JP2003178353A - Collected data processing method and collected data processing system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technique for quickly and precisely performing, in the occurrence of an error by a line abnormality or the like in the collection of data, the following job while eliminating its influence. <P>SOLUTION: The data received by a communication part 110 are transmitted to a data collection part 120, and after confirming the presence of collection error and the presence of collection error in the previous collection, the data are stored in a collected data storage part 112a while adding flags thereto as occasion demands. Thereafter, (i) a data replenishment part 113 performs data replenishment for the data having the collection error, and (ii) a data correction part 114 performs data correction for those in which the collection error was generated in the previous collection but the data are normally collected in this collection. The replenished and corrected data are stored in a collected data storage part 112b. An information analyzing device 115 reads these data as needed and uses them as materials for information analysis. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for collecting data from a plurality of devices connected via a network and then quickly executing a subsequent job based on the collected data.

[0002]

2. Description of the Related Art Conventionally, in a system that collects data from a plurality of terminals on a network, a subsequent job is interrupted until the abnormality in the communication is eliminated as a countermeasure when the abnormality in the communication occurs due to a communication failure or the like. Action was taken, or action was taken to execute the subsequent job after eliminating the communication abnormality. Among these, various proposals have been made so far regarding the technology for promptly and surely eliminating the abnormal reception. For example, there is a case in which the saved data is returned by a certain procedure called a so-called recovery process so that the subsequent process may be performed without receiving the information including the error data.

However, it is difficult to promptly collect information and perform subsequent processing by such measures.
In particular, in a large-scale network system, when a plurality of locations have abnormalities of simultaneous reception, it is very difficult to deal with them.

In addition, as the network environment is rapidly changing, the current B2
It is expected that the B-business will extend far beyond the boundaries of companies, that is, unspecified, and enormous numbers of users will be connected to each other worldwide, and that physical communication will occur. It is conceivable that the frequency of errors will increase dramatically. In the future, it is hoped that countermeasures will be taken against abnormalities in concentration that can cope with such situations.

On the other hand, in recent years, the manner of utilizing the collection information has become diverse. For example, EOS (Electric Orde
ring system) In the information collection by the EDI server in business, since the accuracy of the information is important, the following job is suspended until the abnormality is resolved, or the abnormality is resolved. It is indispensable to take measures to execute subsequent jobs after doing so. However, in a business in which the next action is executed as quickly as possible based on the received information, the speed of the succeeding job is required to be higher than the accuracy of the information. In such a case, the development of a technology to eliminate the influence of the received data and perform subsequent jobs quickly and accurately becomes an important technical issue, beyond the consideration of a method for quickly removing the received data. .

[0006]

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and when an error occurs due to a line abnormality when collecting data, the succeeding job is performed while eliminating the influence thereof. It provides a technique for quickly and accurately performing.

[0007]

According to the present invention, there is provided a method for collecting data from a plurality of devices connected via a network and then executing a subsequent job based on the collected data. The step of collecting data from a plurality of devices, the step of extracting the abnormal concentrator device for which no data was obtained, and the process of normally collecting data from the abnormal concentrator device to the data missing part of the abnormal concentrator device. It is characterized by including a step of replenishing inferred data that is estimated to be obtained when acquired, and a step of executing the subsequent job based on the concentrated data group after replenishing the inferred data. A method of processing concentrated data is provided.

Further, according to the present invention, there are provided a plurality of devices and a concentrated data processing device group for executing a subsequent job based on the data collected from the device, wherein the concentrated data processing device group comprises the plurality of devices. Data gathering means for gathering data from the device, the gathering abnormality extracting means for extracting the gathering abnormality device from which no data was obtained, and the gathering abnormality for the data missing part of the gathering abnormality device. Data replenishing means for replenishing inferred data estimated to be obtained when data is normally acquired from the device, storage means for storing a group of concentrated data after replenishing the inferred data, and replenished for the inferred data And a succeeding job executing unit that executes the succeeding job based on the concentrated collection data group.

Here, the concentrated data processing device group may be composed of a single information processing device or may be configured by connecting a plurality of information processing devices via a network.

In the present invention, for the data missing part,
Since the succeeding job is executed after supplementing the guess data that is supposed to be obtained when the data is normally acquired, the succeeding job is executed promptly and accurately while eliminating the influence of the concentrated data. be able to. Conventionally, regarding the handling of the data missing portion, a subsequent job is interrupted until normal data is acquired, or line restoration processing is performed so that normal data is acquired. On the other hand, in the present invention, the promptness of the succeeding job is ensured by supplementing the estimated data with the data missing portion. Moreover, since this estimated data is data that is estimated to be obtained when the data is normally acquired, the accuracy of the succeeding job can be guaranteed to a certain extent.

