【０１５３】
最後のステップＳ２５において、ステップＳ２４で得られた部品寿命解析結果を記憶部から読み出し、製品使用ＤＢに格納して終了する。
【０１５４】
以上、部品寿命解析エンジン３２の処理を説明した。
【０１５５】
以上説明したように、本実施形態の保全事業支援システムによれば、保全事業シミュレータ２で算出した評価指標の確率分布のデータを利用することにより、個々の顧客の昇降機の稼動の特性を抽出することが容易になる。そして、抽出した特性に基づいて、個々の顧客に応じた保全サービスの立案を支援するための情報を出力することができる。
【０１５６】
また、シミュレーション結果を視覚化して出力することにより、立案した保全サービスの特徴を顧客に対して定量的に説明することを支援する。これらは、料金体系の明確化、簡易なＳＬＡ（サービスレベルアグリーメント）の設定を可能とする。
【０１５７】
また、本実施形態によれば、営業活動を行なうべき顧客を効率的に抽出することを支援し、効果的な顧客へのアプローチ時期を決定することを支援できる。これらは、営業活動の効率化につながる。
【０１５８】
さらに、本実施形態の保全事業支援システムによれば、上述のように営業活動の効率化とともに、事業計画立案支援部３により事業計画の立案を行なうことができる。これにより、保全事業計画の無理／無駄を省くことができる。
【０１５９】
また、サプライヤ３０、自社工場４０とデータを共有することにより、サプライや３０や自社工場４０の生産／配送計画システムと連携して製品開発の支援、部品生産／配送計画の立案支援が可能となる。
【０１６０】
本実施形態によれば、保全事業に関するシミュレータを核とした保全事業支援システムにより、従来は経験に基づく情報に頼っていた業務に対して、シミュレータを用いて論理的な手段により算出された精度の高い支援情報を提供することができ、これにより、保全事業全体の効率的な運営が可能となる。
【０１６１】
なお、本実施形態は、昇降機の保全事業を例にあげて説明をしたが、本発明を適用可能な保全対象は昇降機に限られない。販売後、所定期間保守管理の必要な製品全般の保全事業に適用可能である。
【０１６２】
【発明の効果】
個々のユーザの要求に応じた保全サービスの立案を支援することにより、当該サービスの立案を容易にし、顧客満足度を高めるとともに、事業者にとっても効率のよい保全事業の運営を実現する。
【図面の簡単な説明】
【図１】図１は、本実施形態を適用した保全事業支援システムの全体構成を説明するための図である。
【図２】図２は、本実施形態の保全事業シミュレーション部の機能構成図である。（ａ）は基本的な保全事業シミュレータの機能構成図であり、（ｂ）は、第１の発展例、（ｃ）は第２の発展例である。
【図３】図３（ａ）〜（ｆ）は、それぞれ、本実施形態の保全事業シミュレーション部の出力イメージ図である。
【図４】図４は、サプライヤおよび自社工場の本実施形態に関する機能を抽出した図である。
【図５】図５は、本実施形態の営業支援部の処理フローである。
【図６】図６（ａ）〜（ｃ）は、それぞれ、本実施形態の営業支援部の出力イメージ図である。
【図７】図７は、本実施形態の保全作業の概念を説明するための図である。
【図８】図８は、本実施形態の保全事業シミュレーション部の処理フローである。
【図９】図９は、本実施形態の保全作業シミュレータのシミュレーション対象の概念を説明するための図である。
【図１０】図１０は、本実施形態の保全作業シミュレータの処理フローである。
【図１１】図１１は、本実施形態の部品寿命解析部の機能構成図である。
【図１２】図１２は、本実施形態の部品寿命解析部の処理フローである。
【符号の説明】
１・・・意思決定支援システム、２・・・保全事業シミュレーション部、３・・・事業計画立案支援部、４・・・シミュレーション結果ファイル、５・・・営業支援部、６・・・実績データ収集／解析システム、７・・・稼動実績収集部、８・・・保全実績収集部、９・・・部品寿命解析部、１０・・・データベースシステム、１１・・・製品仕様ＤＢ、１２・・・顧客ＤＢ、１３・・・保全体制ＤＢ、１４・・・稼動実績ＤＢ、１５・・・保全実績ＤＢ、１６−１、１６−２・・・ＧＵＩ部、１６−３、１６−４・・・利用端末、１７−１・・作業実績収集端末、１７−２・・・顧客データ収集端末、１９−１、１９−２、１９−３、１９−４、１９−５、１９−６・・・通信処理部、２０−１、２０−２、２０−３、通信回線、２１−１、２０−２・・・顧客ビル、３０・・・サプライヤ、４０・・・自社工場、２４・・・保全作業シミュレータ、２５・・・ロジスティクス・シミュレータ、２６・・・間接作業シミュレータ、２７・・・入力データ、２８・・・出力データ、３２・・・部品寿命解析エンジン、３３・・・部品寿命解析ナレッジＤＢ、３４・・・生産／配送計画システム[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a technology for supporting a maintenance business of equipment such as an elevator.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, there is a support technology for planning a business strategy as a maintenance business of equipment such as an elevator. For example, when the failure characteristics of component parts of a device are well known, a method for deriving a maintenance method that minimizes the total cost required for maintenance activities according to the failure characteristics has been proposed (Non-Patent Document 1). reference).
[0003]
Furthermore, there has been proposed a method of drafting a maintenance plan for a plurality of elevators, aiming at overall optimization as a base for performing a maintenance business from the viewpoint of costs and man-hours (see Non-Patent Document 2).
[Non-patent document 1]
"Study on Optimal After-Service Strategy for Infrastructure Equipment (1st Report)", Yamashinamoto et al., Journal of Precision Engineering, p1371-1377, Vol. 66, no. 9, 2000
[Non-patent document 2]
"Study on Optimal After-Service Strategy for Infrastructure Equipment (2nd Report)", Yamashinamoto et al., Journal of the Japan Society of Precision Engineering, p1233-1238, Vol. 67, no. 8, 2001
[0004]
[Problems to be solved by the invention]
The methods described in these documents determine the probability that work is performed on each component from the probability distribution of the life of each component (hereinafter referred to as the work probability), and the maintenance required based on the work probability. This is to calculate evaluation indices such as the amount of work, cost, and customer loss due to equipment failure, and to formulate an optimal business plan based on the evaluation indices. These focus on the average value (expected value) of the evaluation indices, with the aim of optimizing from the perspective of the maintenance business provider, and for each customer, the above method was drafted. Maintenance services will be applied uniformly.
