JP3408752B2 - Failure prediction maintenance device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a maintenance device for electronic equipment, and more particularly to a maintenance device having a function of predicting a failure of each equipment unit.

[0002]

2. Description of the Related Art FIG. 15 is a diagram showing a configuration of a conventional maintenance device for electronic equipment. FIG. 15 (a) shows components that function during operation of the electronic equipment, and FIG. 15 (b) shows functions during maintenance of the electronic equipment. It shows the constituent elements. In the figure, 2 is a plurality of device units that constitute an electronic device, 13 is an operating time recording counter that is provided in each of the device units 2 and counts the total operating time of each device unit 2, and 14 is each device unit. The maintenance computer 10 including the device exchange setting period storage memory 15 for storing the respective exchange periods 2 and the device exchange determination software 16 for determining the exchange of each device unit is output from the maintenance computer 14. It is a display device that displays device replacement determination information.

Next, the operation will be described. During operation of the electronic device, the operating time recording counter 13 provided in each device unit 2 counts the total operating time of each device unit 2. A predetermined operation period is set for each device unit 2, and this operation period is stored in the device replacement setting period storage memory 15 in the maintenance computer 14. When servicing an electronic device, the maintenance computer 14 uses the device replacement determination software 16 to compare the total operating time data from the operating time recording counter 13 with the operating period in the device replacement setting period storage memory 15. Then, it is determined whether each device unit 2 is continuously operated or replaced, and the device replacement determination information as the determination result is output to the display device 10. When the device replacement is displayed as the determination information, the device unit is replaced with maintenance equipment (not shown). FIG. 16 is a flowchart showing the processing contents of the device replacement determination software 16, in which the maintenance computer 14 reads out the total operating time data of each device unit 2 from the operating time recording counter 16 and stores the device replacement setting period. Each exchange period data of each device unit 2 in the memory 15 is read (step S2
1) For each device unit 2, it is determined whether or not the total operating time data is larger than the replacement period data (step S22), and when the total operating time data of the device unit 2 is larger than the replacement period data. Outputs the device replacement determination information to the display device 10 as "device replacement" (step S23), and instructs the replacement of the device unit 2. When the total operation time data of the device unit 2 does not exceed the exchange period data, the device exchange determination information is output to the display unit 10 as "continuous operation" (step S24), and the device unit 2 continues. Display that it is operational.

[0004]

As described above, in the conventional maintenance device, the total operating time is set for each device unit 2 constituting the electronic device, and the total operating time of each device unit 2 is preset. After the replacement period, the device unit 2 is unconditionally replaced. When the replacement period is set to the average value of the time when each device unit 2 is considered to be defective, the failure during operation increases, so the replacement period is usually shorter than the average value. Was set to.
Therefore, there is a problem in that even a device unit 2 that can be sufficiently continuously operated is often replaced, resulting in a lot of waste.

The present invention has been made in order to solve the above-mentioned conventional problems, and accurately predicts a failure of a device unit of an electronic device and enables the device unit to be used immediately before the failure. An object of the present invention is to provide a failure prediction maintenance device that can eliminate the waste of replacement.

[0006]

A failure prediction maintenance apparatus according to claim 1 of the present invention is a maintenance equipment used at the time of maintenance, an equipment unit, and the equipment unit mounted on the equipment unit. Equipped with an exchange date / time memory that stores exchange date / time data when the equipment is exchanged, and an environment sensor that detects the operating environment conditions of the device unit such as temperature and humidity and outputs the environmental data of each device unit, and also the rule of thumb. Which is the average failure interval time of each device unit according to environmental conditions such as temperature and humidity preset by M
An MTBF data storage memory for storing TBF data,
An operation time totaling result storage memory that stores operating time data of each device unit, a past failure data storage memory that stores past failure data of each device unit, a previous replacement date / time data and environmental data of each device unit MTBF
It has means for obtaining failure prediction data of each equipment unit from data, operating time data and past failure data, and is provided with a computer for predicting failure of each equipment unit, and an indicator for displaying the failure prediction data. It is a thing.

Further, the failure prediction maintenance device according to claim 2 takes the above operating time data into consideration of the environmental conditions based on the previous exchange date / time data, environmental data, MTBF data and operating time data of each equipment unit. The operation time data is converted into operation time parameters, and failure prediction data of each device unit is obtained from the operation time parameters and past failure data.

