JP2015114849A - Demand prediction device, demand prediction method, and computer program for demand prediction - Google Patents

Abstract

PROBLEM TO BE SOLVED: To precisely predict the number of components necessary for inspection of a machine.SOLUTION: A demand prediction device 1 includes: a prediction unit 2A for obtaining estimated use time during which a machine to be inspected is estimated to be used after the last inspection until the next inspection on the basis of usage record of the machine; a required number calculation unit 2B for obtaining a required number that was requested per one machine in the inspection in the past regarding at least one component used for the machine according to time during which the machine was used; and a required component number calculation unit 2C for obtaining the number of component necessary for the next inspection from the estimated use time and the required number corresponding to the estimated use time.

Description

  The present invention relates to a demand prediction apparatus, a demand prediction method, and a computer program for demand prediction that predict demand for replacement parts.

  When a machine such as an aircraft, a ship, a vehicle, or a gas turbine is inspected, it is necessary to replace a worn part or a new part to improve performance. Patent Document 1 describes a plant demand prediction method and apparatus that can easily cope with changes in demand fluctuation patterns and can reduce the number of operation steps.

JP 2001-188772 A

  By the way, in the inspection of the machine, the parts included in the machine are repaired or replaced according to the degree of deterioration. In the inspection, there is a trade-off between the inventory of parts owned by the entity that undertakes the inspection and the service rate that responds to the request from the party requesting the inspection, and the target service rate can be achieved with the minimum inventory. It is expected to be achieved. For that purpose, it is effective to accurately predict the quantity of necessary parts.

  The required amount of parts used for the inspection is adjusted by the operator based on the average value of past results for each part, and the future required quantity is predicted. As a result, there was a possibility of overstocking or shortage of parts for parts with low demand occurrence and parts for which a sudden decrease or increase in demand occurred in the previous year.

  It is an object of the present invention to accurately predict the number of parts required for machine inspection.

  The present invention relates to a prediction unit that obtains a predicted usage time that is predicted to be used from the time when the machine to be inspected to the next inspection after receiving the last inspection, to the machine. For at least one part to be used, a required number calculation unit for obtaining the required number per machine in the previous inspection according to the time when the machine was used, the predicted usage time, A demand forecasting device comprising: a required number-of-parts calculation unit that obtains the number of parts required in the next inspection from the required number corresponding to the predicted usage time.

  The present invention obtains an estimated usage time that a machine to be inspected is expected to be used after the last inspection until the next inspection, and at least for at least one part used in the machine so far The number of requests required for each machine in the inspection is determined according to the time when the machine is used, and the next time is calculated from the predicted usage time and the required number corresponding to the predicted usage time. This is a demand prediction method for obtaining the number of parts required for inspection.

  The present invention is a computer program for demand prediction that causes a computer to execute the above-described demand prediction method.

  The present invention uses the time obtained from the average value regarding the usage time of the machine for the quantity of each part required in the machine inspection. Therefore, the demand prediction apparatus, the demand prediction method, and the demand prediction computer program according to the present invention can accurately predict the quantity of parts necessary for machine inspection.

  The requested number is preferably classified for each range of time the machine is actually used. By doing so, the prediction accuracy of the number of parts required in the next inspection is improved.

  It is preferable that the number of parts required in the next inspection is obtained from the predicted usage time and the required number corresponding to the predicted usage time and the use area of the machine. Depending on the area where the machine is used, the impact on the machine and the parts of the machine may differ, but the present invention predicts the number of parts required for the next inspection by considering the influence depending on the area of use. Accuracy can be improved.

  It is preferable that the number of parts required in the next inspection is obtained from the predicted usage time and the required number corresponding to the predicted usage time and the usage pace of the machine. Depending on the pace of use of the machine, the influence of the machine and the parts of the machine may be different, but the present invention predicts the number of parts required in the next inspection by considering the influence according to the use pace. Accuracy can be improved.

  The number of parts required in the next inspection considering the use pace or use area of the machine and the number of parts required in the next inspection not considering the use pace or use area of the machine Of these, it is preferable to select the one closer to the past results. By doing in this way, what is closer to the past actual value can be used as the total number of requests in the next inspection, so that the accuracy of demand prediction can be further improved.

  The number of parts required in the next inspection considering the use pace and use area of the machine and the number of parts required in the next inspection not considering the use pace and use area of the machine Of these, it is preferable to select the one closer to the past results. By doing in this way, what is closer to the past actual value can be used as the total number of requests in the next inspection, so that the accuracy of demand prediction can be further improved.

  It is preferable that the number of parts necessary for the next inspection is obtained from the predicted usage time, the predicted usage time, the usage area of the machine, and the required number corresponding to the usage pace of the machine. The present invention can improve the prediction accuracy of the number of parts required in the next inspection by taking into consideration the influence according to the machine use area and the machine use pace.

  The present invention can accurately predict the quantity of parts required for machine inspection.