[0005]
However, in the case of elevators, for example, hospitals want to reduce the number of failures even if the maintenance cost increases, or apartments want to reduce the maintenance cost even if the number of failures increases slightly. different.
[0006]
The above method is insufficient to plan an optimal maintenance service for each user having different needs, and cannot plan a maintenance plan directly related to user satisfaction. For this reason, a maintenance method that is considered to be optimal for each user is determined exclusively based on experience and intuition.
[0007]
In view of such a situation, the present invention outputs information useful for planning a maintenance service according to each user, thereby supporting the planning of the service, increasing customer satisfaction, and improving the business satisfaction for the business operator. The aim is to realize efficient operation of maintenance business.
[0008]
[Means for Solving the Problems]
The maintenance business simulator of the present invention is a probability distribution calculating means for calculating a probability distribution of a combination of the number of maintenance operations of each component by using a component life of each component constituting the device to be maintained and a predetermined maintenance method. Is provided.
[0009]
Further, the maintenance business support system of the present invention is a maintenance business simulator that calculates a probability distribution of an evaluation index related to a maintenance business, and a support system that supports the planning of various maintenance services or business plans using the calculated probability distribution. And
[0010]
The maintenance business simulator calculates a probability distribution of a combination of the number of maintenance operations of each component constituting the maintenance target device.
[0011]
A component life generating means for generating a component life of each of the components by pseudo-random numbers according to a predetermined probability distribution function; a component life generated by the component life generating means and a predetermined replacement time of each of the components; Based on the above, the number of times the component is replaced (post-replacement) due to the occurrence of the failure and the number of times the component is replaced (preventive replacement) due to the replacement time are counted. And a probability distribution calculating means for calculating a probability distribution of a combination of the number of times of each of the parts in the preventive replacement.
[0012]
Further, the maintenance business simulator calculates the probability distribution of the evaluation index including the number of parts required for maintenance work, the cost, and the number of man-hours from the probability distribution by using information on the characteristics of the parts held in advance. An evaluation index calculating unit is further provided. On the other hand, the support system calculates useful information when planning the maintenance service using the probability distribution of the evaluation index calculated by the maintenance work simulator. Support the planning of maintenance services to be provided to customers who own the target equipment.
[0013]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0014]
FIG. 1 is a diagram for explaining the overall configuration of a maintenance business support system for an elevator to which the present embodiment is applied.
[0015]
As shown in the figure, the elevator maintenance business support system simulates a maintenance business, and based on the result, a decision support system 1 that supports the planning of the maintenance business and the proposal of a maintenance service according to each user, A performance data collection / analysis system 6 for acquiring and analyzing performance data of a maintenance target elevator directly from each elevator or by input from a maintenance worker, and holding various data necessary for maintenance business planning. A database system 10 is provided.
[0016]
The decision support system 1, the database system 10, and the performance data collection / analysis system 6 are connected via an intranet 20-3. The intranet 20-3 is connected to the supplier 30 and the company factory 40, and the GUI unit 16-1 and the collection terminal 17-1 are connected via the communication line 20-2.
[0017]
Further, the operation result collection terminals 18-1 and 18-2 of each elevator are connected to the result data collection / analysis system 8 via the communication line 20-1 to acquire data relating to the actual operation of the elevator.
[0018]
Note that the above network configuration is not limited to this diagram, and the communication lines 20-1 and 20-2 may be a telephone line, the Internet, or the like, but are not limited to this as long as data can be transmitted and received.
[0019]
The decision support system 1 includes a maintenance business simulation unit 2, a business plan formulation support unit 3, a simulation result file 4, a sales support unit 5, and a user from the user of the decision support system 1 connected by a bus line. It includes a GUI unit 16-2 including an input device for receiving input data and an output device for presenting a result to a user, and a communication processing unit 19-1 as an interface with the intranet 20-3.
[0020]
The maintenance business simulation unit 2 performs a simulation for calculating useful information for supporting planning of various maintenance services to be provided to customers and business planning. Specifically, a probability distribution such as a combination of the number of maintenance operations and a cost is calculated and output from the life distribution of each component constituting the elevator.
[0021]
FIG. 2 is a diagram showing a functional configuration of the maintenance business simulation unit 2.
[0022]
As shown in FIG. 2A, the maintenance business simulation unit 2 includes a maintenance work simulator 24.
[0023]
The maintenance work simulator 24 simulates the probability distribution of the combination of the number of maintenance work by a predetermined method based on the input data 27, calculates the probability distribution of the evaluation index in consideration of necessary data, and outputs the output data 28 Is output as
[0024]
Here, the input data 27 is a simulation condition and a simulation target that have been input through the sales support unit 5 or the business plan planning support unit 3, and various data stored in the database system 10. The simulation conditions include the number of sample functions to be generated, the simulation period, the simulation time interval, the aggregation period, the time interval at the time of aggregation, the desired evaluation index, and various data stored in the database 10 for the user to perform the simulation. This is the data newly given to. In the present embodiment, a Monte Carlo simulation is used. Desired evaluation indices include the number of parts required, the cost of parts, the number of man-hours, and the cost of work. The simulation target is a customer or a model.
[0025]
The maintenance work simulator 24 changes the life of the parts constituting the product to be subjected to the maintenance work using the pseudo-random number, and, according to a predetermined simulation procedure, calculates the combination of the number of maintenance work that occurs during the designated totalization period. Find the probability distribution. From the probability distribution of the determined combination of the number of maintenance operations, the probability distribution of various evaluation indices such as the cost required for maintenance using various data stored in the database system 10 in accordance with the specified evaluation index. Is calculated and output as output data 28. Details of the simulation procedure will be described later.
[0026]
The maintenance business simulation unit 2 stores the numerical data of the calculated result in the simulation result file 4. Here, in order to make it easy for the user to make various kinds of decisions, it is also possible to configure so as to output to the display screen of the GUI unit 16-2 as illustrated in FIGS. 3 (a) to 3 (f).