In the failure predictive maintenance device according to the third aspect, the MTBF data is data corresponding to all combinations of environmental conditions such as temperature, humidity, vibration, shock and atmospheric pressure.

According to a fourth aspect of the failure predictive maintenance device, the MTBF data is data corresponding to respective environmental conditions such as temperature, humidity, vibration, shock and atmospheric pressure.

Further, the failure prediction and maintenance device according to claim 5 is characterized in that the temperature,
Weighting factors are given for all combinations of environmental conditions such as humidity, vibration, shock, and atmospheric pressure.

According to another aspect of the present invention, there is provided a failure prediction and maintenance device in which a replacement date / time storage memory is mounted on a computer, and each device unit is provided with a device recognition means for recognizing a replacement device.

Further, in the failure prediction maintenance device according to claim 7, replacement date data, environmental data, MTBF data, past failure data, for each module of each equipment unit,
It is provided with operating time data.

[0013]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. In the following description, portions common to the conventional example will be described using the same reference numerals.

Embodiment 1. FIG. 1 is a diagram showing a configuration of a failure prediction maintenance device for an electronic device according to a first embodiment of the present invention. FIG. 1A is a diagram showing components that function during operation of the electronic device, and functions during maintenance of the electronic device. The components are shown in FIG. In the figure, 1 is a maintenance equipment used when servicing an electronic device, 2 is a plurality of device units constituting the electronic device, 3 is each mounted on each of the above device units 2,
An exchange date / time storage memory for storing exchange date / time data when the equipment unit 2 is exchanged with the above-mentioned maintenance equipment 1, and 4 senses the usage status (environmental condition) of the equipment unit such as temperature, humidity, vibration, shock, and atmospheric pressure. An environment sensor 5 for outputting environmental data of the device unit 2 is an MT for storing MTBF data which is an average failure interval time of each device unit according to each environmental condition such as temperature and humidity preset by an empirical rule.
A BF data storage memory 6, a past failure data storage memory 7 for storing past failure data of each device unit,
From the operating time totaling result storage memory 8 that stores the operating time data of each device unit, the previous exchange date / time data of each device unit 2, the environmental data, the MTBF data, the past failure data, and the total operating time data, each device unit. And a failure prediction software 9 which is a means for obtaining the failure prediction data of 1. and is a computer for predicting the failure of each device unit 2. Further, 10 is a display device for displaying failure prediction data output from the computer 5.

Next, the operation of the failure prediction maintenance device having the above configuration will be described with reference to the flowchart showing the processing contents of the failure prediction software 9 shown in FIG. Note that, when the device unit 2 is replaced with the maintenance device 1 during the maintenance of the electronic device, the replacement date and time at that time is written in the replacement date and time storage memory 3 mounted on the device unit 2. . During operation of the electronic device, the failure prediction software 9 of the computer 5 first reads the replacement date / time data of the device unit 2 from the replacement date / time storage memory 3 of the device unit 2 and uses each device unit 2 from the environment sensor 4. The situation (temperature, humidity, vibration, shock, atmospheric pressure, etc.) is read as environmental data (step S1). Next, the operating time of each operating environment of the device unit 2 is totaled (step S2), and the totaled result is added to the operating time totaling result of the operating time totaling result storage memory 8 (step S2).
3). FIG. 3 is a table showing an example of a storage format of the operation time totaling result, and this table is for each device unit 2
It is stored in the operation time totaling result storage memory 8 for each. Further, FIG. 4 is a specific example thereof, and for simplification, FIG.
In both FIGS. 4A and 4B, two environmental condition parameters representing the usage environment of the device unit 2 are temperature (i) and humidity (j). In FIG. 3, in the operation time A21, the temperature (i) range is i = 2 (10 ° C. ± 2.5 ° C.), and the humidity (j) range is j.
It represents the time when the device unit 2 is used under the use condition of = 1 (50% ± 5%). In FIG. 4, the value of the operating time data A21 is A21 = 70000h. If the environmental condition parameters are, for example, five values of temperature (i), humidity (j), vibration (k), shock (l), and atmospheric pressure (m), the table shows each environmental condition. Operation time data Aij corresponding to all combinations of
This is a table that records klm.