FIG. 1 is a schematic diagram illustrating a facility demand prediction apparatus according to the present embodiment. FIG. 2 is a flowchart illustrating a procedure of the demand prediction method according to the first embodiment. FIG. 3 is a diagram for explaining processing in the demand prediction method according to the first embodiment. FIG. 4 is a diagram for explaining processing in the demand prediction method according to the first embodiment. FIG. 5 is a diagram for explaining processing in the demand prediction method according to the first embodiment. FIG. 6 is a diagram for explaining processing in the demand prediction method according to the first embodiment. FIG. 7 is a diagram for explaining processing in the demand prediction method according to the first embodiment. FIG. 8 is a flowchart illustrating a procedure of the demand prediction method according to the second embodiment. FIG. 9 is a diagram for explaining processing in the demand prediction method according to the second embodiment. FIG. 10 is a diagram for explaining processing in the demand prediction method according to the second embodiment. FIG. 11 is a diagram for explaining processing in the demand prediction method according to the second embodiment. FIG. 12 is a diagram for explaining processing in the demand prediction method according to the third embodiment. FIG. 13 is a diagram for explaining processing in the demand prediction method according to the third embodiment. FIG. 14 is a diagram for explaining processing in the demand prediction method according to the third embodiment.

  DESCRIPTION OF EMBODIMENTS Embodiments (embodiments) for carrying out the present invention will be described in detail with reference to the drawings. In the following, an aircraft is taken as an example of a machine, but a machine to which the demand prediction apparatus and the demand prediction method according to the present embodiment can be applied is not limited to an aircraft. The demand prediction apparatus and the demand prediction method according to the present embodiment can also be applied to predicting demand for parts in inspections of ships, vehicles, and the like other than aircraft.

(Embodiment 1)
FIG. 1 is a schematic diagram illustrating a facility demand prediction apparatus according to the first embodiment. The demand prediction apparatus 1 is an apparatus that predicts the number of parts required at the time of inspection for at least one part provided in a machine having a very large number of parts and various kinds of parts, such as an aircraft.

  The demand prediction device 1 includes a processing unit 2, a storage unit 3, and an input / output unit 4. The demand prediction device 1 is, for example, a computer. The processing unit 2 is a processor such as a CPU (Central Processing Unit). The processing unit 2 realizes the demand prediction method according to the present embodiment. Therefore, the processing unit 2 includes a prediction unit 2A, a required number calculation unit 2B, and a necessary component number calculation unit 2C.

  The storage unit 3 is, for example, a semiconductor memory such as a random access memory (RAM) and a read only memory (ROM) or a hard disk device. In the present embodiment, the storage unit 3 includes both a main storage device and an auxiliary storage device. The storage unit 3 stores various computer programs such as a demand prediction computer program (PGM) 11 for causing a computer (processing unit 2) to execute the demand prediction method according to the present embodiment, or the present embodiment. Data necessary for realizing the demand prediction method according to the above is stored. Such data is stored in, for example, the first database 12a, the second database 12b, the third database 12c, and the fourth database 12d.

  In the first database 12a, for example, information on parts necessary for the inspection is described. The information in the first database 12a is based on the actual values accumulated in the previous examinations. In the second database 12b, for example, information such as the inspection time is described. In the third database 12c, for example, at least one of information on the area where the machine is used and information on the pace of use of the machine is described. The fourth database 12d is, for example, a database in which information related to past examination results is accumulated.

  The processing unit 2 sequentially reads and interprets the instruction sequence of the computer program stored in the storage unit 3, and moves or processes data according to the result. The processing unit 2 reads the demand prediction computer program 11 from the storage unit 3 and executes it, thereby realizing the functions of the prediction unit 2A, the required number calculation unit 2B, and the required component number calculation unit 2C. In this way, the demand prediction device 1 realizes the demand prediction method according to the present embodiment. The processing unit 2 may be realized by dedicated hardware.

  The demand forecasting method according to this embodiment is realized by recording the computer program 11 for demand forecasting on a computer-readable recording medium, reading the computer program recorded on the recording medium into a computer system, and executing it. Also good. The “computer system” here includes an OS (Operating System) and hardware such as peripheral devices.

  In the present embodiment, the “computer-readable recording medium” refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, or a recording device such as a hard disk built in a computer system. Say. The computer program described above may be capable of realizing the function of the processing unit 2 in combination with a computer program already recorded in the computer system.

  The demand prediction method according to the present embodiment can be realized by executing a demand prediction computer program 11 prepared in advance by a computer such as a personal computer or a workstation. The demand prediction computer program 11 can be distributed through a communication line 7 such as the Internet.

  The input / output unit 4 connects the demand prediction device 1 with a display device 5 or an input device 6 as an external device to exchange information, or connects the demand prediction device 1 and the communication line 7. The display device 5 is, for example, a liquid crystal display. The input device 6 includes, for example, a keyboard 6K and a mouse 6M. The input device 6 may be a touch panel, for example.

