JP2008263237A - Method and device for managing alarm - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily investigate a cause of alarm occurrence. <P>SOLUTION: An alarm managing device 20 is provided with a data collecting device 22 collecting alarm data which a device of management object 10 emits and device data showing a state of the device of management object 10, a data base device 24 storing data collected by the device 22 and an alarm statistic device 26 generating an alarm statistic by using data stored in the device 24. The alarm statistic device 26 is provided with a link means linking an alarm which the device of management object 10 emits and device data showing the state of the device of management object 10 with generated date/time of the alarm as a key, a trend graph generating means generating a trend graph of prescribed device data in device data and a display means showing the trend graph generated by the trend graph generating means and showing date/time when the previously designated alarm occurs on the trend graph based on the link. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a method and apparatus for managing alarms by creating statistics of alarms issued from managed devices. Examples of the management target device include a semiconductor manufacturing device for manufacturing and testing semiconductors, a liquid crystal display manufacturing device for manufacturing and testing liquid crystal displays, and an ion beam irradiation device for these devices.

  When managing alarms generated by managed devices as described above, conventionally, the number of alarm occurrences is simply accumulated to create a list (so-called worst list) in which the alarms are arranged in descending order, or a percentage graph. And charts for analysis such as Pareto charts.

  In addition, as a filtering method for designating (in other words, extracting) the data subject to the number of alarm occurrences, filtering is usually performed by date or time range, but in addition, lot ID (lot identification information) There is also proposed a technique of performing filtering by using (see, for example, Patent Document 1).

JP-A-6-332910 (paragraph numbers 0008 and 0012, FIG. 1)

  The conventional alarm management techniques as described above have the following problems (1) and (2).

(1) The way of taking alarm statistics does not match the purpose The purpose of taking alarm statistics is mainly to improve the utilization efficiency of the management target device by using the statistics. Various troubles, failures, malfunctions, and the like that cause a reduction in usage efficiency are recognized as alarms by the device user and the device supplier. Then, the device user or the device supplier analyzes the alarm statistics and tries to improve the device utilization efficiency by taking measures sequentially from the ones that seem to be serious. Therefore, the alarm statistics need to reflect the degree of influence of each alarm on the device usage efficiency as accurately as possible.

  However, in the conventional alarm statistics, since the influence degree for each alarm is not considered, it is simply interpreted that “the alarm with the largest number is the most serious”. And this interpretation is often incorrect. An example is given below.

  For example, in a certain period, the number of occurrences of a certain alarm (for example, alarm ID 1001) is 20, the average device stop time per alarm is 1 hour, and other alarms (for example, alarm ID 3050) Assume that the number of occurrences is 5, and the average device stop time per alarm is 10 hours. According to the conventional alarm statistics, it is interpreted that the alarm with the alarm ID = 1001 is four times more serious, but the apparatus stop time by each alarm during the period is 20 hours by the alarm with the alarm ID = 1001. From the point of view of the degree of influence on the utilization efficiency of the management target device, the alarm of alarm ID = 3050 is the second alarm of alarm ID = 1001. 5 times larger.

  In the above example, alarm statistical analysis aimed at improving the utilization efficiency of managed devices was described, but there are various other purposes such as reducing maintenance costs of managed devices and reducing the number of maintenance personnel, It is sufficient if there is a method for evaluating the degree of influence suitable for each purpose, but such an evaluation cannot be performed by the conventional alarm management method that performs the alarm statistics as described above.

(2) Alarm statistics cannot be used to investigate the cause of alarm occurrence Another problem with conventional alarm management techniques is the poor link (combination) function with device data representing the status of managed devices. Once a critical alarm to be addressed is identified, the role of alarm statistics does not end there. The alarm statistics only need to be useful for identifying the cause of alarm occurrence, but it is difficult with conventional alarm statistics. This is because the alarm data and the device data are not well linked.

  In this regard, the technique of filtering alarms by lot ID as shown in Patent Document 1 is not sufficient. This is because the lot ID is associated with a plurality of device data, and the lot ID and the plurality of device data are not necessarily uniquely connected. That is, just because the lot ID is the same, the operation parameters of the apparatus when the lot is processed are not always the same. More specifically, even if a process recipe that is a processing condition when the lot is processed can be identified from the lot ID, the process recipe is realized by a plurality of apparatus operation parameters (this can also be regarded as apparatus data). Even if some of the operating parameters seem to be related to the cause of the alarm, they may have the same value in different process recipes. In the process recipe unit filtering according to, the accuracy is too low, and it may be difficult to specify the operation parameter causing the alarm.

