Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
METHOD OF ASSESSMENT OF A RISK INVOLVED IN OPERATING A MEASUREMENT DEVICE ON A MEASUREMENT SITE IN AN INDUSTRIAL PROCESS
Document Type and Number:
WIPO Patent Application WO/2013/060522
Kind Code:
A1
Abstract:
A method of assessment of a risk (R) involved in operating a measurement device on a measurement site in an industrial process is described, which takes into account a maintainability (M) of the device, comprising the steps of: determining a degree of difficulty of a maintainability (M) of the device, determining a probability (P) of an undetected defect of the device, determining a severity (S) of the consequences of the defect of the device, weighing the determined probability (P), the determined severity (S) and the determined degree of difficulty of the maintainability (M) with corresponding predefined weighing factors (gS, gP, gM) corresponding to a contribution of the difficulty of maintainability (M), the probability (P) and the severity (S) to the risk (R), and determining the risk (R) based on a length (LR) of a vector (I) whose coordinates are given by the weighted degree of difficulty of the maintainability (gM M), the weighted probability (gP P) and the weighted severity (gS S).

Inventors:
VAISSIERE DIMITRI (FR)
Application Number:
PCT/EP2012/067830
Publication Date:
May 02, 2013
Filing Date:
September 12, 2012
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
ENDRESS & HAUSER MESTECHNIK GMBH & CO KG (DE)
VAISSIERE DIMITRI (FR)
International Classes:
G05B23/02
Other References:
IRAJ ELYASI KOMARI ET AL: "Extended Dependability Analysis of Information and Control Systems by FME(C)A-technique: Models, Procedures, Application", DEPENDABILITY OF COMPUTER SYSTEMS, 2009. DEPCOS-RELCOMEX '09. FOURTH INTERNATIONAL CONFERENCE ON, IEEE, PISCATAWAY, NJ, USA, 30 June 2009 (2009-06-30), pages 25 - 32, XP031531357, ISBN: 978-0-7695-3674-3
THOMAS LANZISERO: "Applied Safety Science and Engineering Techniques (ASSETâ ): Taking HBSE to the next level", PRODUCT COMPLIANCE ENGINEERING (ISPCE), 2010 IEEE SYMPOSIUM ON, IEEE, PISCATAWAY, NJ, USA, 18 October 2010 (2010-10-18), pages 1 - 6, XP031802310, ISBN: 978-1-4244-7498-1
ROBERT BORGOVINI ET AL: "Failure Mode, Effects and Criticality Analysis (FMECA)", CRTA-FMECA, 1 April 1993 (1993-04-01), Rome, NY 13442-4700, pages 1 - 101, XP055022662
Attorney, Agent or Firm:
ANDRES, Angelika (Weil am Rhein, DE)
Download PDF:
Claims:
Claims

1 . Method of assessment of a risk (R) involved in operating a measurement device on a measurement site in an industrial process, comprising the steps of:

- determining a degree of difficulty of a maintainability (M) of the device,

- determining a probability (P) of an undetected defect of the device,

- determining a severity (S) of the consequences of the defect of the device,

- weighing the determined probability (P), the determined severity (S) and the determined degree of difficulty of the maintainability (M) with corresponding predefined weighing factors (gs, gp, gM ) corresponding to a contribution of the degree of difficulty of the maintainability (M), the probability (P) and the severity (S) to the risk (R), and

- determining the risk (R) based on a length (LR) of a vector ( V ) whose coordinates are given by the weighted difficulty of maintainability (gM M), the weighted probability (gP P) and the weighted severity (gs S).

2. Method according to claim 1 , wherein the degree of difficulty of the

maintainability (M) of the device is determined based on an availability of resources for maintenance and/or repair of the device, an availability of spare and/or wear parts for the device and/or an availability of suitable replacement devices.

3. Method according to claim 1 , wherein the probability (P) of an undetected defect of the device is determined based

- on device specific properties, in particular on an expected life time of the device, its operating hours and/or its statistical failure rate,

- on measurement site specific properties, in particular on environment conditions prevailing at the measurement site, in particular an exposure of the measurement device to high or varying temperatures and/or pressures, to aggressive and/or abrasive media and/or to vibrations, and/or

- on means available at the measurement site for detecting a defect of the device and a frequency of their application.