In the present invention, data collection is usually
A configuration that is regularly or irregularly repeated is adopted. The estimated data can be supplemented based on the past collection results. The estimated data in the present invention can be, for example, an average value of the collected data of the abnormal collection device for a certain past period. In this way, accurate estimation data can be easily obtained.

In the concentrated data processing method according to the present invention, when normal data is obtained by the subsequent concentrated processing in the data missing portion, the estimated data is replaced with the correction data calculated based on the normal data. A replacement configuration can be adopted. Further, in the concentrated data processing system according to the present invention, in the data missing portion, when the normal data is obtained by the subsequent concentrated processing, the estimated data is replaced with the correction data calculated based on the normal data. It is possible to adopt a configuration further including a data correction unit that stores the correction data in the storage unit after replacement. With this configuration, the estimated data can be supplemented on the basis of the more accurate reception record in the subsequent reception process. That is, accurate estimation data can be obtained in the subsequent concentration processing while ensuring the promptness of the subsequent job in the certain concentration processing.
In the concentrated data processing method of the present invention, together with the data collected from the plurality of devices, offline data is collected from a source other than the plurality of devices without passing through a network, and the offline data missing portion On the other hand, it is possible to supplement the estimated data that is estimated to be obtained when the data is normally acquired from the source. Further, in the concentrated data processing system of the present invention, offline data collecting means for collecting offline data together with data collected from the plurality of devices from a collection source other than the plurality of devices without using a network. And offline data replenishing means for replenishing the inferred data estimated to be obtained when the data is normally acquired from the concentrator with respect to the missing portion of the offline data. it can. By doing this, it is possible to handle the received data by collective processing in a system that has both online and offline devices.
Efficient data processing can be realized without causing a delay in subsequent jobs.

[0013]

BEST MODE FOR CARRYING OUT THE INVENTION The present invention is more effective when applied to a system in which data is periodically collected from devices connected via a network. This is because in such a system, the estimated data to be filled in the data missing portion can be more accurately obtained based on the data collected in the past.

The present invention is more effective when applied to a system or the like in which a succeeding job can be executed if data with a certain degree of accuracy is obtained, such as simulation. In such a system, it is not necessary to acquire definite data, but quickness of data processing is required, so that the effect of the present invention is remarkably exhibited.

According to the present invention, for example, a sales forecast is made on the basis of sales record data transmitted from a plurality of selling devices at regular intervals, and the timing and amount of replenishment of products to each selling device are determined based on the sales forecast. Can be suitably applied to the system. In such a system, since the data is regularly collected, it is possible to more accurately obtain the inferred data to be filled in the data missing portion, and the collected data is used for the simulation of the sales forecast, so the confirmed data is This is because it does not need to be acquired.

The present invention will be described below by taking as an example a system that collects information on the sales volume from a plurality of vending machines, and based on this information, executes a sales forecast and a product loading plan formulation for the vending machines. explain.

FIG. 1 shows an automatic vending machine 101 and a distribution server 1.
2 is a diagram showing a network system including 02, a network 103, and a concentrating server 104. FIG. here,
In the present embodiment, the collection server 104 is the vending machine 10.
1 and the distribution server 102 include a form in which they are wirelessly connected, a form in which they are connected by wire, and an offline form in which they are not connected to the server.

The inside of the concentrating server 104 can take various forms, and for example, it can be configured as shown in FIG. In FIG. 2, the concentration server 104 includes a communication unit 110, a data concentration unit 120, a concentration data storage unit 112a, and a concentration data storage unit 112b. Data collection unit 12
0 is the data processing unit 121 and the data correction unit 1
22 are included.

The data received by the communication section 110 is sent to the data concentrating section 120, and the presence or absence of a concentrating error and the presence or absence of a concentrating error at the time of the previous concentrating are checked, and a flag is added if necessary. And stores the data in the received data storage unit 112a. This process is executed as a transaction process. Regarding the data stored in the collected data storage unit 112a, the data supplement unit 113 and the data correction unit 114
The following processing is performed by.

(I) The data replenishing unit 113 replenishes the data for those having a collection error.

(Ii) The data correction unit 114 corrects the data for which the reception error has occurred in the previous reception but the data has been normally received by the current reception.