  Therefore, the main object of the present invention is to facilitate the investigation of the cause of alarm occurrence.

  One of the alarm management methods according to the present invention is an alarm management method for managing an alarm issued by a management target device, wherein (A) a data collection device displays alarm data generated by the management target device and a state of the management target device. Collecting the device data to be represented and storing the collected data in the database device; (B) the alarm statistics device using the data stored in the database device and the alarm issued by the device to be managed and the management Based on the link, the device data representing the state of the target device are linked to each other using the alarm occurrence date as a key, the trend graph of predetermined device data in the device data is created and displayed, and On the trend graph, the date and time when the alarm specified in advance is displayed. -Up, is characterized in that it comprises.

  According to this alarm management method, the alarm and device data generated by the device to be managed are linked to each other using the alarm occurrence date and time as a key, so the relationship between the alarm occurrence and the predetermined device data can be known. That is, it is possible to easily know at what value the predetermined device data is generated by using the trend graph together with the alarm. As a result, it becomes easier to investigate the cause of the alarm occurrence.

  Another one of the alarm management methods according to the present invention is an alarm management method for managing an alarm issued by a management target device, wherein (A) a data collection device includes alarm data generated by the management target device and the management target device. Collecting device data representing a state and storing the collected data in a database device; and (B) the alarm statistics device emits the managed device using the data stored in the database device The statistics of alarms are created, and only the alarms that are generated when the predetermined device data of the device data take a predetermined value or a range of values when creating the statistics are targeted for statistics. The step of carrying out is provided.

  According to this alarm management method, it is possible to know the state of occurrence of an alarm when the management target device is in a specific operation state based on the statistics created as described above, and therefore it becomes easy to investigate the cause of the alarm. . The statistics created as described above are also useful when it is desired to preferentially take countermeasures only for alarms in a specific situation.

  One of the alarm management devices according to the present invention is an alarm management device that manages an alarm issued by a management target device, and is a data collection that collects alarm data issued by the management target device and device data representing a state of the management target device Apparatus, a database device that stores data collected by the data collection device, and an alarm statistics device that creates alarm statistics using the data stored in the database device, and the alarm statistics device comprises: Link means for linking an alarm generated by the management target device and device data representing the status of the management target device with each other using an alarm occurrence date as a key, and creating a trend graph of predetermined device data in the device data Created by trend graph creation means and this trend graph creation means And displays the trend graph, on the trend graph on the basis of the link, is characterized in that it comprises a display means for displaying the date and time the alarm was generated previously specified.

  According to this alarm management device, since the alarm and device data generated by the device to be managed are linked to each other using the alarm occurrence date and time as a key, the relationship between the occurrence of the alarm and predetermined device data can be known. That is, it is possible to easily know at what value the predetermined device data is generated by using the trend graph together with the alarm. As a result, it becomes easier to investigate the cause of the alarm occurrence.

  Another one of the alarm management devices according to the present invention is an alarm management device that manages an alarm issued by a management target device, and collects alarm data issued by the management target device and device data representing a state of the management target device. A data collection device; a database device that stores data collected by the data collection device; and an alarm statistics device that creates alarm statistics using the data stored in the database device. Is a filtering means for extracting only alarms generated when predetermined device data of the device data has a value or range of values designated in advance, and is extracted by the filtering means. And a statistical creation means for creating the statistics of alarms There.

  According to this alarm management device, it is possible to know the state of occurrence of an alarm when the management target device is in a specific operation state based on the statistics created as described above, and therefore it becomes easy to investigate the cause of the alarm. . The statistics created as described above are also useful when it is desired to preferentially take countermeasures only for alarms in a specific situation.

  Since this invention is comprised as mentioned above, there exist the following effects.

  According to the first and fourth aspects of the present invention, the alarm and device data issued by the management target device are linked to each other using the alarm occurrence date and time as a key. I can know. That is, it is possible to easily know at what value the predetermined device data is generated by using the trend graph together with the alarm. As a result, it becomes easier to investigate the cause of the alarm occurrence.

  According to the inventions of claims 2 and 5, the statistics generated as described above can know the alarm occurrence status when the management target device is in a specific operation state. Investigation becomes easy. The statistics created as described above are also useful when it is desired to preferentially take countermeasures only for alarms in a specific situation.