4 . Method according to claim 1 , wherein severity (S) of the consequences of the undetected defect is determined based on the consequences of the defect, in particular on consequences regarding safety hazards for people, for the environment and/or for the measurement site and an industrial production plant it is embedded in, on consequences regarding a violation of rules, regulations and/or requirements the device is required to comply to, on consequences regarding quality, quantity and/or availability of end -, intermediate - or by-products produced by the industrial process, and/or on consequence regarding operating costs involved in running the industrial process.

5. Method according to claim 1 , wherein a quantitative determination of

the probability (P), the severity (S) and the degree of difficulty of the maintainability (M) is performed on discrete scales given by predefined levels for the probability (P), severity (S) and the degree of difficulty of the maintainability (M), in particular a low level (LOW), a medium level

(MEDIUM) and a high level (HIGH) for the probability (P), the severity (S) and the degree of difficulty of the maintainability (M).

6. Method according to claim 5, wherein a fixed numerical value is assigned to each level (LOW, MEDIUM, H IGH) of the probability (P), the severity (S) and the degree of difficulty of the maintainability (M)

7. Method according to claim 1 , wherein the risk (R) involved in operating the measurement device is classified based on the length (LR) of a vector ( V ) on a discrete scale given by predefined risk levels, in particular a low level comprising a range of vector lengths representing a low risk, a medium level comprising a range of vector lengths representing a medium risk, a high level comprising a range vector lengths representing a high risk, and an

unbearable level comprising a range of vector lengths representing an unbearable risk.

Description:
Method of assessment of a risk involved in operating a measurement device on a measurement site in an industrial process

The present invention concerns a method of assessment of a risk involved in operating a measurement device on a measurement site in an industrial process.

Measurement devices are used in nearly all branches of industry for measuring physical quantities, in particular quantities related to ongoing production processes. Measurement indications indicating the value of the quantity measured by the device are for example commonly used in process automation for monitoring, controlling and/or regulating an industrial process, in particular a production process. Thus measurement devices play a vital part in industrial processes, and a defect of a device may have severe consequences. Industrial production sites are quite often very complex sites including a large number of measurement devices on various measurement sites. In order to ensure and/or improve quality and safety of the industrial process it is advantageous, to perform a criticality analysis preferably for every measurement device on the site. Today criticality of a device is typically determined based on a product of a probability of a defect of the respective device and a severity of the consequences of this defect. Criticality analysis thus allows to identify the device forming the greatest risk in terms of probability and severity of the consequences to the overall performance of the process. Thus it is a valuable means in identifying candidates for reliability improvement and/or optimizing maintenance and calibration or verification of measurement devices as well as safety measures, such as provisions of redundant systems. Even though criticality analysis is a vital instrument, e.g. in optimizing

maintenance procedures for the devices, it does not take into account, that maintenance of a device may not always be possible. This is for example the case, when spare and/or wear parts for the device or suitable replacement devices are no longer available on the market. It is an object of the invention to provide a method of assessment of a risk involved in operating a measurement device on a measurement site in an industrial process, which takes into account a maintainability of the device. To this extend the invention comprises a method of assessment of a risk involved in operating a measurement device on a measurement site in an industrial process, comprising the steps of:

- determining a degree of difficulty of a maintainability of the device,

- determining a probability of an undetected defect of the device,

- determining a severity of the consequences of the defect of the device,

- weighing the determined probability, the determined severity and the determined degree of difficulty of the maintainability with corresponding predefined weighing factors corresponding to a contribution of

the degree of difficulty of the maintainability, the probability and the severity to the risk, and

- determining the risk based on a length of a vector whose

coordinates are given by the weighted difficulty of maintainability, the weighted probability and the weighted severity. According to a refinement of the invention, the degree of difficulty of the maintainability of the device is determined based on an availability of resources for maintenance and/or repair of the device, an availability of spare and/or wear parts for the device and/or an availability of suitable replacement devices. According to a further refinement the probability of an undetected defect of the device is determined based

- on device specific properties, in particular on an expected life time of the device, its operating hours and/or its statistical failure rate,