[0027]
Here, an output image example of the maintenance business simulation unit 2 is shown in FIG.
[0028]
FIG. 3 (a) shows the maintenance parameters (replacement cycle, inspection cycle, number of warehouse bases, number of personnel, etc.) on the horizontal axis, and the maintenance cost, the number of failure occurrences, etc. on the vertical axis. It is intended to support preliminary examination at times.
[0029]
FIG. 3 (b) confirms time-series changes such as maintenance costs and the number of failure occurrences.
[0030]
In FIG. 3 (c), parameters relating to the customer (running time, number of floors, etc.) are plotted on the horizontal axis, and maintenance costs, the number of failure occurrences, etc. are plotted on the vertical axis.
[0031]
FIG. 3D is a graph of the probability distribution of the obtained evaluation index. This is used when considering not only the expected value but also the variance. The output of the probability distribution of the evaluation index can serve as basic data having a high degree of accuracy when, for example, a service level agreement (SLA) contract that specifically guarantees a service level is devised in the sales support unit 5 described later. Details of the calculation method will be described later.
[0032]
FIG. 3 (e) shows the amount of parts transported between each warehouse / sales office, and is used when confirming the transportation cost when examining the maintenance system or the like.
[0033]
FIG. 3 (f) shows the stock of parts at each warehouse / sales office, and is used when checking inventory costs or the like when examining a maintenance system or the like.
[0034]
The simulation result file 4 stores the simulation result of the maintenance business simulation unit 2.
[0035]
The sales support unit 5 causes the maintenance business simulation unit 2 to perform a necessary simulation in accordance with the instruction received via the GUI unit 16-2, and adds necessary data to the result, and displays the result on the display unit of the GUI unit 16-2. , The information necessary for planning the maintenance service to be provided to the customer is presented to the planner.
[0036]
According to the data of the probability distribution of the evaluation index obtained by the maintenance business simulation unit 2, it is easy to extract the operating characteristics of the elevator of each customer. The sales support unit 5 uses this data to assist in drafting various maintenance service proposals suitable for individual customers.
[0037]
The sales support unit 5 has each function of an SLA planning support function 51, a customer proposal time study support function 52, and an investigation required customer extraction function 53. It is not necessary for the sales support unit 5 to have all of these functions, and it is sufficient that at least one of them is provided.
[0038]
Next, the function of the SLA planning support function 51 will be described. Here, for example, there is a maintenance contract that defines a service level such that the number of failures per year is reduced to a predetermined number or less, and a penalty is paid if the number of failures exceeds that. When proposing such a contract to a customer, it is necessary to determine the probability of complying with the SLA and examine profitability.
[0039]
The SLA planning support function 51 calculates the above-described SLA compliance rate and the like from the simulation result of the number of failures of the model owned by each customer and the maintenance results, and displays the result on the output device of the GUI unit 16-2. . This provides support for the SLA contract formulator. SLA targets include not only the number of failures, but also recovery time, maintenance staff arrival time, and the like.
[0040]
The materials created by this function can be used for customer presentations and the like.
[0041]
The customer suggestion time study support function 52 calculates the average value of the evaluation index (for example, the number of failures) of the equipment owned by each customer for each specified aggregation period, and outputs the result to the output device of the GUI unit 16-2. To be displayed. This will help determine when to approach customers to sign new maintenance contracts.
[0042]
The investigation required customer extraction function 53 calculates the probability distribution of the general evaluation index of the model owned by each customer and the maintenance work performance, and displays them on the output device of the GUI unit 16-2. This assists in determining whether a survey is needed for the customer. The detailed processing flow of each function will be described later.
[0043]
The business plan drafting support unit 3 makes the maintenance business simulation unit 2 perform a necessary simulation according to the instruction received via the GUI unit 16-2, and displays the result on the display device of the GUI unit 16-2. , And present the information necessary for planning a conservation project.
[0044]
The business plan drafting support unit 3 optimizes a plan required for the maintenance business by causing the maintenance business simulation unit 2 to repeatedly perform simulation by changing input parameters, and for example, drafts a maintenance work schedule. .
[0045]
In addition, when an instruction for presenting information necessary for drafting an inventory / delivery plan is received, the maintenance business simulation unit 2 calculates the required component amount required for the planning by changing conditions and presents the result. . When an instruction to present information necessary for drafting a personnel plan (employment, allocation, etc.) is received, conditions are changed, the maintenance work amount is calculated by the maintenance business simulation unit 2, and the result is presented. When an instruction to present information necessary for planning a management plan (such as an investment plan) is received, the maintenance business simulation unit 2 calculates costs and profits and presents the results.
[0046]
A person who intends to draft a business plan can proceed with drafting a business plan by viewing the various results described above displayed on the display device of the GUI unit 16-2.
[0047]
These can be realized using the method proposed in Non-Patent Document 2 and the like.
[0048]
The above-described decision support system 1 generally includes an input / output unit as an interface for exchanging data, an arithmetic unit including a CPU for performing arithmetic processing, and a storage unit including a memory for storing data and the like and a hard disk. Is built on a typical information processing device.
[0049]
The maintenance business simulation unit 2, the business plan drafting support unit 3, and the business support unit 5 are realized by the CPU executing a predetermined program stored in the storage unit in advance. The simulation result file 4 uses a memory or a hard disk.
[0050]
In the present embodiment, the above-described units constituting the decision support system 1 will be described by using an example in which the units are physically constructed on the same information processing apparatus. However, the units are not necessarily constructed on the same information processing apparatus. do not have to. The information processing apparatus may be constructed by being distributed to a plurality of information processing apparatuses.
[0051]
The performance data collection / analysis system 6 includes an operation performance collection unit 7, a maintenance performance collection unit 8, a component life analysis unit 9, and an input device for inputting necessary data and a user. And a communication processing unit 19-2, which is an interface between the GUI unit 16-3 and the communication line 20-1, and a communication processing unit 19-3, which is an interface with the intranet 20-3. And
[0052]
The operation result collection unit 7 operates the elevator such as traveling time, door opening / closing frequency, and energization time collected by the information collection terminals 18-1 and 18-2 installed in the elevator, the ambient temperature of the elevator, and the salt content. Data relating to the use environment such as sulfur and sulfur content is acquired and stored in a predetermined database such as the customer database 12 and the operation result database 14 of the database system 10, respectively.