  In the present embodiment, the demand prediction device 1 may be connected to the communication line 7. In this case, the demand prediction apparatus 1 acquires information from computers 8A and 8B such as terminal devices or servers connected to the communication line 7, and receives commands from the computers 8A and 8B as terminal devices. The computers 8A and 8B are devices used by workers who maintain and inspect aircraft, for example. In this case, for example, the operator can order parts necessary for inspection of the machine, in this embodiment, the aircraft from the computers 8A and 8B.

  FIG. 2 is a flowchart illustrating a procedure of the demand prediction method according to the first embodiment. 3 to 7 are diagrams for explaining processing in the demand prediction method according to the first embodiment. The demand prediction method according to the present embodiment predicts the amount of parts required for the next inspection of the aircraft from the past use results of the aircraft that is the inspection target. This prediction is also referred to as demand forecast as appropriate.

  When the demand prediction apparatus 1 shown in FIG. 1 executes the demand prediction method according to the present embodiment, the processing unit 2 reads the demand prediction computer program 11 from the storage unit and executes instructions described therein. In step S <b> 101, the processing unit 2 acquires information on an aircraft body scheduled to be inspected in the next inspection in the present embodiment in the future (hereinafter, referred to as airframe information as appropriate). In the present embodiment, the inspection is performed, for example, every four years. Aircraft information includes at least the airframe number, the total number of aircraft to be inspected NAL, the flight time (use time) when it is brought into the inspection facility for inspection, and the flight time (use time) when it is carried out from the inspection facility May be included. The machine body information may be stored in advance in the storage unit 3, or may be input to the processing unit 2 from the input device 6 illustrated in FIG. 1, or the processing unit 2 is connected to this via the communication line 7. May be obtained from the computers 8A, 8B, etc.

  In step S102, the prediction unit 2A of the processing unit 2 sets the predicted usage time as the time that the aircraft as the machine to be inspected is predicted to be used after the last inspection until the next inspection, Obtained based on the actual use of the aircraft to be inspected. FIG. 3 shows the actual use of the aircraft to be inspected. This usage record may be input to the processing unit 2 from the input device 6 or the like as machine information, for example. Moreover, the use results regarding the airframe to be inspected may be databased and stored in the storage unit 3 of the demand prediction apparatus 1.

  The aircraft number MT is A in the aircraft that is the target of the usage record shown in FIG. Hereinafter, this aircraft is referred to as an aircraft having an aircraft number A as appropriate. FIG. 3 describes the actual use of the aircraft with the machine number A from 1996 to 2012. The information in FIG. 3 is acquired from, for example, the second database 12b shown in FIG. In FIG. 3, tin represents the carry-in time, and tout represents the carry-out time. t1, t3, t5, t7, and t9 are loading times tin in the inspections of 1996, 2000, 2004, 2008, and 2012, respectively, and t2, t4, t6, t8, and t10 are respectively 1996, It is the carry-out time tout in the inspections of 2000, 2004, 2008, and 2012. The last inspection is 2012, and in this embodiment, the inspection of the aircraft is performed every four years, so the next inspection is 2016. For example, at the time of 2013, the processing unit 2 predicts the number of parts required for the 2016 inspection.

  In the present embodiment, it is assumed that there is an actual use record from 1996 to 2012 for the aircraft with the machine number A. From this usage record, the prediction unit 2A obtains the predicted time that the aircraft with the machine number A is expected to fly in the four years from 2016 when the final inspection is completed to the next inspection in 2016, that is, the predicted usage time. . In the present embodiment, the prediction unit 2A obtains a value Δt (= t9−t2) obtained by subtracting the carry-out time tout = t2 of the oldest inspection included in the use record from the carry-in time tin = t9 in the last inspection. By dividing by a period ΔY from 1996 to 2012 (16 years in this embodiment), an average flight time (average use time) tav per year of the aircraft of the aircraft number A is obtained. tav = Δt / ΔY. The prediction unit 2A can obtain the predicted usage time tg predicted by the aircraft of the aircraft number A flying in the four years from 2012 to 2016 by 4 × tav. That is, the predicted usage time tg = 4 × tav.

  Next, in step S103, the required number calculation unit 2B of the processing unit 2 shown in FIG. 1 requests at least one part used for the aircraft as the machine to be inspected per aircraft in the previous inspections. The requested number is determined according to the time the aircraft has been used. Information about the components can be acquired from, for example, the first database 12a shown in FIG. FIG. 4 shows the number of parts Pa used for the aircraft in the range of the flight time, that is, the usage time of 1200 hours (h) or more and less than 1400 hours (h) among the aircrafts inspected at the time of the 2004 inspection. It is. FIG. 4 shows the actual usage of the aircraft.

  In the example shown in FIG. 4, at the time of inspection in 2004, the number of parts Pa (number of uses) AD used for aircrafts with aircraft numbers MT of B, C, D, E, and F is different for each aircraft. It has been described. In the inspection in 2004, three parts Pa are used for five aircraft. The required number calculation unit 2B also calculates the number of aircraft (number of aircraft) MN that used parts Pa and the number of used parts (number of uses) AD in past inspections such as 2000 and 2008. Seeking a relationship.