- on measurement site specific properties, in particular on environment conditions prevailing at the measurement site, in particular an exposure of the measurement device to high or varying temperatures and/or pressures, to aggressive and/or abrasive media and/or to vibrations, and/or

- on means available at the measurement site for detecting a defect of the device and a frequency of their application. According to a further refinement of the invention, severity of the consequences of the undetected defect is determined based the on the consequences of the defect, in particular on consequences regarding safety hazards for people, for the environment and/or for the measurement site and an industrial production plant it is embedded in, on consequences regarding a violation of rules, regulations and/or requirements the device is required to comply to, on consequences regarding quality, quantity and/or availability of end -,

intermediate - or by - products produced by the industrial process, and/or on consequence regarding operating costs involved in running the industrial process.

According to a further refinement a quantitative determination of

the probability, the severity and the degree of difficulty of the maintainability is performed on discrete scales given by predefined levels for the probability, severity and the degree of difficulty of the maintainability, in particular a low level, a medium level and a high level for the probability, the severity and the degree of difficulty of the maintainability.

According to a refinement of the last mentioned refinement, a fixed numerical value is assigned to each level of the probability, the severity and the degree of difficulty of the maintainability.

According to a further refinement of the invention, the risk involved in operating the measurement device is classified based on the length of a vector on a discrete scale given by predefined risk levels, in particular a low level comprising a range of vector lengths representing a low risk, a medium level comprising a range of vector lengths representing a medium risk, a high level comprising a range vector lengths representing a high risk, and an unbearable level comprising a range of vector lengths representing an unbearable risk.

The invention and further advantages are explained in more detail using the figure of the drawing.

Fig. 1 shows: a three dimensional graphic illustration of the assessed risk as a vector in a three-dimensional coordinate system described by three coordinates given by the weighted discrete level of the determined probability, the weighted discrete level of the determined severity and the weighted discrete level of the determined maintainability.

The method according to the invention concerns an assessment of a risk R involved in operating a measurement device on a measurement site in an industrial process.

It comprises the steps of:

- determining a probability P of an undetected defect of the device,

- determining a severity S of the consequences of the defect of the device, and

- determining a degree of difficulty of a maintainability M of the device.

Quantitative determination of the probability P, the severity S and the degree of difficulty of the maintainability M is preferably performed on a discrete scale given by predefined levels of probability, severity and difficulty of

maintainability. In order to do so the ranges of probabilities, severities and degrees of difficulty of maintainability are for example partitioned into a low level, a medium level and a high level range, and a fixed numerical value is assigned to each level. To this extend the low level can be assigned the low numerical value LOW = 1 , the medium level can be assigned the intermediate numerical value MEDIUM = 2 and the high level can be assigned the high numerical value HIGH = 3. Application of discrete scales facilitates

determination of these properties in a reproducible way.

A device is considered defect, in case it does no longer perform measurements with the measurement accuracy specified for it.

The probability P of an undetected defect of the device is preferably determined based on device specific properties as well as on measurement site specific properties and based on means provided at the measurement site, for detecting a defect of the device. The device specific properties preferably include an expected life time of the device, its operating hours and/or its statistical failure rate. Obviously, the probability P of an undetected defect will be lower for a device with proven robustness then for a device which is known to be more sensitive.

The measurement site specific properties of the device preferably include environment conditions prevailing at the measurement site, like for example an exposure of the measurement device to harsh conditions such as high or varying temperatures and/or pressures, aggressive and/or abrasive media and/or vibrations.

Means for detecting a defect of the device are for example redundant measurement systems or means for checking the measurement properties of the device. In case a redundant system is provided at the measurement site, a defect of the device will in almost all cases be detected immediately by comparison of the measurement results of the device and the redundant device. In consequence a defect of the device will only remain undetected, in case both devices exhibit the same defect at the same time. Thus there is a very low probability P of occurrence of an undetected defect.

Means for checking the measurement properties of the device are for example performances of regular on-site verifications of the device or periodic or sporadic executions of measurements of the property measured by the device with another device of the same or a different type. When means for checking the measurement properties of the device are foreseen, the probability P of an undetected defect is inversely proportional to the frequency of their application.