Next, the failure prediction software 9 is the MTB.
The MTBF data of the device unit 2 is read from the F data storage memory 6 (step S4). This MTBF data is the MTBF (Mean Time Between F) of each device unit according to the use environment condition preset by the rule of thumb.
ailures; average failure interval time-average time to operate without failure), and MTBF data storage memory 6 stores all environmental conditions such as temperature, humidity, vibration, shock, and atmospheric pressure. MTB corresponding to the combination
F data is stored. FIG. 5 is a table showing an example of the saving format of the MTBF data, and FIG.
It is a table which shows the specific example. For simplicity,
The environmental condition parameters were only temperature (i) and humidity (j). In FIG. 5, the MTBF 21 has a temperature (i) range i = 2 (10 ° C. ± 2.5 ° C.) and a humidity (j) range j =.
1 represents MTBF data under a use condition of 1 (50% ± 5%), and in FIG.
The value of F21 is MTBF21 = 105000h. In the case where the above-mentioned environmental condition parameters are the temperature (i), the humidity (j), the vibration (k), the shock (l), and the atmospheric pressure (m), the MTBF data is the operating time data Aijklm described above. As in the case, it is represented as MTBFijklm.

The failure prediction software 9 calculates an operating parameter Ts considering environmental conditions from the operating time data Aijklm and the MTBF data MTBFijklm according to the following equation (1) (step S5), and the operating parameter Ts is calculated. Ts is compared with the past failure data T stored in the past failure data storage memory 7, which is the average value of operating time parameters from the past maintenance to the failure. Ts = (A11111 / MTBF11111 + A11112 / MTBF11112 + ... + Aijklm / MTBFijklm) (1) For example, when the environmental condition parameters are only temperature and humidity, the operating time data Aij in FIG. 4 and the MTBF data MTBFij in FIG. Therefore, the operating time parameter Ts in consideration of the environmental conditions is Ts = (A11 / MTBF11 + A12 / MTBF12 + A21 / MTBF21 + A22 / MTBF22 + A31 / MTBF31) = (0/110000 + 0/115000 + 70000/105000 + 5000/100000 + 10000/982) = 0.82. The value of the operating time parameter Ts increases as the operating time of each device unit 2 increases.

Before comparing the operating time parameter Ts with past failure data T, the device unit 2 concerned
Is checked whether or not there is a failure (step S6).
If the device unit 2 is out of order, the operating time parameter Ts is written in the past failure data storage memory 7 (step S7), and the past failure data T of the device unit 2 is recalculated (step S8). ). In step S6, if the device unit 2 has not failed, the process proceeds to step S9, the average value of past operating time parameters Ts is calculated, and this is set as past failure data T. However, if the device unit 2 has not failed in the past, T = 1. Next, it is determined whether or not the value of the current operating time parameter Ts is larger than the past failure data T (step S10). When the value of the current operating time parameter Ts exceeds the past failure data T, the failure prediction software 9 determines that the device unit 2 may fail, and displays the failure prediction data on the display 10. Output (step S11) and instruct replacement of the device unit 2. When the value of the current operating time parameter Ts is less than or equal to the past failure data T, it is determined that the device unit 2 can be continuously operated, and the failure time of the next device unit 2 is predicted.

As described above, according to the first embodiment, not only the operating time of each device unit 2 is recorded, but also the last exchange date / time data, environment data, MTBF data, and operating time of each device unit are recorded. The operating time parameter Ts is calculated from the data and the environmental conditions during operation of the device unit 2 such as temperature, humidity, vibration, shock, atmospheric pressure, etc. are calculated, and this operating time parameter Ts and past failure data T are compared. Since the failure prediction was made, the equipment unit 2
It is possible to accurately predict the failure time. Therefore, since the device unit 2 can be used until just before a failure, it is possible to eliminate waste of replacement of the device unit 2, reduce the life cycle cost of the device unit 2, and achieve resource saving. You can