[0053]
The maintenance result collection unit 8 performs maintenance work results such as an inspection / maintenance date, a preventive replacement date, a failure history, failure data, and a collected part status input on the information collection terminal 17-1 carried by a maintenance worker or a salesperson. And maintenance business results such as the use status of elevators such as customer names, charges, and the number of times of mischief are acquired, and stored in predetermined databases such as the customer database 12 and the maintenance result database 15 of the database system 10, respectively.
[0054]
The component life analysis unit 9 calculates the life distribution of each component constituting the elevator from the actual data stored in the database system 10 and the like. Details such as the analysis method will be described later.
[0055]
Next, the details of the supplier 30 and the company factory 40 will be described. FIG. 4 is a diagram in which portions related to the present embodiment are extracted.
[0056]
As shown in the figure, the supplier 30 and the company factory 40 respectively include an information collection terminal 17-2, a GUI unit 16-4 that receives inputs related to the maintenance business support system and displays necessary outputs, And a production / delivery planning system 34 for supporting the planning of a delivery plan and a delivery plan, and a communication processing unit 19-6 as an interface with the intranet 20-3.
[0057]
The supplier 30 and the in-house factory 40 support the maintenance business, such as product life test results such as product specification data, numerical data and knowledge data, and component manufacturing / delivery lead time, at the time of product development via the information collection terminal 17-2. Various kinds of data necessary for the operation are collected and stored in a predetermined database of the database system 10 via the intranet 20-3.
[0058]
Further, the supplier 30 and the company factory 40 can also acquire component life analysis data, failure data, failure data, and the like from the database system 10 via the GUI unit 16-4. The acquired data can be used for the next product development.
[0059]
The database system 10 holds data necessary for the system, such as a product specification database 11, a customer database 12, a maintenance system database 13, an operation results database 14, and a maintenance results database 15.
[0060]
The product specification database 11 stores data related to product specifications, such as product configuration, component life, man-hours, and unit price.
[0061]
The customer database 12 stores customer name, address, delivery information such as products, data on customers such as usage environment of products, and data on usage environment of elevators which should be noted for each customer.
[0062]
The maintenance system database 13 stores data on the maintenance system on the business side, such as the location of a sales office, the number of workers, and the location of a warehouse.
[0063]
The operation result database 14 stores data relating to the operation of the elevator, which is collected via the operation result collection unit 7, such as the running time, the number of times of opening and closing the door, and the energizing time.
[0064]
The maintenance performance database 15 stores data relating to the performance of elevator maintenance, such as preventive replacement dates, inspection / maintenance dates, early warning information, and failure histories, which are input by a worker at the site via a collection terminal.
[0065]
Next, the processing operation of the sales support unit 5 in the decision support system 1 will be described with reference to the drawings.
[0066]
FIG. 5 shows the processing flow of the sales support section 5 together with the processing of the maintenance business simulation section 2. The following process is a process executed according to a predetermined program stored in a memory or the like.
[0067]
The sales support unit 5 receives the selection of the processing content by the user via the GUI unit 16-2 and stores the selection in the memory (step S501). The contents of the process are selected from processes that can be performed by the sales support unit 5, such as SLA planning support, customer proposal time study support, and investigation of customers requiring investigation.
[0068]
Upon receiving an instruction for SLA planning support, the SLA planning support function 51 uses the input items required to create the SLA planning support material, such as a desired evaluation index, a simulation condition, a simulation target, and a compliance rate of the SLA to be displayed ( A screen that accepts input of information such as multiple input is read from the memory and presented to the GUI unit 16-2 (step S502).
[0069]
Here, the simulation conditions are the number of generated sample functions, the simulation period, the simulation time interval, the aggregation period, and the aggregation time interval, and the simulation target is the name of the SLA drafting customer. In addition, here, as an evaluation index, the number of failures of an elevator held by a planning target customer will be described as an example.
[0070]
When the input of the above conditions is received through the input screen, the conditions are stored in a memory, and the SLA planning support function 51 outputs, from the received information, information necessary for executing the simulation, that is, a desired evaluation index, a simulation condition, and a simulation target. Is transmitted to the maintenance business simulation unit 2 (step S503).
[0071]
The maintenance business simulation unit 2 acquires necessary data from the database system 10 and stores it in the memory, and also stores the data received in step S503 in the memory. Then, using the data stored in the memory, the maintenance work simulator 24 calculates the probability distribution of the evaluation index of the customer to be simulated (step S504), stores the probability distribution in the simulation simulation result file 4, and stores it in the sales support section. 5 (step S505).
[0072]
The SLA planning support function 51 calculates the number of failures that achieve the received SLA compliance rate based on the probability distribution of the received evaluation index (here, the number of failures), stores the number of failures in the memory (step S506), and The number of failures achieving the compliance rate is superimposed on the probability distribution of the number of times and output (step S507). The output is displayed on a display device of the GUI unit 16-2, for example.
[0073]
Here, an example of the output is shown in FIG. In the figure, the x-axis direction is a simulation target, and the y-axis direction is a probability. The value that achieves the SLA compliance rate input here is superimposed.
[0074]
When planning an SLA, it is necessary to determine the probability of compliance with the SLA and examine the profitability. However, the user can determine the relationship between the predetermined SLA compliance rate and the number of failures by looking at the output results in this diagram. It can be easily grasped. For example, assuming that the predetermined compliance rate is 80%, the number of failures can be easily determined to be 5 times / year from the output result of this figure, and the customer is assured that the number of failures is 5 times or less per year. If SLA is proposed, it can be determined that it can be complied with an 80% probability.
[0075]
In step 501, when the selection of the customer proposal time study support is received, the customer proposal time study support function 52 performs an input item necessary for creating a material for supporting the customer proposal time study, for example, a desired evaluation index. Then, a screen for accepting input of information such as a simulation condition, a simulation target, and a reference change amount is displayed on the GUI unit 16-2 (step S502).