The severity S of the consequences of the defect of the device is preferably determined based on information pertinent to the industrial process and the measurement site embedded in this process. This information is for example retrieved by a thorough analysis of the production process in which preferably all potential consequences of the defect of the device are taken into account. The consequences of the defect include for example safety hazards the defect causes for people, for the environment and/or for the measurement site and the industrial production plant it is embedded in, consequences regarding a violation of rules, regulations and/or requirements the device is required to comply to, consequences regarding the quality, quantity and the availability of end -, intermediate - or by-products produced by the industrial process, as well as consequence regarding the operating costs involved in running the industrial process.

The degree of difficulty of the maintainability M of the device is preferably determined based on the availability of resources for maintenance and/or repair of the device, the availability of spare and/or wear parts for the device and/or suitable replacement devices. A high level of difficulty of maintainability M will for example be assigned to devices which have been phased-out for a long time, e.g. for more than four years. Also a high level of difficulty of

maintainability M will for example be assigned to devices for which spare parts or repair will only be available for a short time, e.g. for less than one year.

Correspondingly a medium level of difficulty of maintainability M will for example be assigned to devices which have been phased out not so long ago, e.g.

longer than 2,5 years ago but shorter than four years ago, or for which spare parts or repair will be available for a slightly longer time, e.g. for the next two years. A low level of difficulty of maintainability M will for example be assigned to devices for which a long term availability of replacements, spare parts and repair is given.

In a next step, the contributions of probability P, severity S and difficulty of maintainability M to the overall risk involved in operating the device are taken into account by weighing the determined probability P, the determined severity S and the determined difficulty of maintainability M each with a corresponding predefined weighing factor g s , gp, gM . Typically the contribution of the severity S of the consequence of the defect to the overall risk is larger than the contribution of the probability P of an undetected defect, and the difficulty of maintainability M has the least effect on the overall risk. Depending on the device, the measurement site, and the industrial process the weighing factors g s , gp, gM can for example be determined to be g s = 3, gp = 2 and gwi = 1■

This allows for every device used in the industrial process to be graphically illustrated as a vector V pointing to a discrete point in a three-dimensional coordinate system described by three coordinates given by the weighted discrete level of the determined probability g P P, the weighted discrete level of the determined severity g s S and the weighted discrete level of the determined difficulty of maintainability g M M. Fig 1 shows an example of the vector V in a scaled coordinate system, wherein each coordinate axis is represented scaled proportional to an inverse of the respective weighing factor g s , gp, gM.

In a last step the risk R involved in operating the device on the measurement site in this industrial process is determined based on a length L R of this vector

V given by:

L R = J(g s S) 2 + (g P P) 2 + (g M M) 2

Based on the length L R of the vector V , the risk R involved is preferably classified based on a discrete scale given by predefined risk levels. In order to do so the range of the possible lengths L R of the vector V is preferably partitioned into a range of vector lengths representing a low risk, a range of vector lengths representing a medium risk, a range of vector lengths representing a high risk, and a range of vector lengths representing an unbearable risk.

An example of these risk levels based on the weighted discrete level of the determined probability g P P, the weighted discrete level of the determined severity g s S and the weighted discrete level of the determined difficulty of maintainability g M M, using the numerical values LOW = 1 , MEDIUM = 2 and HIGH = 3, and the weighing factors g s = 3, gp = 2 and gM = 1 as described above is given in the table below.

Thus the invention provides a structured and relevant way to assess the risk R involved in operating the measurement device on the measurement site in this industrial process.

The method according to the invention can be performed for a single

measurement device on the industrial site, for all measurement devices used in a particular process performed on the industrial site, for all devices of a given type used on the industrial site, for all measurement devices applied in a selected area of the industrial site, or for all measurement devices applied on the industrial site.

It allows to identify highly risky situations and builds a tangible foundation to define pragmatic and cost optimized actions on the measurement site within the industrial process. It provides a profound basis for maintenance, calibration and device replacement schemes, as well as for the application of additional safety measures, like for example provisions of redundant measurement systems or means for periodic or sporadic checking of the measurement properties of individual devices.