Embodiment 2. In the first embodiment, M
The case where the TBF data is data corresponding to all combinations of environmental conditions such as temperature, humidity, vibration, shock, and atmospheric pressure has been described, but the MTBF data is described as temperature, humidity, vibration, shock, atmospheric pressure, and the like. Environmental conditions (i = 1, 2,
The current operating time parameter Tis (i = 1, 1) of each device unit 2 as data corresponding to each of
2, 3, ...) Values are calculated, and past failure data Ti (i = 1, 2 ,,) set for each environmental condition is calculated.
3, ...) may be compared to make a failure prediction. FIG. 7 shows MT corresponding to each of the above environmental conditions.
In the figure, MTBFij is an environmental condition parameter i (for example, i = 1; temperature, i = 2; humidity, ...), and an environmental level (parameter value). The range is j), and specifically MTBF21 has an environmental condition parameter of humidity (i = 2) and a level of 1 (50%).
.About.55%), which is the MTBF data under the use condition. FIG. 8 is a specific example in which the environmental condition parameters are only temperature and humidity, and the value of MTBF21 is MTBF.
21 = 110,000 h. FIG. 9 is a table showing an example of a storage format of the operation time totaling result corresponding to the MTBF data. Aij has environmental condition parameters i (1; temperature, 2; humidity, ...) And environmental level. Represents operating time data with j. FIG. 10 is a specific example in which the environmental condition parameters are only temperature and humidity, for example, humidity (i = 2) and the level is 1 (50% to 55%).
%) Of the operating time data A21 is set to A21 = 11000h.

When the number of environmental condition parameters is n, the failure prediction software 9 uses the following equation (2-) from the operating time data Aij and the MTBF data MTBFij.
1) to (2-n), a plurality of operating parameters Tis (i = 1 to n) in which environmental conditions are added are calculated, and the past Tis and the past of the device unit 2 stored in the failure data storage memory 7 are calculated. The failure data Ti (i = 1 to n) is compared. If there is no past failure data, T
Let i = 1. T1s = (A11 / MTBF11 + A12 / MTBF12 + ... + A1j / MTBF1j) ... (2-1) T2s = (A21 / MTBF21 + A22 / MTBF12 + ... + A2j / MTBF2j) ... (2-2) ... Tis = (Fi1 + Ai1 + MT2). /MTBFi2+...+Aij/MTBFij) (2-i) Tns = (An1 / MTBFn1 + An2 / MTBFn2 + ... + Anj / MTBFnj) (2-n) For example, the environmental condition parameter is temperature (i = 1). And humidity (i = 2) only, operating time parameters T1s and T2s considering environmental conditions are calculated from the MTBF data MTBFij in FIG. 8 and the operating time data Aij in FIG. 10 as follows. be able to. T1s = (A11 / MTBF11 + A12 / MTBF12 + ... + A1j / MTBF1j) = (15000/105000 + 10000/100000) = 0.24 T2s = (A21 / MTBF21 + A22 / MTBF12 + ... + A2j / MTBF2j) = (11000/108000) 18000/108000 + 1 The values of T1s and T2s both increase as the operating time of each device unit 2 increases. Therefore, when there are i environmental condition parameters, if one operating time parameter Tms value among the operating time parameters Tis (= 1 to n) is larger than the value of the past failure data Tm, the failure occurs. The prediction software 9 judges that the device unit 2 is at risk of failure, and the display unit 1
The failure prediction data is output to 0. Further, when all the values of the i current operating time parameters Tks do not exceed the respective past failure data Tk, it is determined that the device unit 2 can be continuously operated, and the next device unit 2 is determined.
Predict the failure of.

Embodiment 3. In the first embodiment, the case has been described where the MTBF data is data corresponding to all combinations of environmental conditions such as temperature, humidity, vibration, shock, and atmospheric pressure. However, the MTBF of each device unit is described.
A time weighting coefficient corresponding to all combinations of environmental conditions such as temperature, humidity, vibration, shock, and atmospheric pressure is given to the data, and the value of the current operating time parameter Ts of each device unit 2 is calculated. As a result, the same effect can be obtained. That is, the MTBF for each device unit 2
Data is set, and further, a temporal weighting coefficient Bijklm corresponding to all combinations of environmental conditions of temperature, humidity, vibration, shock, and atmospheric pressure is set for the MTBF data. FIG. 11 is a table showing an example of a storage format of the MTBF data and the temporal weighting coefficient Bijklm, and
FIG. 12 is a table showing a specific example thereof. However,
For simplicity, the environmental condition parameters are only temperature (i) and humidity (j). In FIG. 11, B21 indicates that the temperature (i) is i = 2 (10 ° C. ± 2.5 ° C.) and the humidity (j) is j.
It represents a temporal weighting coefficient under the use condition of = 1 (50% ± 5%). In FIG. 12, the value of the temporal weighting coefficient B21 is B21 = 1.05. When the above-mentioned environmental condition parameters are temperature (i), humidity (j), vibration (k), shock (l), and atmospheric pressure (m), the temporal weighting coefficient is expressed as Bijklm. The operating time data is represented as Aijklm, as in the first embodiment.