  Next, the requested number calculation unit 2B obtains the requested number ADSav requested (used) per aircraft in the previous inspections. In the present embodiment, the required number ADSav is obtained from the results of the past three inspections in 2000, 2004, and 2008. In the past three inspections in 2000, 2004, and 2008, the required number ADSav in the range where the usage time is 1200 hours (h) or more and less than 1400 hours (h) is divided by the total number of aircraft MNA. Is required. That is, the required number ADSav = total number of use ADS / total number of aircraft MNA. In the example shown in FIG. 4, since the total number of aircraft MNA is 7 and the total number of use ADS is 9, the required number ADSav is 9/7 = 1.3 (pieces).

  The total use number ADS is the total number of parts Pa used in the past inspection (three times in the present embodiment), and is obtained by adding the use number AD used in each inspection for all inspections. The total number of aircraft MNA is the total number of aircraft whose parts Pa have been replaced in the past inspection (three times in this embodiment), and is obtained by adding the number of aircraft MN used in each inspection for all inspections. It is done.

  In the example shown in FIG. 4, the required number ADSav is obtained using information (the actual use of the aircraft) in the past three inspections, but the information used for obtaining the requested number ADSav is not limited to this. For example, the information used when obtaining the required number ADSav may be all the information existing at the time of predicting the demand for parts required for the next inspection, or may be a part of the information. After all the existing information is statistically processed, it may be information excluding specific values. The same applies to the following embodiments.

  The requested number calculation unit 2B similarly obtains the requested number ADSav for the parts Pa even in different usage time ranges. FIG. 5 shows the range 1 in which the usage time RT is 400 hours (h) or more and less than 600 hours (h), the range 2 in which the usage time RT is 600 hours (h) or more and less than 800 hours (h), and the usage time RT is 800 hours. (H) Range 3 for 1000 hours (h) or more, Range 4 for use time RT of 1000 hours (h) or more and less than 1200 hours (h), Use time RT for 1200 hours (h) or more and less than 1400 hours (h) The requested number ADSav corresponding to the range 5 is shown.

  When the number of parts subject to demand prediction is 2 or more, the required number calculation unit 2B calculates the required number ADSav for each range of each usage time RT for each part. FIG. 6 is an example of a result obtained by obtaining the required number ADSav for each range of usage time RT for a part Pb different from the part Pa described above. Thus, in the present embodiment, the requested number calculation unit 2B classifies the requested number ADSav for each range of time when the machine was actually used, that is, when the aircraft actually flew.

  When the required number ADSav is obtained for every range of the usage time RT for all the parts to be demanded in the next inspection, the process proceeds to step S104. In step S104, the required number-of-parts calculation unit 2C of the processing unit 2 included in the demand prediction apparatus 1 shown in FIG. 2 has the number of parts required for the next inspection (total number of requests) for all aircraft to be inspected. ) Find PNAL. In the present embodiment, the required number-of-components calculation unit 2C calculates the total number of requests PNAL from the predicted usage time tg and the required number ADSav corresponding to the predicted usage time tg.

  FIG. 7 shows the relationship between the predicted usage time tg and the requested number ADSav corresponding to the predicted usage time tg for the eight aircrafts whose aircraft numbers MT are A to H. In the present embodiment, it is assumed that nine aircrafts whose aircraft numbers MT are A to H are to be subjected to the next inspection. As described above, the predicted usage time tg is obtained in step S101, and the required number ADSav is obtained in step S103. FIG. 7 shows information about the part Pa. Before the necessary part number calculation unit 2C obtains the total request number PNAL, information about the part that is the target of demand prediction in the next inspection is obtained.

  For example, the required number-of-parts calculation unit 2C determines which usage time range the predicted usage time tg of each aircraft belongs in the relationship between the usage time range obtained in step S103 and the required number ADSav. The required number ADSav corresponding to the usage time range to which the predicted usage time tg belongs is the required number ADSav required for the aircraft of the predicted usage time tg.

  For example, for the part Pa, the required number ADSav of each aircraft can be obtained by giving the predicted usage time tg of each aircraft to the relationship shown in FIG. The required number-of-parts calculation unit 2C can obtain the total number of requests PNAL for the parts Pa by adding the required number ADSav of each aircraft for all the aircrafts to be subjected to the next inspection. In the example shown in FIG. 7, the total required number PNAL of parts Pa in the next inspection is 6.6. The demand prediction apparatus 1 can determine the total number of requests PNAL for all parts required in the next inspection by executing the above-described processing for all parts.

  The demand forecasting apparatus 1 and the demand forecasting method according to the present embodiment are the average value regarding the usage time of the machine (flight time in the case of an aircraft), the number of parts required for inspection of a machine, for example, an aircraft. Then, the predicted usage time tg obtained from the average usage time tav is used. For this reason, the demand prediction method according to the demand prediction device 1 and the present embodiment is compared with the method of obtaining the quantity of each part required at the time of inspection based on the average value of past use results, Even if the change cannot be predicted accurately, the prediction accuracy of the total request number PNAL, which is the total request amount, can be improved. In addition, the demand prediction apparatus 1 and the demand prediction method according to the present embodiment can improve the prediction accuracy as the overall required amount even if the secular change of each part cannot be strictly predicted.