[0076]
In the following processing, the customer suggestion time study support function 52 stores the result obtained by the processing of each step in the memory, and uses the result in the subsequent processing.
[0077]
Here, the simulation conditions and objects are the same as those for the SLA planning support. In addition, various evaluation indexes, such as cost and work amount, can be considered. Here, the number of failures will be described as an example. The reference change amount is given by the determination time and the reference amount that changes during that time.
[0078]
When the input of the above conditions is received via the input screen, the customer suggestion time study support function 52 outputs, from the received information, information necessary for executing the simulation, that is, the desired evaluation index, the simulation condition, and the information of the simulation target. It is sent to the maintenance business simulation unit 2 (step S503).
[0079]
After the processing in steps S504 and S505, the customer proposal time study support function 52 calculates the average value of the number of failures for each aggregation time based on the probability distribution of the received evaluation index (here, the number of failures). (Step S506), the change amount (judgment time and criterion) is superimposed and output to, for example, the display device of the GUI unit 16-2 (Step S507).
[0080]
Here, an example of the output is shown in FIG. This drawing extracts the approach time to the customer from the time series data of the evaluation index such as the number of failures, cost, and profit calculated using the maintenance business simulation unit 2. In the figure, at a predetermined determination time, the time at which the evaluation index is larger than the predetermined determination criterion is extracted as “rate revision proposal time”, “renewal proposal time”, and the like.
[0081]
Such a quantitative output result not only contributes to the decision at the time of planning, but also as a persuasive material that shows the legitimacy of the maintenance fee when proposing and explaining maintenance service products to customers. Can be presented, leading to a transparent fee structure.
[0082]
Further, in step 501, when the selection of the customer requiring investigation is received, the customer-exquiring-requiring function 53 causes the customer-requiring-customer extracting function 53 to input items necessary for creating a material from which the customer requiring investigation can be extracted, for example, a desired evaluation. A screen for accepting input of information such as an index, a simulation condition, a simulation target, and a necessity determination criterion is displayed on the GUI unit 16-2 (step S502).
[0083]
In the following processing, the investigation required customer extraction function 53 stores the result obtained by the processing of each step in the memory, and uses it in the subsequent processing.
[0084]
Here, the simulation conditions and objects are the same as those for the SLA planning support. In addition, various evaluation indexes, such as cost and work amount, can be considered. Here, the number of failures will be described as an example.
[0085]
When the input of the above conditions is received through the input screen, the customer-required-for-investigation extraction function 53 maintains the information necessary for executing the simulation among the received information, that is, the desired evaluation index, the simulation condition, and the information of the simulation target. This is sent to the business simulation unit 2 (step S503).
[0086]
After the processing of steps S504 and S505, the customer-in-survey extraction function 53 obtains the maintenance results of the simulation target customer from the database system 10, and, together with the inputted judgment criteria, the probability of the received evaluation index (here, the number of failures) The data is superimposed on the distribution (step S506) and output to, for example, the display device of the GUI unit 16-2 (step S507). Here, an example of the output is shown in FIG.
[0087]
The user of the sales support unit 5 can easily extract the customer requiring investigation by looking at the output result shown in FIG. For example, in the case of an elevator with many parts replacements and breakdowns, the situation is considered that there are many mischiefs, the usage status has changed and the operation frequency has increased rapidly, and the user instructs the sales staff and security staff to investigate. Then, the situation can be grasped, and the proposal of a new maintenance service such as a revision of a security service or a fee or a proposal of a security camera can be considered.
[0088]
The processing operation of the sales support unit 5 has been described above.
[0089]
Next, the processing operation of the maintenance business simulation unit 2 will be described.
[0090]
FIG. 7 is a conceptual diagram of a maintenance work for replacing specific components assumed in the present embodiment. In the present embodiment, two types of maintenance work are considered: preventive replacement and post-replacement (corresponding to a name).
[0091]
Here, the preventive replacement is to replace a part after a predetermined period (hereinafter, referred to as a preventive replacement interval) for each part from the mounting of the part, and is performed when the life of the part is longer than the preventive replacement interval. It is. On the other hand, the maintenance work (replacement of parts) by post-replacement is to replace parts due to a failure, and is performed when the life of the parts is shorter than the preventive replacement interval.
[0092]
The reason why both maintenance works are treated separately is that they are different when calculating work man-hours and costs. For example, costs and man-hours (emergency delivery, faulty component identification man-hours, etc.) are generated during an emergency response, so that "post-replacement" generally requires more cost and man-hours than "preventive replacement".
[0093]
In the present embodiment, an example will be described in which the Monte Carlo method, which can be implemented relatively easily, is used to calculate the number of maintenance operations. In the Monte Carlo method, the life of each component of an elevator is generated using a pseudo-random number according to a predetermined random variable, and each life is compared with a predetermined preventive replacement scheduled interval for each component. Thus, the number of preventive replacements and subsequent replacements are counted. Of course, the simulation method is not limited to the Monte Carlo method. For example, a method of modeling a maintenance work process as a regeneration process may be used.
[0094]
In addition, in actual maintenance work, there are other inspection / maintenance work, legal inspection work, etc. By comparing the interval with the component life generated by the pseudorandom number, it is possible to count the number of times such as inspection work or post-replacement.
[0095]
Next, the flow of processing in the maintenance business simulation unit 2 will be described.
[0096]
FIG. 8 shows a processing flow of the maintenance business simulation unit.
[0097]
As shown in the figure, when the maintenance business simulation unit 2 receives data such as simulation conditions and targets from the business support unit 5 or the business plan planning support unit 3, the maintenance work simulator 24 is activated, and the received data is processed. It is stored in the memory (step S801).
[0098]
Here, the input data such as the simulation condition and the target are the number of generated sample functions, the simulation period, the simulation time interval, the centralization period, the total time interval, the simulation target customer, and the calculation evaluation index.
[0099]
The maintenance work simulator 24 first accesses the database system 10, acquires data necessary for executing the simulation, and stores the data in a memory (step S802).