  According to the third and sixth aspects of the invention, the data volume to be handled can be reduced, and the alarm display can be simplified.

  FIG. 1 is a block diagram showing an example of an alarm management apparatus and a management target apparatus for implementing the alarm management method according to the present invention. An alarm management device 20 is connected to the management target device 10 via a signal transmission path 18.

  In this example, the management target device 10 includes several sub-controllers 13 to 16 that control various devices (not shown) constituting the management target device 10 and a main controller that performs overall control of these sub-controllers 13 to 16. 12 and a loop-like signal transmission path 18 connecting these controllers 12 to 16, and these constitute a network.

  Taking the case where the management target apparatus 10 is an ion implantation apparatus which is a kind of ion beam irradiation apparatus as an example, the sub-controllers 13 to 16 are as follows, for example.

  The sub-controller 13 is an implantation controller that controls ion implantation operations on the wafer, such as mechanical scanning of the wafer to be implanted, ion beam scanning, and dose amount management.

  The sub-controller 14 is a beam controller that controls various power sources for various devices for generating an ion beam and transporting it to a wafer.

  The sub-controller 15 is an end station controller that controls wafer conveyance.

  The sub-controller 16 is a system controller that controls various devices other than the power source and controls vacuum exhaust.

  Each alarm generated by the management target device 10 may be managed in a text format that directly represents the content of the alarm, but in this example, it is managed by an alarm ID that is identification information for identifying each alarm. . This is preferable because the data capacity to be handled can be reduced. Also, the alarm display is simplified. Therefore, in this example, the alarm ID is transmitted from the management target device 10 and from the management target device 10 to the alarm management device 20. The alarm ID and the alarm content that it represents are uniquely associated. In order to make it easy to grasp the relationship between the two, a list that links the alarm ID and the alarm contents in the text format represented by the alarm ID is stored in the main controller 12, for example, and is read out and used as necessary. Also good.

  In this example, in addition to the above function, the main controller 12 configures the alarm management device 20 with alarm data including the alarm ID and occurrence date and time of the alarm when the own controller or the sub-controllers 13 to 16 generates an alarm. It has a function of transmitting to the data collection device 22. When the operator removes the cause of the alarm and presses an alarm release button (not shown) of the main controller 12, the main controller 12 releases the alarm, and the alarm data including the alarm ID and the release date / time of the alarm is stored. It also has a function of transmitting to the collection device 22. That is, the contents of the alarm data generated by the management target device 10 (more specifically, in this example by the main controller 12) are an alarm ID and its occurrence date and time when an alarm occurs, and an alarm ID and its release date and time when the alarm is released.

  In this example, the alarm management device 20 includes a data collection device 22, a database device 24, and an alarm statistics device 26, which are connected to each other via a signal transmission path 19.

  The data collection device 22 collects the alarm data issued by the management target device 10. Furthermore, when realizing a link (which will be described later) between device data indicating the state of the management target device 10 and an alarm, device data emitted by the management target device 10 is also collected. The apparatus data includes, for example, the above-described process recipe, apparatus operation parameter setting values, apparatus operation parameter monitor values, and the like. The data collection device 22 performs control to collect these data and store them in the database device 24. It also has a function of notifying the alarm statistics device 26 that alarm data has been received.

  The database device 24 stores the alarm data collected by the data collection device 22 in a database. One example of the alarm data AD stored in the database device 24 is shown in FIG. This alarm data AD includes an alarm ID, an occurrence date and time, and a release date and time. The weighting factor will be described later. When realizing the link, the database device 24 also stores the device data collected by the data collection device 22 in the database.

  The alarm statistics device 26 creates alarm statistics using the alarm data stored in the database device 24 or using the alarm data and the device data. The means and the like will be described in detail below for (A) a case where a weight is assigned to each alarm and (B) a case where alarms and device data are linked.

  First, a case where a weight is assigned to each alarm (A) will be described.

  In this case, as shown in the example shown in FIG. 3, the alarm statistical device 26 includes a weighting factor assigning unit 30 that assigns one or more types of weighting factors to each alarm, and the one or more types of weighting factors for each alarm. A total weight coefficient calculating means 32 for multiplying each of the total weight coefficients by multiplication, a weighted number calculating means 34 for multiplying the total weight coefficient by 1 representing the number of cases for each alarm case, There is provided a statistic creation means 36 for accumulating the number of weighted cases for alarms and creating statistics of the number of weighted alarm occurrences.