The failure prediction software 9 uses the operation time data Aijklm, the MTBF data, and the temporal weighting coefficient of the MTBF data from Bijklm to calculate the following equation (3).
Thus, the operating parameter Ts in consideration of the environmental condition is calculated, and the Ts is compared with the past failure data T of the device unit 2 in the past failure data storage memory 7. If there is no past failure data, T = 1. Ts = (A11111 · B11111 + A11112 · B11112 + ... + Aijklm · Bijklm) / M TBF (3) For example, when the environmental condition parameters are only temperature and humidity, the MTBF data of FIG. 12 and the temporal weighting coefficient are Bijklm and From the above-mentioned operating time data Aij in FIG. 4, the operating time parameter Ts in consideration of the environmental condition is: Ts = (A11 ・ B11 + A12 ・ B12 + A21 ・ B21 + A22 ・ B22 + A31 ・ B31) / MTBF = (0.1.10 + 0.11.15 + 70000)・ 1.05 + 5000 ・ 1.00 + 10000 ・ 0.98) /100000=0.88. The value of the operating time parameter Ts increases as the operating time of each device unit 2 increases, and when the operating time parameter Ts becomes larger than the value of the past failure data T, the failure prediction software 9 Judges that the device unit 2 is at risk of failure, and outputs failure prediction data to the display 10.

Fourth Embodiment In the first embodiment, the exchange date / time storage memory 3 for storing the exchange date / time of each equipment unit is provided for each equipment unit. However, as shown in FIG. 13, the exchange date / time storage memory 3 is installed in the computer 5, Each device unit 2 has a device recognition chip 1 which is a device recognition means.
The same effect can be obtained even if the inside of 1 is provided and the replacement date and time of each device unit 2 is managed. In the fourth embodiment, the exchange date / time storage memory 3
The data transmission from the device unit 2 to the computer 5 is simplified by installing the computer 5 in the computer 5, and the processing speed of failure prediction in the computer 5 can be increased.

Embodiment 5. In the first embodiment, exchange date / time data, environment data, M
Although the configuration is such that the TBF data, the past failure data, and the total operation time data are provided, as shown in FIG. 14, for each module 12 in each device unit 2, replacement date / time data, environmental data, MTBF data. Since the failure prediction can be performed for each module 12 by including the past failure data and the total operation time data, the accuracy of the failure prediction of the device unit 2 can be further improved.

[0026]

As described above, according to the first aspect of the present invention.
According to the described invention, the failure prediction of each device unit is made based on the previous exchange date / time data, environment data, MTBF data, operating time data, and past failure data of each device unit. The failure time can be accurately predicted, and the equipment unit can be used until just before the failure. Therefore, it is possible to eliminate the waste of replacing the device unit, reduce the life cycle cost of the device unit, and achieve resource saving.

According to the second aspect of the invention, the operating time data in which the environmental conditions are added to the operating time data based on the previous exchange date / time data, environmental data, MTBF data, and operating time data of each equipment unit. Is converted into the operating time parameter and the failure prediction data of each device unit is obtained from the operating time parameter and the past failure data. Therefore, the failure prediction in consideration of environmental conditions can be performed more accurately.

According to a third aspect of the present invention, the MTB
Since the F data is data corresponding to all combinations of environmental conditions such as temperature, humidity, vibration, shock, and atmospheric pressure,
It is possible to accurately predict the failure time of the equipment unit.

According to the invention described in claim 4, the MTB
Since the F data is data corresponding to each environmental condition such as temperature, humidity, vibration, shock, and atmospheric pressure, it is possible to improve the accuracy of predicting the failure time of the equipment unit.

According to the fifth aspect of the present invention, weighting factors corresponding to all combinations of environmental conditions such as temperature, humidity, vibration, shock, and atmospheric pressure are given to the MTBF data of each equipment unit. Therefore, the failure time of the equipment unit can be accurately predicted.

According to the sixth aspect of the present invention, the replacement date storage memory is installed in the computer, and the device recognition means for recognizing the replacement device is provided in each device unit. It is possible to accurately predict, and it is possible to increase the processing speed of failure prediction.