The data thus supplemented and corrected is stored in the received data storage section 112b. The information analysis device 115 reads this data from time to time and utilizes it as a material for information analysis. Since this embodiment relates to a system for managing vending machines, the analysis by the information analysis device 115 is used for simulation of sales trends and the like. In managing vending machines, it is desirable to check the daily sales volume and accurately grasp the inventory quantity before formulating a plan for loading products into the vending machine. According to the present embodiment, the latest sales volume data is always stored in the collection data storage unit 112, and appropriate values are replenished for the data missing due to the collection error. Without stopping, it is possible to perform a series of processes from collecting information, grasping sales trends, simulating sales trends, and formulating a loading plan.

Next, data processing by the error check unit 116 and the flag addition unit 117 in the collection server 104 will be described with reference to FIG. After the start of the collection, the error check unit 116 in FIG. 2 confirms the presence or absence of the collection error (FIG. 3, S101). When a collection error is confirmed, the flag addition unit 117 of FIG. 2 causes the error flag in the item of the sender vending machine in which the error has occurred in the collection data table stored in the collection data storage unit 112a. Is added (S102). Further, the data search unit 118 searches for the collection data at the previous collection from the collection data stored in the collection data storage unit 112,
Whether or not there is an error is confirmed (S103). As a result, an error at the time of the previous reception was confirmed, and a supplement flag is added to the vending machine for which the normal value was obtained in the current reception (S
104). As described above, if there is no collection error,
Those to which the error flag has been added and those to which the complement flag has been added are stored in the received data storage units 112a and 112b in FIG. 2 (S105).

Next, the flow of processing by the data supplement unit 113 and the data correction unit 114 in FIG. 2 will be described with reference to FIG.
Will be described with reference to. First, the data correction unit 122 accesses the concentrated data storage unit 112b to search the concentrated data (S201). Next, it is determined whether or not there is an error flag (S202). If there is no error flag, move to the next process. If there is an error flag, a data supplement process is performed (S203). After that, the presence or absence of the complement flag is determined (S204). If there is no complement flag, the process ends. If there is a complement flag, data correction processing is performed (S20).
5). Through the above processing, data replenishment and correction are performed.

Next, the method of processing the concentrated data according to the present invention will be described with reference to specific examples. FIG. 5 shows sales data of vending machines collected on the night of April 3. In the figure,
“VMID” means the ID number of the vending machine. This figure shows the transition of the sales volume and theoretical inventory of VMID100101. On April 1st and April 2nd, the sales volume is 10, and the inventory 60 on April 2nd is fixed. However, a communication error occurred on April 3 and the sales volume data for that day was missing. Therefore, in the present embodiment, the sales quantity on April 3 is predicted to be “10”, and this theoretical data is stored in the sales quantity cell to calculate the theoretical inventory on that day (FIG. 5). During,
Displayed as "fill in holes when abnormal". This forecast of sales volume can be, for example, an average value of past sales volumes. FIG. 8 schematically shows a method for obtaining estimated data. Here, for the data missing portion at t _n , t ₀ to t
Mean values between _n-1 are supplemented as inferred data.

FIG. 6 shows sales data of the automatic vending machines collected on the night of April 4. In the figure, up to the inventory 60 on April 2, is fixed. However, on April 3rd and April 4th, there was an abnormality in concentration, and the sales volume on that day is unknown. Therefore, in this embodiment, April 3 and 4
The theoretical inventory is calculated by predicting that the sales quantity on the 4th of a month is "10" and storing these values in the sales quantity cell.

FIG. 7 shows sales data of vending machines collected on the night of April 5th. Up to the inventory 60 on April 2 was confirmed, but an abnormality in concentration occurred on both April 3 and April 4, and accurate inventory cannot be confirmed. However, the line returned to normal on the night of April 5, and the value of "20" was obtained as the inventory for the day. Therefore, the total sales volume for 3 days from April 3 to April 5 is obtained by subtracting the inventory 20 of April 5 from the inventory 60 of April 2, and the value is "40". Become. This quantity is distributed almost equally over three days, and the sales quantity on April 3 and April 4 is 13, and the sales quantity on April 5 is 14. This correction of data ensures consistency with the theoretical inventory. In this embodiment, the quantity is distributed almost evenly for correction, but a coefficient indicating the sales ratio for each day of the week can be reflected here. FIG. 9 shows distribution of sales volume by such a method. Where t
Distribution is performed by multiplying _{k-3 to} tk _-1 by the sales ratio for each day of the week.

In the present embodiment, as shown in FIG. 2, two types of collected data storage units are provided, but they may be configured as one storage unit. FIG. 10 shows the configuration of such a concentration server 104. In this server, the collection data is stored in the collection data storage unit 112 via the communication unit 110 and the database control unit 111. The data replenishing unit 113 replenishes the received data and the data correcting unit 114 corrects the received data.