(Embodiment 2)
The second embodiment is the same as the first embodiment, but the requested number ADSav is classified for each use region of the machine, and the next time from the predicted use time tg and the request number ADSav corresponding to the predicted use time tg and the use region, The difference is that the total required number PNAL, which is the number of parts required in the inspection, is obtained. Except for this point, the second embodiment is the same as the first embodiment.

  FIG. 8 is a flowchart illustrating a procedure of the demand prediction method according to the second embodiment. 9 to 11 are diagrams for explaining processing in the demand prediction method according to the second embodiment. The demand prediction method according to the present embodiment is realized by the demand prediction apparatus 1 shown in FIG. When the demand prediction apparatus 1 shown in FIG. 1 executes the demand prediction method according to the present embodiment, the processing unit 2 reads the demand prediction computer program 11 from the storage unit and executes instructions described therein. Since step S201 and step S202 of the demand prediction method according to the present embodiment are the same as the demand prediction method according to the first embodiment, description thereof will be omitted.

  In step S203, the required number calculation unit 2B of the processing unit 2 shown in FIG. 1 requests the request per aircraft in the previous inspection for at least one part used in the aircraft as the machine to be inspected. The number ADSav is determined according to the time the aircraft has been used. In the present embodiment, the requested number calculation unit 2B further classifies the obtained requested number ADSav for each use area (flight area) of the aircraft. In the example shown in FIG. 9, the use area of the aircraft is set to three places of X, Y, and Z, but the use area is not limited to three places.

  FIG. 9 is used for aircraft in the range where the flight time, that is, the use time RT is 1200 hours (h) or more and less than 1400 hours (h) among the aircrafts inspected at the time of inspection in 2000, 2004, and 2008. The number of the parts Pa is described for each use region. MN in FIG. 9 represents the number of aircraft. For example, the information regarding the area of use is described in the third database of the storage unit 3 shown in FIG.

  As shown in FIG. 9, in 2000, the number of aircraft MN and the number of usage AD are 0 in the usage areas X and Y, and in the usage area Z, the number of aircraft MN is 1 and the number of usage AD is 6. In 2004, in the area of use X, the number of aircraft MN is 2, and the number of uses AD is 0. In the area of use Y, the number of aircraft MN is 2, the number of uses AD is 1, and in the area of use Z, the number of aircraft MN. Is 1, and the usage number AD is 2. In 2008, in the area of use X, the number of aircraft MN is 1 and the number of uses AD is 0. In the area of use Y, the number of aircraft MN is 0 and the number of uses AD is 0. Is 0, and the usage number AD is 0. In this way, the requested number calculation unit 2B classifies the obtained use number AD for each use region of the aircraft.

  From these results, the requested number calculation unit 2B obtains the requested number ADSav. In the present embodiment, the required number ADSav is obtained for each use region X, Y, Z and for each range of different use time RT. For example, in the range where the usage time RT is 1200 hours (h) or more and less than 1400 hours (h), the total usage number ADS in the usage region X is 0, and the total number of aircrafts MNA is 3, so the requested number ADSav as shown in FIG. Is 0 (pieces). Since the total use number ADS of the use area Y is 1 and the total number of aircrafts MNA is 2, the required number ADSav is 0.5 (pieces) as shown in FIG. Since the total use number ADS of the use area Z is 8 and the total number of aircrafts MNA is 2, as shown in FIG. 10, the required number ADSav is 4 (pieces). FIG. 10 shows the required number ADSav of parts Pa for different usage regions and different usage time ranges obtained by the required number calculation unit 2B.

  When the number of parts subject to demand prediction is 2 or more, the required number calculation unit 2B calculates the required number ADSav for each range of each usage time RT for each part. As described above, in the present embodiment, the requested number calculation unit 2B classifies the requested number ADSav for each time range in which the machine is actually used, that is, for each time range and usage region in which the aircraft actually flew.

  When the required number ADSav is obtained for each range of use time RT and for each use region for all parts that are targets of demand prediction in the next inspection, the process proceeds to step S204. In step S204, the required number-of-parts calculation unit 2C of the processing unit 2 included in the demand prediction apparatus 1 shown in FIG. 2 performs the number of parts required for the next inspection (total number of requests) for all aircraft to be inspected. ) Find PNAL. In the present embodiment, the required number-of-parts calculation unit 2C obtains the total required number PNAL from the predicted usage time tg and the predicted usage time tg and the required number ADSav corresponding to the usage region.

  FIG. 11 shows the relationship between the predicted usage time tg, the requested number ADSav corresponding to the predicted usage time tg, and the usage region PL for nine aircraft with aircraft numbers MT of A to H. Also in the present embodiment, it is assumed that nine aircrafts having an aircraft number MT of A to H are targets for the next inspection. In FIG. 11, A / X indicates that an aircraft with an aircraft number MT of A was flying in the usage region X, and C / Z was an aircraft having an aircraft number MT of C flying in the usage region Y. It shows that.