  In this example, the alarm statistic device 26 is preferably provided. The alarm statistic device 26 is a display unit 38 for displaying statistics created by the statistic creation unit 36, an input / output unit 39 for performing necessary input / output, and stores various data. The storage unit 40 and weight coefficient setting unit 42 and weight coefficient determination unit 44 described later are further provided. These units are connected to each other by a signal transmission path 54 in this example. The display means 38 is a display, for example.

  The one or more types of weighting coefficients are handled as attributes of each alarm, are assigned to the alarm data AD for each alarm ID, and are stored in the database device 24 in this example together with the alarm data AD. An example of the alarm data AD to which the weight coefficient is given is shown in FIG. In FIG. 2, two weighting factors are shown as an example, but the present invention is not limited to this. The same applies to FIG.

  The weighting factor can be roughly classified into the following two.

(1) Weight coefficient uniquely determined for each alarm type In this example, as described above, since the alarm is managed by the alarm ID, this weight coefficient is a weight coefficient uniquely determined for each alarm ID. You can also. Details will be described below.

(2) Weighting coefficient uniquely determined for each alarm occurrence The term "every alarm occurrence" here is not limited to the timing of alarm generation or release. It can also be said to be a weighting coefficient determined by other than the alarm ID. That is, even with an alarm having the same alarm ID, there is a possibility that the weighting coefficient differs for each occurrence. Details will be described below.

  The weighting factor (1) can be set by the user as a unique value for each type of alarm (that is, for each alarm ID). The weight coefficient setting means 42 performs this setting. The weight coefficient setting means 42 may be, for example, a touch panel provided on the front surface of the display constituting the display means 38 or a keyboard. The weighting factor set by the weighting factor setting means 42 is stored in the storage means 40 as a table together with the corresponding alarm ID. Then, when the alarm is released, the alarm statistics device 26 searches this table using the alarm ID as a key, determines the weight coefficient of the alarm, and assigns it to the alarm data AD in the database device 24. An example of such data is shown in FIG.

  An example of the use of the weighting factor (1) is to exclude a specific alarm from the alarm statistics. For example, during maintenance or failure repair, an alarm that the storage container door of the management target device 10 has been opened may frequently occur. These alarms are, of course, important when working in the field, but are not necessary for alarm statistics for improving the utilization efficiency of the management target device 10, for example. Therefore, the weighting factor is set to 0 for these alarm IDs, and any other value other than 0 (for example, 1 or more) is set for the alarm IDs necessary for other alarm statistics depending on the purpose. By keeping it, unnecessary alarms can be easily excluded from the alarm statistics.

  The weight coefficient (2) can be set by the user every time an alarm occurs. This setting is also performed by the weight coefficient setting means 42 in this example. That is, in this example, the weighting factor setting unit 42 serves as both a unit for setting the weighting factor (1) and a unit for setting the weighting factor (2). The setting of the weighting factor (2) is performed, for example, when the alarm is canceled after the alarm is generated. More specifically, for example, it can be set (input) according to a dialog of a control screen displayed on the display unit 38 when the alarm is canceled. The alarm statistics device 26 assigns the weight coefficient set in this way to the alarm data AD in the database device 24 using the alarm ID as a key. An example of such data is shown in FIG.

  However, since it is not always possible to input a weighting coefficient when there is a person at the alarm statistical device 26 when the alarm is released, the alarm statistical device 26 is provided with a function for inputting the weighting coefficient later. You can leave it. As the method, for example, (a) according to a control screen displayed on the display means 38, an alarm ID and an alarm release date and time are designated, and a necessary weight coefficient is input from the weight coefficient setting means 42; b) The alarm statistics device 26 outputs an alarm release list in the form of a file by the input / output means 39, adds a necessary weighting factor to the file, and inputs it by causing the alarm statistics device 26 to read by the input / output means 39. There is a method. In the case of (b), it can be said that the input / output means 39 for reading the file to which the weight coefficient is added is the weight coefficient setting means.

  As an example of using the function (a) or (b), there is a case where the weighting factor of the alarm is determined and input using the work man-hours and work costs generated in association with the alarm. . For example, the operator may calculate the weighting factor by summing up the values of the work man-hours and work costs, and input them when the alarm is released. Further, these values aggregated by the operator may be reported to the administrator, and the administrator may input the weighting coefficient in units that are gathered to some extent such as weekly or monthly.