Although an example of the abnormalities in communication and the restoration to the normal state related to the online real-time processing has been described above, the same processing can be performed for an offline vending machine which is not connected to the network 103 in FIG. The sales data of the offline vending machine is collected when the products are loaded. From the viewpoint of the concentration server 104, when there is no data collection, the state is the same as the concentration abnormality in the online machine. Therefore, when the offline machine is not collecting data, the subsequent job can be smoothly executed by treating it in the same manner as the abnormality in the online machine. In this way, in a system in which online machines and offline machines coexist, it is possible to execute the concentrated data processing by batch processing. Note that FIG.
As described with reference to FIG. 4 and the like, the flag is added to the predetermined data in the present embodiment, but this is merely an example, and in short, it suffices that the abnormal reception data can be recognized. Various methods of can be adopted.

[0030]

As described above, according to the present invention, even if a collection error occurs, the succeeding job can be performed quickly and accurately while eliminating the influence of the collection data.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a concentrated data processing system according to the present invention.

FIG. 2 is a schematic configuration diagram of a concentration server that constitutes the concentration data processing system according to the present invention.

FIG. 3 is a diagram showing a concentrated data processing method according to the present invention.

FIG. 4 is a diagram showing a concentrated data processing method according to the present invention.

FIG. 5 is a diagram showing an example of a concentrated data processing method according to the present invention.

FIG. 6 is a diagram showing an example of a concentrated data processing method according to the present invention.

FIG. 7 is a diagram showing an example of a concentrated data processing method according to the present invention.

FIG. 8 is a diagram for explaining a method of calculating supplementary data.

FIG. 9 is a diagram for explaining a method of creating correction data.

FIG. 10 is a schematic configuration diagram of a concentrated data processing system according to the present invention.

[Explanation of symbols]

101 vending machine 102 distribution server 103 network 104 Concentration server 110 Communication unit 111 Database control unit 112, 112a, 112b Collected data storage unit 113 Data Replenishment Department 114 Data correction unit 115 Information analysis device 116 Error check section 117 flag addition unit 118 Data search section 120 Data collection unit 121 Data processing department 122 Data Correction Section

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Toshiya Takashima             2-2-1 Toranomon, Minato-ku, Tokyo, Japan             Bako Sangyo Co., Ltd. F-term (reference) 3E044 AA01 CC03 DC01 DE01 DE02                 5B083 BB03 CE01 GG04

Claims (6)

[Claims] 1. A method of collecting data from a plurality of devices connected via a network and then executing a subsequent job based on the collected data, the method comprising collecting data from the plurality of devices. And a step of extracting a concentration abnormality device for which no data has been obtained, and it is presumed to be obtained when data is normally acquired from the concentration abnormality device for the data missing part of the concentration abnormality device. A method for processing concentrated data, comprising: a step of supplementing estimated data; and a step of executing the subsequent job based on a concentrated data group after supplementing the estimated data. 2. The concentrated data processing method according to claim 1, wherein when the normal data is obtained by the subsequent concentrated processing, the estimated data is the correction data calculated based on the normal data. A method of processing concentrated data, characterized in that 3. The method of processing concentrated data according to claim 1, wherein offline data is collected from a collection source other than the plurality of devices together with the data collected from the plurality of devices without passing through a network. A method of collecting data, comprising collecting data and supplementing the missing portion of the offline data with estimated data that is estimated to be obtained when data is normally acquired from the collection source. 4. A plurality of devices, and a concentrated data processing device group for executing a subsequent job based on data collected from the device, wherein the concentrated data processing device group receives data from the plurality of devices. A data collecting means for collecting data, a collection abnormality extracting means for extracting a collection abnormality device for which no data has been obtained, and a data missing portion of the collection abnormality device, which normally receives data from the collection abnormality device. Data replenishing means for replenishing the inferred data estimated to be obtained, storage means for storing the concentrated data group after replenishing the inferred data, and concentrating data after replenishing the inferred data A succeeding job executing means for executing the succeeding job based on the data group,
A concentrating data processing system comprising: 5. The concentrated data processing system according to claim 4, wherein when the normal data is obtained by the subsequent concentrated processing, the estimated data is corrected data calculated based on the normal data. The received data processing system further comprises a data correction unit that stores the correction data in the storage unit after the replacement. 6. The collected data processing system according to claim 4, wherein the data collected from the plurality of devices is combined with offline data from a source other than the plurality of devices without passing through a network. Offline data collecting means for collecting data; and offline data replenishing means for replenishing inferred data estimated to be obtained when the data is normally acquired from the concentrating source with respect to the missing portion of the offline data. A collection data processing system characterized by further comprising.