  For example, in the relationship between the usage time range obtained in step S203 and the required number ADSav corresponding to the usage region X, Y, Z, the required number-of-parts calculation unit 2C determines which usage the predicted usage time tg of each aircraft has. Find out if it belongs to a time range. The requested number ADSav corresponding to the usage time range and the usage region to which the predicted usage time tg belongs is the required number ADSav required for the aircraft of the predicted usage time tg.

  For example, for the part Pa, the required number ADSav can be obtained in consideration of the use area of the aircraft by giving the predicted use time tg of each aircraft to the relationship shown in FIG. The required number-of-parts calculation unit 2C can obtain the total number of requests PNAL for the parts Pa by adding the required number ADSav of each aircraft for all the aircrafts to be subjected to the next inspection. In the example shown in FIG. 11, the total required number PNAL of parts Pa in the next inspection is 5.2. The demand prediction apparatus 1 can determine the total number of requests PNAL for all parts required in the next inspection by executing the above-described processing for all parts.

  Depending on the area of use of the aircraft, the impact on the aircraft and aircraft components may also vary. For this reason, in the second embodiment, the total number of requests PNAL is obtained in consideration of the area where the aircraft is used, so that the accuracy of demand prediction can be improved in consideration of the influence according to the area used.

  In the present embodiment, the total number of requests PNAL obtained in consideration of the area where the aircraft is used may be used. However, the total number of requests PNAL obtained without considering the area where the aircraft is used, that is, the demand according to the first embodiment. Compared with the total number of requests PNAL obtained by the prediction method, the one closer to the past actual value may be adopted in the next inspection. Steps S205 and S206 are processes for realizing this method.

  In step S205, the processing unit 2 obtains past performance values. For example, the processing unit 2 acquires the number of parts actually used in the past inspection from the fourth database 12d of the storage unit 3, and statistically processes the acquired information to obtain a statistical value. This statistical value is used as a past actual value. The statistical value is, for example, an average value of the number of parts used in the past inspection. Then, the processing unit 2 compares the past actual value with the total number of requests PNAL in consideration of the use area of the aircraft obtained in step S204. Further, the processing unit 2 compares the past actual value with the total request number PNAL obtained without considering the area where the aircraft is used.

  In step S <b> 206, the processing unit 2 calculates the next request value PNAL that takes into account the area where the aircraft is used and the total request number PNAL that is calculated without taking the area where the aircraft is used into Adopted for inspection. By doing in this way, what is closer to the past actual value can be used as the total number of requests PNAL in the next inspection, so that the accuracy of demand prediction can be further improved.

(Embodiment 3)
The second embodiment is the same as the second embodiment, but the requested number ADSav is classified according to the usage pace of the machine instead of being classified according to the usage region, and corresponds to the predicted usage time tg, the predicted usage time tg, and the usage pace. The difference is that the total required number PNAL, which is the number of parts required in the next inspection, is obtained from the required number ADSav. Except for this point, the third embodiment is the same as the second embodiment.

  12 to 14 are diagrams for explaining processing in the demand prediction method according to the third embodiment. The demand prediction method according to the present embodiment is realized by the demand prediction apparatus 1 shown in FIG. The process of the demand prediction method according to the present embodiment is as shown in the flowchart of FIG. 8 showing the process of the demand prediction method according to the second embodiment. In the following, a description will be given centering on differences from the second embodiment.

  In step S203, the required number calculation unit 2B of the processing unit 2 shown in FIG. 1 requests the request per aircraft in the previous inspection for at least one part used in the aircraft as the machine to be inspected. The number ADSav is determined according to the time the aircraft has been used. In the present embodiment, the requested number calculation unit 2B further classifies the obtained requested number ADSav for each use pace of the aircraft. Information on the pace of use is described, for example, in the third database of the storage unit 3 shown in FIG.

  As shown in FIG. 12, the usage pace is set to 400 hours / year as a reference value PS, and when the number of years exceeding 400 hours / year is zero between tests, PS × 0, the year exceeding 400 hours / year Is PS × 1 when it is once, and PS × 2 when the year exceeding 400 hours / year is twice. In the example shown in FIG. 12, the usage pace of the aircraft is three stages of PS × 0, PS × 1, and PS × 2, but the usage pace is not limited to three stages. Also, the definition of the pace of use is not limited to the example of this embodiment.

  FIG. 12 is used for an aircraft in the range where the flight time, that is, the use time RT is 1200 hours (h) or more and less than 1400 hours (h) among the aircrafts inspected at the time of inspection in 2000, 2004, and 2008. The number of parts Pa is described for each use pace. MN in FIG. 12 represents the number of aircraft.