  Among the weighting factors in (2) above, there are those that can be automatically determined by the alarm statistics device 26 based on a predetermined rule set in advance. The weight coefficient determination means 44 performs this determination. The weighting factor determining unit 44 automatically determines a weighting factor based on a predetermined rule when the alarm is released, and the weighting factor assigning unit 30 adds this to the alarm data AD in the database device 24 as one of the weighting factors. . As a result, the weighting factor can be determined without human intervention, so that labor can be saved.

  As an example of using this function, there is an example in which a time corresponding to the stop time of the management target device 10 associated with the alarm is employed as a weighting factor. When the alarm is released, the weighting factor determination unit 44 searches the alarm occurrence time from the database device 24 using the alarm ID as a key, and subtracts the alarm occurrence time from the current alarm release time, so that the target alarm is generated. The time from release to release can be obtained. The weighting factor determining unit 44 determines a weighting factor corresponding to this time, and the weighting factor assigning unit 30 adds this to the alarm data AD in the database device 24 as one of the weighting factors. Thereby, the degree of influence on the utilization efficiency of the management target device 10 can be added to the alarm statistics.

  The alarm statistics device 26 preferably has a function capable of switching between valid / invalidity for each of the above-described one or more weighting factors when the statistics creation means 36 creates the alarm statistics. In this example, this function is given to the weighting coefficient adding means 30. The total weight coefficient calculation means 32 multiplies the weight coefficients validated by the weight coefficient provision means 30 with each other to obtain one total weight coefficient for each alarm. The weighted number calculation means 34 calculates the weighted number by multiplying 1 representing the number of each alarm by one of the total weighting factors. For example, if the total weight coefficient is 2.5, the number of weighted cases is 1 × 2.5 = 2.5 [cases]. The statistics creating means 36 adds up the weighted cases for each alarm, and creates statistics of the weighted alarm occurrences.

  The statistical form of the number of weighted alarm occurrences may be, for example, a so-called worst list in which alarms are arranged in descending order, or may be visualized in the form of an analysis chart such as a percentage graph or Pareto chart. good. In this case, as a method for representing each alarm, it is simple and preferable to use an alarm ID, but text corresponding to the alarm ID may be used.

  The alarm management device 20 uses the alarm statistics device 26 to attach one or more types of weighting factors to each alarm generated by the management target device 10 and multiply the one or more types of weighting factors for each alarm. The total weighting factor is obtained, and 1 representing the number of each alarm is multiplied by this total weighting factor to obtain the number of weighted items, and the number of weighted items is accumulated for each alarm, respectively. An alarm management method of creating statistics on the number of alarm occurrences can be implemented.

  According to the alarm management method and the alarm management apparatus 20 that implements the alarm management method, it is possible to create alarm statistics including value judgment by adopting a weighting factor instead of simply integrating the number of alarm occurrences as in the prior art. Can do. In other words, each alarm issued by the management target device 10 can be given a weighting factor according to the purpose, so that the degree of influence of each alarm can be evaluated according to the purpose based on the statistics of the number of occurrences of weighted alarms. it can.

  Next, a case where the alarm (B) is linked to device data will be described.

  In this case, as in the example shown in FIG. 4, the alarm statistics device 26 links the alarm generated by the management target device 10 and the device data representing the state of the management target device 10 with each other using the alarm occurrence date as a key. A link means 46, a trend graph creation means 48 for creating a trend graph of predetermined device data in the device data, a trend graph created by the trend graph creation means 48 are displayed, and based on the link On the trend graph, there is provided display means 38 for displaying the date and time when the alarm specified in advance occurs. A trend graph is a graph in which the horizontal axis is time and the vertical axis is a value, and represents a temporal trend / transition of the certain value. An example of the display means 38 is as described above.

  In this example, the alarm statistics device 26 further includes the input / output means 39 and the storage means 40, which are preferably provided. The connection through the signal transmission path 54 is as described above.

  The alarm management device 20 including such an alarm statistics device 26 links an alarm generated by the management target device 10 and device data representing the state of the management target device 10 to each other using the alarm occurrence date and time as a key, Alarm management in which a trend graph of predetermined device data out of device data acquired from the management target device 10 is created, and a date and time when a predetermined alarm is generated is displayed on the trend graph based on the link The method can be carried out.