[0100]
Here, the data acquired from the database system 10 is necessary for calculating the components constituting the elevator held by the simulation target customer who has received the input, the preventive replacement interval of each component, and the evaluation index for which the input has been received. The data includes, for example, the unit price of each part, the number of man-hours required for replacing each part, the cost of a worker, and the unit of part replacement.
[0101]
Next, using the input data read from the memory, the maintenance work simulator 24 calculates the probability of the combination of the number of preventive and post-replacement replacements of each component of the lift that the customer has by the Monte Carlo method, and (Step S803). Details of the calculation by the Monte Carlo method will be described later.
[0102]
Then, the maintenance work simulator 24 reads from the memory the probability distribution of the calculated combination of the number of component replacements and the data obtained from the database system 10, calculates the evaluation index for which the input has been received, and stores it in the memory (step S804). .
[0103]
The maintenance business simulation unit 2 stores the result calculated by the maintenance work simulator 24 in step S804 in the simulation result file 4 and transfers it to the instructed sales support unit 5 or the business plan planning support unit 3 (step S805).
[0104]
Next, the details of the processing in which the maintenance work simulator 24 counts the number of maintenance works (preventive replacement and post-replacement of parts) by the Monte Carlo method and obtains the probability distribution of the combination will be described below.
[0105]
First, a simulation target of the maintenance work simulator 24 will be described.
[0106]
FIG. 9 is a conceptual diagram of a simulation target of the maintenance work simulator 24.
[0107]
As shown in FIG._i(I = 1,..., NC) are NM, respectively._iEquipment M to be maintainedⁱ _j(J = 1,... NM_i).
[0108]
In addition, the maintenance target device Mⁱ _jIs NPⁱ _jParts to be maintained_iP^j _k(K = 1,... NPⁱ _j).
[0109]
Here, the parts to be maintained_IP^j _kParts life_iτ^j _kAre the probability density functions_If^j _k(_iτ^j _k). The part life probability density function may be a multivariable function.
[0110]
Next, a processing procedure of the maintenance work simulator 24 will be described.
[0111]
First, symbols used in the following description are defined.
[0112]
(Equation 1)

FIG. 10 shows a processing flow of the maintenance work simulator 24.
[0113]
As described above, in the present embodiment, the Monte Carlo method is used. The basic process flow of the simulation according to the present embodiment is to generate a component life using a pseudo-random number sequence and count the number of maintenance operations.
[0114]
The following processing is performed by the maintenance work simulator 24 in accordance with a predetermined program stored in the memory.
[0115]
When the maintenance work simulator 24 is started, in step S2, in step S2, the model and the number of elevators owned by the designated customer, the components of each elevator, the probability distribution function of the life of each component, the preventive replacement interval of each component, Are acquired from the customer database 12, the product specification database 11, the maintenance result database 15, etc. of the database system 10, respectively, and stored in the memory.
[0116]
In the following step S3, the number of created sample functions stored in the memory is compared with the number of generated sample functions received in the memory upon receiving an input, and it is determined whether a desired number of sample functions have been generated. Judge and store the judgment result in the memory. Then, the determination result stored in the memory is read, and if [Yes], the process proceeds to step S4, and if [No], the process proceeds to step S14.
[0117]
In step S4, the simulation time t, the work number counter
_iNWP^j _k(T_a),_iNWB^j _k(T_a) Is initialized to 0 and stored in memory. Here, it is assumed that the simulation time t is discretized to take an integer value by using a certain time interval.
[0118]
In the following step S5, each component_iP^j _kLifetime_iτ^j _kIs the probability density function_if^j _k(_iτ^j _k) Is used to determine the life of the component, and the determined life is digitized using the time interval and stored in a memory.
[0119]
Next, using the component life stored in the memory,_iP^j _kAt which the end of life is reached_itl^j _k= T +_iτ^j _kIs calculated and stored in the memory.
[0120]
In the following step S6, an exchange operation occurs at the current simulation time t (t =_itl^j _kOr, the time elapses by the preventive replacement interval from the component mounting). It is determined whether a component exists by comparing the component life stored in the memory with the current simulation time t, and the determination result and the replacement work are determined. Information for identifying the generated component is stored in the memory.
[0121]
Then, the determination result stored in the memory is read out. If [Yes], the process proceeds to step S7, and if [No], the process proceeds to step S11.
[0122]
In step S7, it is determined whether the current simulation time t stored in the memory is within the evaluation index totaling period by comparing it with the end time of the evaluation index totaling period stored in the memory, and the determination result is stored in the memory. I do. The determination result stored in the memory is read out. If [Yes], the process proceeds to step 8, and if [No], the process proceeds to step S10.
[0123]
In step S8, the component determined to require replacement work in step S6 is read from the memory, and the read component is read._iP^j _kThe time when the end of life is reached_itl^j _kCalculate time interval dt_aAnd store it in a memory, and then go to step S9. Where [x]⁻Is a Gaussian symbol and represents the largest integer value not exceeding x.
[0124]
(Equation 2)

In the following step S9, the parts determined to require replacement work in step S6_iP^j _kAnd the current simulation time t are read from the memory, and by comparing the two, the type of the replacement work (preventive replacement or post-replacement) performed on the part is determined, and the counter of the number of the corresponding work stored in the memory is determined. (_iNWP^j _k(T_a) Or_iNWB^j _k(T_a)) Is read out, incremented by 1 and stored again in the memory, and the process proceeds to step S10.
[0125]
In step S10, the part determined to require replacement in step S6 is read from the memory, and_iP^j _kNew component life_iτ^j _kSet the life time_itl^j _kIs calculated and stored in the memory, and the process proceeds to step S6.
[0126]
In step S11, the simulation time t read from the memory is increased by 1 and stored in the memory again, and the process proceeds to step S12.
[0127]
In the following step S12, the simulation time t and the simulation period set by the user are read from the memory, and the two are compared to determine whether the simulation time t has exceeded the simulation period set by the user. The result is stored in the memory. The determination result is read from the memory. If [Yes], the process proceeds to step S13, and if [No], the process proceeds to step S6.
[0128]
In a succeeding step S13, the number of work for each part_iNWP^j _k(T_a),_iNWB^j _k(T_a), The following formula 3 is read from the memory, incremented by 1 and stored in the memory again, and the process proceeds to step S3.
