Method of determining a calibration or maintenance time interval

1. Method of determining a maintenance time interval after which a specific measurement device of a given type for measuring a quantity to be measured on a measurement site of an industrial site shall be maintained again, comprising the steps of:

determining a criticality (C) of the specific device,

based on the criticality (C) setting a reliability target (RT) for the device, wherein the reliability target (RT) denominates the probability of the device to be compliant according to a predefined criterion at the end of the maintenance time interval to be determined by this method,

defining deviation ranges for deviations of a property of devices of the given type from an ideal status,

selecting a reliability model for a reliability of the device as a function of a normalized time (t n ) and a set of at least one parameter (c 1 , . . . , c m ) from a variety of predefined reliability models,

determining a reliability function for each deviation range by determining a separate set of parameters for the selected reliability model for each of the deviation ranges based on prescribed reliability expectation values (RV(t n pd )) for each of the deviation ranges, which a reliability function associated with this deviation range shall comply to at at least one predefined normalized time (t n pd ),

determining the deviation of the specific measurement device and based on the deviation determining the corresponding deviation range for the specific measurement device,

determining a normalized maintenance time (t n ) as the normalized time, at which the reliability function (R(t n )) determined for the deviation range corresponding to the deviation of the specific device and given by the selected reliability model and the set of parameters determined for this deviation range equals the reliability target (RT), and

determining the next maintenance time interval, after which the specific measurement device shall be maintained again, based on a product of this normalized maintenance time (t n ) and a given reference time interval (T R ).

2. Method according to claim 1 , wherein

an ideal status of a property of measurement devices of the given type is given as a reference, and

devices of this type are considered compliant as long as their deviations from this ideal status do not exceed a maximum permissible deviation (MPD).

3. Method according to claim 1 , wherein the maintenance time interval is set to be equal to a maximum permissible maintenance time interval (T max ), in case the product of the normalized maintenance time (t n ) and the reference time interval (T R ) is longer than the maximum permissible maintenance time interval (T max ).

4. Method according to claim 1 , wherein the reference time interval (T R ) is determined

based on a length of a previous calibration time interval,

as a given percentage of an expected life time of the type of device,

according to conditions and demands the specific measurement device will be exposed to when put back into operation,

as a product of a given percentage of the expected life time and a factor accounting for the conditions and demands the specific device will be exposed to, or

based on the reliability expectation values (RV(t n pd )) and a known expected reliability value for this type of device.

5. Method according to claim 1 , wherein criticality (C) of a device is determined based on a probability (P) of a defect of the device and a severity (S) of consequences of the defect.

6. Method according to claim 1 , wherein criticality (C) of a device is determined based on a simplified severity (S) of consequences of the defect, which is based on an impact of the measurement result of the specific measurement device on the application it is used in.

7. Method according to claim 6 , wherein

the deviation range is either determined based on an effective deviation determined based on the deviation determined during maintenance and an additional term reflecting the operating conditions, or

the deviation range

is set equal to the deviation range corresponding to the determined deviation, in case the device will be exposed to normal operating conditions and/or the deviation range corresponding to the determined deviation is the highest or lowest deviation range, and

is set equal to the deviation range adjacent to the deviation range corresponding to the determined deviation, comprising higher deviations than the deviation range corresponding to the determined deviation, in case the device will be exposed to harsh operating conditions and the deviation range corresponding to the determined deviation is lower than the highest deviation range, and

is set equal to the deviation range adjacent to the deviation range corresponding to the determined deviation, comprising lower deviations than the deviation range corresponding to the determined deviation, in case the device will be exposed to ideal operating conditions and the deviation range corresponding to the determined deviation is higher than the lowest deviation range.

8. Method according to claim 1 , wherein the predefined reliability models comprise an exponential model, a Weibull model and/or a mortality drift model.

9. Method according to claim 1 , wherein one predefined normalized time (t n pd ) is the largest normalized time (t n pd =1).

10. Method according to claim 1 , wherein the sets of model parameters (c 1 , . . . , c m ) are determined such, that the resulting reliability functions (R(t n ; c 1 , . . . , c m )) cover a normalized area ([0, 1] 2 ) defined by reliabilities (R(t n )) ranging from 0 to 1 and normalized calibration times (t n ) ranging from 0 to 1.