  According to the alarm management method and the alarm management device 20 that implements the alarm management method, the alarm generated by the management target device 10 and the device data are linked to each other using the alarm occurrence date and time as a key. You can know the relationship with device data. That is, it is possible to easily know at what value the predetermined device data is generated by using the trend graph together with the alarm. As a result, it becomes easier to investigate the cause of the alarm occurrence.

  For example, it is assumed that an alarm “implanted ion mass number abnormality (for example, alarm ID = 3025)” has occurred. The link means 46 designates device data called “implanted ion mass number set value” as data to be linked to this alarm ID. Usually, the link time is also specified. The trend graph creating means 48 receives, for example, an instruction to create a trend graph, accesses the database device 24, searches the device data (injected ion mass number setting value) within a specified period, and retrieves the device data. Create a trend graph. An example is shown in FIG.

  The display means 38 displays a trend graph as shown in FIG. 5 created by the trend graph creation means 48. Further, based on the link, the date and time when the alarm specified in advance (for example, alarm ID = 3025 and 3026) is generated is displayed on the trend graph by means of an icon 64 or the like. In other words, the icon 64 representing the occurrence of the alarm is displayed at the date and time when the alarm occurred. FIG. 5 shows an example of a state in which an icon 64 with an alarm ID = 3025 is displayed. In the example of FIG. 5, since the horizontal axis is time, the date is displayed on the date display unit 60. The horizontal axis and the date display unit 60 can display the date and time when the alarm occurred. However, the horizontal axis may be the date. The alarm ID may be displayed together with the icon 64, but instead, it is preferable to provide a legend display unit 62 for displaying the correspondence between the icon 64 and the alarm ID as in this example. By doing so, the display around the icon 64 becomes simple and easy to see.

  In the example of FIG. 5, since it is possible to easily know the tendency that an alarm with alarm ID = 3025 is frequently generated when the set value of the implanted ion mass number is 40, this can be used to help investigate the cause of the alarm. It can be.

  As another link function, an alarm data filtering function may be employed. In this case, the alarm statistics device 26 extracts only the alarms generated when the predetermined device data of the device data has a value or a range of values designated in advance, as in the example shown in FIG. A filtering means 50 to be subjected to statistics, and a statistics creating means 52 for creating statistics of alarms extracted by the filtering means 50. The filtering unit 50 can be set with designation of the predetermined device data and the designated value or range of values.

  In this example, the alarm statistics device 26 is also preferably provided with the display means 38, the input / output means 39, and the storage means 40 described above. The connection through the signal transmission path 54 is as described above.

  The alarm management device 20 including such an alarm statistics device 26 is a device that is acquired from the management target device 10 and represents the state of the device 10 when creating statistics of alarms issued by the management target device 10. It is possible to implement an alarm management method in which only an alarm that is generated when predetermined device data in the data has a predetermined value or a range of values is subject to statistics.

  According to the alarm management method and the alarm management device 20 that implements the alarm management method, it is possible to know the alarm occurrence status when the management target device 10 is in a specific operation state by the statistics created as described above. Investigate the cause of the alarm. The statistics created as described above are also useful when it is desired to preferentially take countermeasures only for alarms in a specific situation.

  For example, a specific value (for example, 11) of the apparatus data “implanted ion mass number setting value” is designated as a filter in the filtering means 50. As a result, alarm statistics using only the alarm generated when the mass number of implanted ions is 11 can be created. Thereby, it is possible to know the alarm occurrence state in a specific situation where the number of implanted ions is 11, which is useful when it is desired to take preferential measures only for the alarm under the specific situation.

  FIG. 1 shows an example in which an alarm generated by one managed device 10 is managed by one alarm management device 20, but an alarm generated by a plurality of managed devices 10 is managed by one alarm. It can also be managed by the device 20. An example in that case is shown in FIG.

  In the example of FIG. 7, five management target devices 10 a to 10 e are managed. The configuration of each of the management target devices 10a to 10e is the same as that of the management target device 10 illustrated in FIG. 1, for example. In this example, the alarm management device 20 includes five data collection devices 22a to 22e that collect data from the respective management target devices 10a to 10e. The functions of the data collection devices 22a to 22e are the same as those of the data collection device 22 shown in FIG. One database device 24 and one alarm statistics device 26 can be shared.