According to the invention described in claim 7, exchange date data, environmental data, and
Since the MTBF data, the past failure data, and the operating time data are provided, failure prediction can be performed for each module, and the accuracy of failure prediction of the equipment unit can be further improved.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a failure prediction maintenance device according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing processing contents of failure prediction software according to the first embodiment of the present invention.

FIG. 3 is a diagram showing a format of an operation time totaling result according to the first embodiment.

FIG. 4 is an example of an operation time totaling result according to the first embodiment.

FIG. 5 is a diagram showing a format of MTBF data according to the first embodiment.

FIG. 6 is an example of MTBF data according to the first embodiment.

FIG. 7 is a diagram showing a format of MTBF data according to the second embodiment.

FIG. 8 is an example of MTBF data according to the second embodiment.

FIG. 9 is a diagram showing a format of an operation time totaling result according to the second embodiment.

FIG. 10 is an example of an operation time totaling result according to the second embodiment.

FIG. 11 is a diagram showing a format of MTBF data according to the third embodiment.

FIG. 12 is an example of MTBF data according to the third embodiment.

FIG. 13 is a diagram showing a configuration of a failure prediction maintenance device according to a fourth embodiment of the present invention.

FIG. 14 is a diagram showing a configuration of a failure prediction maintenance device according to a fifth embodiment of the present invention.

FIG. 15 is a diagram showing a configuration of a conventional maintenance device.

FIG. 16 is a flowchart showing the processing contents of conventional device replacement determination software.

[Explanation of symbols]

1 maintenance equipment, 2 equipment unit, 3 replacement date / time storage memory, 4 environment sensor, 5 calculator, 6 MTBF data storage memory, 7 past failure data storage memory, 8
Operation time totaling result storage memory, 9 Failure prediction software, 10 Display, 11 Device recognition chip, 12
module.

─────────────────────────────────────────────────── ─── Continuation of the front page (56) Reference JP-A-6-331507 (JP, A) JP-A-62-276470 (JP, A) JP-A-57-109061 (JP, A) JP-A-7- 49712 (JP, A) JP 5-100890 (JP, A) JP 64-88892 (JP, A) JP 7-159289 (JP, A) JP 10-21211 (JP, A) JP-A-5-10857 (JP, A) (58) Fields investigated (Int.Cl. ⁷ , DB name) G01M 19/00 G07C 3/00

Claims (7)

(57) [Claims] 1. A maintenance equipment used for maintenance, an equipment unit, a replacement date / time storage memory mounted on the equipment unit, for storing exchange date / time data when the equipment unit is replaced with the maintenance equipment, and a temperature. Equipped with an environmental sensor that outputs environmental data by sensing the usage environment status of equipment units such as humidity and humidity, and the mean failure interval of each equipment unit according to environmental conditions such as temperature and humidity preset by empirical rules M to store MTBF data which is time
TBF data storage memory, operating time summary result storage memory that stores operating time data of each device unit, past failure data storage memory that stores past failure data of each device unit, and previous replacement of each device unit A means for obtaining failure prediction data of each device unit from date / time data, environmental data, MTBF data, operating time data, and past failure data, and a computer for predicting failure of each device unit; and the failure prediction data described above. A failure prediction / maintenance device comprising a display for displaying. 2. The operating time data is converted into operating time parameters, which are operating time data in consideration of environmental conditions, based on previous exchange date / time data, environmental data, MTBF data, and operating time data of each equipment unit. The failure prediction maintenance device according to claim 1, wherein failure prediction data of each device unit is obtained from the operating time parameter and past failure data. 3. The failure prediction maintenance device according to claim 1, wherein the MTBF data is data corresponding to all combinations of environmental conditions such as temperature, humidity, vibration, shock, and atmospheric pressure. 4. The failure prediction maintenance device according to claim 1, wherein the MTBF data is data corresponding to respective environmental conditions such as temperature, humidity, vibration, shock, and atmospheric pressure. 5. The weight coefficient corresponding to all combinations of environmental conditions such as temperature, humidity, vibration, shock, and atmospheric pressure is given to the MTBF data of each equipment unit. Failure prediction maintenance device. 6. The failure prediction and maintenance device according to claim 1, wherein the replacement date and time storage memory is mounted on the computer, and each device unit is provided with a device recognition means for recognizing the replacement device. 7. The failure prediction maintenance apparatus according to claim 1, wherein each module of each equipment unit is provided with replacement date / time data, environment data, MTBF data, past failure data, and operating time data. .