  As shown in FIG. 12, in 2000, the usage pace is PS × 0 and PS × 1, the number of aircraft MN and the usage number AD are 0, the usage pace is PS × 2, the number of aircraft MN is 1, and the usage is The number AD is 6. In 2004, when the usage pace is PS × 0, the number of aircraft MN is 2, and the usage number AD is 0. When the usage pace is PS × 1, the number of aircraft MN is 2, the usage number AD is 1, and the usage pace is Is PS × 2, the number of aircraft MN is 1, and the usage number AD is 2. In 2008, in the area of use X, the number of aircraft MN is 1 and the number of uses AD is 0. In the area of use Y, the number of aircraft MN is 0 and the number of uses AD is 0. Is 0, and the usage number AD is 0. In this way, the requested number calculation unit 2B classifies the obtained use number AD for each use pace of the aircraft.

  From these results, the requested number calculation unit 2B obtains the requested number ADSav. In the present embodiment, the required number ADSav is obtained for each usage pace and for each range of different usage time RT. For example, in the range where the usage time RT is 1200 hours (h) or more and less than 1400 hours (h), the total usage number ADS of the usage pace PS × 0 is 0, and the total number of aircrafts MNA is 3, so the request is shown in FIG. The number ADSav is 0 (number). Since the total usage number ADS of the usage pace PS × 1 is 1 and the total aircraft number MNA is 2, the required number ADSav is 0.5 (pieces) as shown in FIG. Since the total usage number ADS of the usage pace PS × 2 is 8 and the total number of aircrafts MNA is 2, the required number ADSav is 4 (pieces) as shown in FIG. FIG. 10 shows the required number ADSav of parts Pa for different usage paces and different usage time ranges obtained by the required number calculation unit 2B.

  When the number of parts subject to demand prediction is 2 or more, the required number calculation unit 2B calculates the required number ADSav for each range of each usage time RT for each part. As described above, in the present embodiment, the requested number calculation unit 2B classifies the requested number ADSav for each time range in which the machine is actually used, that is, for each time range and usage region in which the aircraft actually flew.

  When the required number ADSav is obtained for each range of usage time RT and for each usage pace for all parts that are targets of demand prediction in the next inspection, the process proceeds to step S204. In step S204, the required number-of-parts calculation unit 2C of the processing unit 2 included in the demand prediction apparatus 1 shown in FIG. 2 performs the number of parts required for the next inspection (total number of requests) for all aircraft to be inspected. ) Find PNAL. In the present embodiment, the required number-of-parts calculation unit 2C calculates the total required number PNAL from the predicted usage time tg and the required number ADSav corresponding to the predicted usage time tg and the usage pace.

  FIG. 14 shows the relationship between the predicted usage time tg, the requested number ADSav corresponding to the predicted usage time tg, and the usage pace PS for nine aircraft with aircraft numbers MT of A to H. Also in the present embodiment, it is assumed that nine aircrafts having an aircraft number MT of A to H are targets for the next inspection. In FIG. 13, A / 0 indicates that the aircraft whose aircraft number MT is A was flying at a usage pace of PS × 0, and C / 2 is that the aircraft whose aircraft number MT is C was PS × It shows that it was flying by 2.

  For example, in the relationship between the range of use time obtained in step S203 and the required number ADSav corresponding to the pace of use, the required number-of-parts calculation unit 2C belongs to which use time range the predicted use time tg of each aircraft belongs to Ask for. The required number ADSav corresponding to the range of usage time and the usage pace to which the predicted usage time tg belongs is the required number ADSav required for the aircraft of the predicted usage time tg.

  For example, for the part Pa, the required number ADSav is obtained in consideration of the pace of use of the aircraft by giving the predicted use time tg of each aircraft to the relationship shown in FIG. The required number-of-parts calculation unit 2C can obtain the total number of requests PNAL for the parts Pa by adding the required number ADSav of each aircraft for all the aircrafts to be subjected to the next inspection. In the example shown in FIG. 14, the total required number PNAL of parts Pa in the next inspection is 8.2. The demand prediction apparatus 1 can determine the total number of requests PNAL for all parts required in the next inspection by executing the above-described processing for all parts.

  Depending on the pace of use of the aircraft, the impact on the aircraft and aircraft components may also vary. For this reason, in the third embodiment, the total number of requests PNAL is obtained in consideration of the pace of use of the aircraft, and therefore the accuracy of demand prediction can be improved in consideration of the influence according to the pace of use.

  In the present embodiment, the total number of requests PNAL obtained in consideration of the pace of use of the aircraft may be used. However, the total number of requests PNAL obtained without considering the pace of use of the aircraft, that is, the demand according to the first embodiment. Compared with the total number of requests PNAL obtained by the prediction method, the one closer to the past actual value may be adopted in the next inspection.

  In step S205, the processing unit 2 obtains past performance values. For example, the processing unit 2 acquires the number of parts actually used in the past inspection from the fourth database 12d of the storage unit 3, and statistically processes the acquired information to obtain a statistical value. This statistical value is used as a past actual value. The statistical value is, for example, an average value of the number of parts used in the past inspection. Then, the processing unit 2 compares the past actual value with the total request number PNAL in consideration of the aircraft use pace obtained in step S204. Further, the processing unit 2 compares the past actual value with the total request number PNAL obtained without considering the pace of use of the aircraft.