[0129]
(Equation 3)

In step S14, the maintenance work counter is read from the memory, divided by the value of the number of created sample functions also read from the memory, and the number of maintenance work is calculated._iNWP^j _k,_iNWB^j _kIs calculated and stored in the memory.
[0130]
(Equation 4)

In the subsequent step S15, the number of maintenance operations stored in the memory is read out, if necessary, and using the following data read out from the database system 10 and stored in the memory, the probability distribution of maintenance costs and maintenance man-hours, An evaluation index such as an expected value is calculated and stored in a memory.
[0131]
Here, an example of the evaluation index calculation formula is shown below.

In the subsequent step S16, the GUI unit 16-reads out the result calculated in step S15 from the memory, outputs it to the simulation result file 4, and transfers it to the sales support unit 5 and the business planning support unit 3 as necessary. Then, the process is terminated, for example, by outputting to the display screen of No. 2.
[0132]
As described above, the maintenance work simulation unit 2 of the present embodiment can output various evaluation indices by the processing of the maintenance work simulator 24. These outputs can be used to support sales activities and business plan planning.
[0133]
The above-described maintenance business simulation unit 2 uses a relatively simple model including a maintenance work simulator 24 that calculates evaluation indices (number of parts replacement, cost, work amount, etc.) related to the maintenance work.
[0134]
The maintenance business simulation unit 2 is not limited to this. Other examples are shown in FIGS. 2 (b) and 2 (c).
[0135]
The maintenance business simulation unit 2 shown in FIG. 2B is a development of the one shown in FIG. 2A, and has a logistics simulator 25 and an indirect work simulator 26 in addition to the maintenance work simulator 24.
[0136]
The logistics simulator 25 and the indirect work simulator 26 each receive the result of the maintenance work simulator 24 in addition to the input data 27, and evaluate an index for parts distribution (inventory, delivery amount, etc.) and an index for indirect operations (report). This is a model in which only one-way coupling from the maintenance work simulator 24 is considered.
[0137]
Here, “coupling” in the maintenance business simulation unit 2 of the present embodiment means that the simulation results by the respective simulators affect each other. For example, if parts are not delivered due to lack of parts, etc., or if there are insufficient personnel to perform indirect tasks such as parts ordering work, work will not be performed as scheduled, resulting in breakdowns, emergency parts arrangements, For example, transportation is required.
[0138]
The maintenance business simulation unit 2 shown in FIG. 2 (c) is a further development of the one shown in FIG. 2 (b), and the maintenance work simulator 24, the logistics simulator 25, and the indirect work simulator 26 are mutually coupled. Is a model that takes account of the above.
[0139]
The logistics simulator 25 and the indirect work simulator 26 used in FIGS. 2B and 2C can be realized by a Monte Carlo method or the like using as input the probability distribution of the number of maintenance work occurrences calculated by the maintenance work simulator 24. it can.
[0140]
2B and 2C, the case where the maintenance work simulator 24, the logistics simulator 25, the indirect work simulator 26, and the like are used separately can be similarly considered.
[0141]
Next, details of the component life analysis unit 9 will be described.
[0142]
FIG. 11 is a configuration diagram of the component life analysis unit 9. As shown in the figure, the component life analysis unit 9 includes a component analysis engine 32 and a component life analysis knowledge database 33.
[0143]
The component life analysis engine 32 is an analysis engine that calculates a component life distribution based on given data, and the component life analysis knowledge DB 33 stores knowledge related to component life analysis used at the time of analysis.
[0144]
Next, the processing operation of the component life analysis engine will be described.
[0145]
FIG. 12 is a processing flowchart of the component life analysis engine 32.
[0146]
Note that the following processing is performed by the CPU using data stored in the storage unit according to a predetermined program.
[0147]
Upon receiving the execution instruction, the component life analysis engine 32 acquires knowledge related to component life analysis from the component life analysis knowledge DB 33 in step S21, and stores it in the storage unit.
[0148]
Here, the knowledge regarding the component life analysis is “parts used near a hot spring and parts used in a normal environment are separately analyzed”, and “part X is not the cumulative running time, but the life based on the number of years of use. Is determined, "and" the life of the component Y follows the Weibull distribution. "
[0149]
In the next step S22, component life test data (test data at the time of product development, remaining life test data of collected components, etc.) and analysis source data such as actual failure data are acquired from the database system 10 and stored in the storage unit. .
[0150]
In the next step S23, the analysis source data acquired in step S22 and the knowledge relating to the component life analysis acquired in step S21 are read, the analysis source data is selected according to the knowledge, and the result is stored in the storage unit. At this time, failure data that has failed due to factors (such as mischief) caused by the customer instead of the component life is also removed.
[0151]
In step S24, the original data selected in step S23 is read from the storage unit, a component life analysis is performed using the read data, and the result is stored in the storage unit.
[0152]
There are many known methods for component life analysis, such as Weibull hazard analysis, and these methods are used.
[0153]
In the last step S25, the component life analysis result obtained in step S24 is read from the storage unit, stored in the product use DB, and the process ends.
[0154]
The processing of the component life analysis engine 32 has been described above.
[0155]
As described above, according to the maintenance business support system of the present embodiment, by using the probability distribution data of the evaluation index calculated by the maintenance business simulator 2, the operation characteristics of the elevator of each customer are extracted. It becomes easier. Then, based on the extracted characteristics, it is possible to output information for supporting planning of a maintenance service according to each customer.
[0156]
Further, by visualizing and outputting the simulation result, it assists in quantitatively explaining the features of the planned maintenance service to the customer. These make it possible to clarify the fee system and set a simple SLA (Service Level Agreement).
[0157]
Further, according to the present embodiment, it is possible to assist in efficiently extracting customers who should perform sales activities, and to assist in determining an effective approach time to customers. These lead to more efficient sales activities.
[0158]
Furthermore, according to the maintenance business support system of the present embodiment, the business plan can be drafted by the business plan drafting support unit 3 while the efficiency of the sales activities is improved as described above. As a result, it is possible to eliminate excessive / wasteful maintenance business plans.