  In this case, if it can be identified from which of the five management target devices 10a to 10e each of the alarms is generated, the case where there is one management target device will be described later. Similarly, alarms can be managed. For this purpose, as in the example shown in FIG. 8, a device ID that is identification information for identifying a management target device may be added to the alarm data AD described above. The same alarm ID may be used for the five managed devices 10a to 10e. In this way, since the alarm can be specified by the device ID and the alarm ID, the alarm statistical device 26 shown in FIGS. 3, 4 and 6 can be used almost as it is, and each of the above-described each An alarm management method can be implemented.

  In addition, although there is currently only one device to be managed as in the example shown in FIG. 1, for example, even if there is a possibility of increasing to a plurality of devices in the future, the data as shown in FIG. 8 can be handled. However, a single value (for example, 1) may be entered in the device ID. By doing so, it is possible to easily cope with an increase in the number of devices to be managed in the future.

  In addition, you may implement | achieve the alarm management apparatus 20 of each said Example using a computer.

It is a block diagram which shows an example of the alarm management apparatus which enforces the alarm management method concerning this invention, and its management object apparatus. It is a conceptual diagram which shows an example of the data which provided the weighting coefficient to alarm data. It is a block diagram which shows an example of a structure of an alarm statistics apparatus. It is a block diagram which shows the other example of a structure of an alarm statistics apparatus. It is a figure which shows an example of a trend graph. It is a block diagram which shows the further another example of a structure of an alarm statistics apparatus. It is a block diagram which shows the other example of the alarm management apparatus which enforces the alarm management method concerning this invention, and its management object apparatus. It is a conceptual diagram which shows the other example of the data which provided the weighting coefficient to alarm data.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Management object apparatus 20 Alarm management apparatus 22 Data collection apparatus 24 Database apparatus 26 Alarm statistics apparatus 30 Weight coefficient provision means 32 Total weight coefficient calculation means 34 Weighted number calculation means 36 Statistics creation means 38 Display means 42 Weight coefficient setting means 44 Weight Coefficient determination means 46 Link means 48 Trend graph creation means 50 Filtering means 52 Statistics creation means

Claims (6)

In an alarm management method for managing alarms generated by managed devices,
(A) a step in which the data collection device collects alarm data issued by the device to be managed and device data representing the state of the device to be managed, and stores the collected data in a database device;
(B) Using the data stored in the database device, the alarm statistics device links the alarm issued by the managed device and the device data indicating the status of the managed device to each other using the alarm occurrence date as a key. A step of creating and displaying a trend graph of predetermined device data in the device data, and a step of displaying a date and time when a predetermined alarm is generated on the trend graph based on the link; An alarm management method characterized by comprising: In an alarm management method for managing alarms generated by managed devices,
(A) the data collection device collects alarm data issued by the managed device and device data representing the state of the managed device, and stores the collected data in a database device; and
(B) The alarm statistics device uses the data stored in the database device to create statistics of alarms issued by the device to be managed, and when the statistics are created, An alarm management method, comprising: a step of subjecting only an alarm generated when device data has a value or a range of values specified in advance to statistics.   The alarm management method according to claim 1, wherein each alarm is managed by an alarm ID for identifying the alarm. In an alarm management device that manages alarms that are issued by managed devices,
A data collection device that collects alarm data issued by the device to be managed and device data representing the state of the device to be managed, a database device that stores data collected by the data collection device, and data stored in the database device And an alarm statistics device that creates alarm statistics using
The alarm statistics device includes a link means for linking an alarm generated by the management target device and device data indicating the status of the management target device with each other using an alarm occurrence date as a key, and a predetermined device among the device data A trend graph creating means for creating a data trend graph and a trend graph created by the trend graph creating means are displayed, and a date and time when a predetermined alarm is generated is displayed on the trend graph based on the link. An alarm management device comprising a display means. In an alarm management device that manages alarms that are issued by managed devices,
A data collection device that collects alarm data issued by the device to be managed and device data representing the state of the device to be managed, a database device that stores data collected by the data collection device, and data stored in the database device And an alarm statistics device that creates alarm statistics using
The alarm statistics device includes a filtering unit that extracts only alarms that are generated when predetermined device data of the device data has a value or range of values specified in advance, and is a target of statistics. An alarm management device comprising: a statistic creation unit that creates a statistic of an alarm extracted by the filtering unit.   The alarm management device according to claim 4 or 5, wherein each alarm is managed by an alarm ID for identifying the alarm.

Images