  In step S <b> 206, the processing unit 2 calculates the next request value PNAL that takes into account the pace of use of the aircraft and the total request number PNAL that is obtained without taking into account the pace of use of the aircraft, Adopted for inspection. By doing in this way, what is closer to the past actual value can be used as the total number of requests PNAL in the next inspection, so that the accuracy of demand prediction can be further improved.

  In the present embodiment, the total request number PNAL obtained in consideration of the aircraft use pace, the total request number PNAL obtained in consideration of the aircraft use region, and the aircraft use pace and use region are not taken into consideration. The total number of requests PNAL may be compared and a value close to the past actual value may be adopted in the next inspection. By doing in this way, the candidate of the total request number PNAL in demand prediction can be increased.

Further, in the present embodiment, the demand prediction apparatus 1 may classify the requested number ADSav for each aircraft use pace and each use region, and obtain the total request number PNAL in consideration of the aircraft use pace and the use region. . For example, in the example shown in FIG. 12, each usage pace is further classified for each usage region, and the total number of requests PNAL is obtained by totaling the number of requests ADSav. Further, in the second embodiment, each usage region may be further classified for each usage pace, and the total number of requests PNAL may be obtained by counting the number of requests ADSav. By doing in this way, the candidate of the total request number PNAL in demand prediction can be increased. In addition to the first embodiment, by increasing n classification criteria, 2 ⁿ total request number PNAL candidates can be generated (n is an integer of 1 or more). The classification standard is, for example, the above-described area of use of the aircraft and the pace of use of the aircraft.

  As described above, the present invention is not limited by the contents described in the first to third embodiments. The constituent elements of the first to third embodiments include those that can be easily assumed by those skilled in the art, those that are substantially the same, and those that are in an equivalent range. Furthermore, the constituent elements described in the first to third embodiments can be appropriately combined. In addition, various omissions, substitutions, or changes of the constituent elements can be made without departing from the spirit of the first to third embodiments.

DESCRIPTION OF SYMBOLS 1 Demand prediction apparatus 2 Processing part 2A Prediction part 2B Required number calculation part 2C Necessary part number calculation part 3 Storage part 4 Input / output part 5 Display device 6 Input device 7 Communication line 11 Computer program for demand prediction

Claims (13)

A prediction unit that obtains a predicted usage time that is predicted to be used by the next inspection after the machine to be inspected after the last inspection;
For at least one part used in the machine, a required number calculation unit for obtaining the required number per machine in the previous inspection according to the time the machine was used;
From the predicted usage time and the required number corresponding to the predicted usage time, a required component number calculation unit for obtaining the number of components required in the next inspection,
Demand forecasting device including The requested number calculation unit
The demand prediction apparatus according to claim 1, wherein the requested number is classified for each time range in which the machine is actually used. The required number of parts calculation unit
The demand according to claim 1 or 2, wherein the number of parts required in the next inspection is obtained from the predicted usage time, and the required usage time corresponding to the predicted usage time and the usage area of the machine. Prediction device. The required number of parts calculation unit
The demand according to claim 1 or 2, wherein the number of parts required in the next inspection is obtained from the predicted usage time and the required number corresponding to the predicted usage time and the pace of use of the machine. Prediction device. The required number of parts calculation unit
The number of parts required in the next inspection is determined from the predicted usage time, the predicted usage time, the usage area of the machine, and the required number corresponding to the usage pace of the machine. Item 3. The demand prediction device according to Item 2.   Of the number of parts required in the next inspection considering the use pace or use area of the machine and the number of parts required in the next inspection not considering the use pace or use area of the machine, The demand forecasting device according to claim 5 which chooses a nearer past performance.   Of the number of parts required in the next inspection considering the use pace and use area of the machine, and the number of parts required in the next inspection not considering the use pace and use area of the machine, The demand forecasting device according to claim 5 which chooses a nearer past performance. Find the expected usage time that the machine to be inspected is expected to be used after the last inspection until the next inspection,
With respect to at least one part used in the machine, a required number per one machine in the inspection so far is obtained according to a time when the machine is used,
The demand prediction method which calculates | requires the number of said components required in the said next test | inspection from the said estimated usage time and the said request | requirement number corresponding to the said estimated usage time.   The demand prediction method according to claim 8, wherein the required number is classified for each time range in which the machine is actually used.   The demand according to claim 8 or 9, wherein the number of parts required in the next inspection is obtained from the predicted usage time and the required number corresponding to the predicted usage time and the area where the machine is used. Prediction method.   The demand according to claim 8 or 9, wherein the number of parts required in the next inspection is obtained from the predicted usage time and the required number corresponding to the predicted usage time and the pace of use of the machine. Prediction method.   The number of parts required in the next inspection is obtained from the predicted usage time, the predicted usage time, the usage area of the machine, and the required number corresponding to the usage pace of the machine. Item 10. The demand prediction method according to Item 9.   The computer program for demand prediction which makes a computer perform the demand prediction method of any one of Claims 8-11.

Images