[0159]
In addition, by sharing data with the supplier 30 and the own factory 40, it becomes possible to support product development and support the planning of parts production / delivery plan in cooperation with the supply and the production / delivery planning system of the own factory 40. .
[0160]
According to the present embodiment, the maintenance business support system using a simulator for maintenance business as a nucleus provides the accuracy of the accuracy calculated by a logical means using a simulator to a business that conventionally relies on information based on experience. High support information can be provided, which allows efficient operation of the entire maintenance business.
[0161]
In the present embodiment, the maintenance business of the elevator is described as an example, but the maintenance object to which the present invention can be applied is not limited to the elevator. After the sale, it can be applied to the general maintenance business of products requiring maintenance for a predetermined period.
[0162]
【The invention's effect】
By supporting the planning of a maintenance service according to the requirements of each user, the planning of the service is facilitated, the satisfaction of the customer is increased, and the maintenance business that is efficient for the business operator is realized.
[Brief description of the drawings]
FIG. 1 is a diagram for explaining an overall configuration of a maintenance business support system to which the present embodiment is applied.
FIG. 2 is a functional configuration diagram of a maintenance business simulation unit of the present embodiment. (A) is a functional configuration diagram of a basic maintenance business simulator, (b) is a first development example, and (c) is a second development example.
FIGS. 3A to 3F are output image diagrams of a maintenance business simulation unit of the present embodiment, respectively.
FIG. 4 is a diagram in which functions related to a present embodiment of a supplier and a company factory are extracted.
FIG. 5 is a processing flow of a sales support unit according to the embodiment;
6 (a) to 6 (c) are output image diagrams of the sales support unit of the present embodiment, respectively.
FIG. 7 is a diagram for explaining the concept of the maintenance work according to the embodiment;
FIG. 8 is a processing flow of a maintenance business simulation unit of the present embodiment.
FIG. 9 is a diagram for explaining the concept of a simulation target of the maintenance work simulator according to the embodiment;
FIG. 10 is a processing flow of the maintenance work simulator of the present embodiment.
FIG. 11 is a functional configuration diagram of a component life analysis unit of the embodiment.
FIG. 12 is a processing flow of a component life analysis unit of the embodiment.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Decision support system 2 ... Maintenance business simulation part 3 ... Business plan drafting support part 4 ... Simulation result file 5 ... Sales support part 6 ... Actual data Collection / analysis system, 7: operation results collection unit, 8: maintenance results collection unit, 9: parts life analysis unit, 10: database system, 11: product specification DB, 12,.・ Customer DB, 13 ・ ・ ・ Maintenance system DB, 14 ・ ・ ・ Operation results DB, 15 ・ ・ ・ Maintenance results DB, 16-1, 16-2 ・ ・ ・ GUI parts, 16-3, 16-4 ... · Usage terminal, 17-1 ··· Work result collection terminal, 17-2 ··· Customer data collection terminal, 19-1, 19-2, 19-3, 19-4, 19-5, 19-6 ··· -Communication processing unit, 20-1, 20-2, 20-3, communication line, 21-1 20-2: Customer building, 30: Supplier, 40: Own factory, 24: Maintenance work simulator, 25: Logistics simulator, 26: Indirect work simulator, 27: Input data, 28 ... output data, 32 ... parts life analysis engine, 33 ... parts life analysis knowledge DB, 34 ... production / delivery planning system

Claims (9)

A maintenance method, comprising: a probability distribution calculation unit configured to calculate a probability distribution of a combination of the number of maintenance operations of each component by using a component life of each component constituting the maintenance target device and a predetermined maintenance method. Business simulator. A maintenance business simulator for calculating a probability distribution of a combination of the number of maintenance operations of each part constituting a maintenance target device,
Component life generating means for generating a component life of each of the components by pseudo random numbers according to a predetermined probability distribution function,
Based on the component life generated by the component life generating means and a predetermined replacement time of each of the components, the number of times the component is replaced due to the failure (post-replacement), And a probability distribution calculating unit that counts the number of times of replacement of parts (preventive replacement) due to the occurrence of the component and calculates a probability distribution of a combination of the number of times of the parts of the post-replacement and the preventive replacement. Features a maintenance business simulator. The maintenance business simulator according to claim 2,
From the probability distribution calculated by the probability distribution calculating means, an evaluation for calculating an evaluation index including the number of parts required for the maintenance work, the cost, and the number of man-hours of the work, using information about the characteristics of the parts held in advance. A maintenance business simulator further comprising an index calculation means. A maintenance service planning support system that supports planning of a maintenance service to be provided to a customer who owns the maintenance target device from data on operation results of the maintenance target device,
A maintenance business simulator according to claim 3,
A maintenance support system that calculates useful information when planning the maintenance service using the evaluation index calculated by the maintenance business simulator. The maintenance service planning support system according to claim 4,
The maintenance service planning support system, wherein the useful information is information indicating a predetermined compliance rate of a maintenance target device held by the customer with a service level agreement contract. The maintenance service planning support system according to claim 4 or 5,
The maintenance service planning support system, wherein the useful information is information serving as a judgment material for determining a time at which a new maintenance service is presented to the customer. The maintenance service planning support system according to claim 4, 5, or 6,
The maintenance service planning support system, wherein the useful information is information obtained by superimposing operation results of the maintenance target device held by the customer on the evaluation finger. A maintenance business support system that supports maintenance service planning based on data on the operation results of equipment to be maintained and also supports maintenance business planning.
A maintenance service planning support system according to any one of claims 4 to 7,
A maintenance project planning support system that supports the maintenance project planning by causing the maintenance project simulator to calculate the evaluation index a plurality of times. A maintenance work number probability distribution calculation method for calculating a probability distribution of a combination of the number of maintenance work times of each part constituting a maintenance target device,
A component life generating step of generating the component life of each of the components by a pseudorandom number according to a predetermined probability distribution function,
Based on the component life generated in the component life generation step and a predetermined replacement time of each of the components, the number of times of replacement (post-replacement) of the component due to the occurrence of the failure, and the replacement time, respectively. And a probability distribution calculating step of calculating the probability distribution of the combination of the number of times of the parts of the post-replacement and the preventive replacement. A characteristic method of calculating the probability distribution of the number of maintenance operations.

Images