SYSTEM AND METHOD FOR CALCULATING A PRODUCTIVITY OF AN INDUSTRIAL INSTALLATION

• US 20170161654A1
• Filed: 06/27/2014
• Published: 06/08/2017
• Est. Priority Date: 06/27/2014
• Status: Active Grant

• Current Assignee: Siemens AG
• Original Assignee: Siemens AG
• Inventors: Montrone, Francesco
• Granted Patent: US 10,628,776 B2
• Time in Patent Office: Days
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/0633 Workflow analysis
• CPC Class Codes: G06Q 10/067 Enterprise or organisation modelling
• CPC Class Codes: Y02P 90/80 Management or planning

Abstract

A method and a system for calculating a productivity of an industrial system consisting of system modules is provided. A Markov model is stored for each system module, the Markov model indicating the probabilities that a system module is in different operating states and indicating transition rates of state changes between the operating states of the respective system module for each operating state of the respective system module with a different productivity. A Cartesian product is calculated from a first Markov model of a first system module and from a second Markov model of a second system module linked to the first system module. The calculated Cartesian product is then reduced to a Markov model of the combined sub-system. The calculation of the Cartesian product and the subsequent reduction are carried out successively until the combined sub-system forms the entire industrial system.

Claims

1. A method for calculating a productivity of an industrial installation including installation modules:

storing a Markov model for each installation module, said Markov model indicating, for operating states of the relevant installation module with a different productivity, probabilities of the relevant installation module being in the different operating states and transition rates of state changes between the operating states of the relevant installation module,

calculating a Cartesian product from a first Markov model of a first installation module and from a second Markov model of a second installation module linked with the first installation module within the industrial installation or of a sub-installation linked with the first installation module within the industrial installation, said Cartesian product indicating, for operating states of a combined sub-installation including the two linked installation modules or of the sub-installation linked with the first installation module, probabilities of the combined sub-installation being in different operating states and transition rates of state changes between the operating states of the combined sub-installation, wherein the calculated Cartesian product is subsequently reduced to a Markov model of the combined sub-installation by those operating states of the calculated Cartesian product which have the same productivity being combined to form one operating state,

carrying out progressively, the calculation of the Cartesian product and the subsequent reduction thereof, until the combined sub-installation forms the entire industrial installation.

2. The method as claimed in claim 1, wherein the installation modules and sub-installations in each case have a state time proportion distribution indicating for at least two discrete operating states of the respective installation module or of the respective sub-installation, the time proportion of an operating time during which the installation module or the sub-installation is in the respective operating state.

3. The method as claimed in claim 2, wherein a time proportion of an operating time in which an installation module or a sub-installation is in an inactive operating state comprises a planned time proportion, in particular for carrying out maintenance and repair measures on the respective installation module, and a time proportion caused by failure events, in particular for restarting the installation module or the sub-installation.

4. The method as claimed in claim 1, wherein the state time proportion distributions are linked by linkage elements, which are selected in accordance with a configuration of the industrial installation, to form a layout model of the industrial installation, wherein, on the basis of the generated layout model, the productivity of the industrial installation with regard to at least one end product is calculated by a calculation unit.

5. The method as claimed in claim 4, wherein the state time proportion distributions of the selectable installation modules are read from a database to which the calculation unit has access.

6. The method as claimed in claim 1, wherein the time proportions indicated in the state time proportion distribution for the different operating states of the respective installation module are set depending on installation parameters and/or characteristic Figures of the respective installation module and/or depending on ambient parameters.

7. The method as claimed in claim 4, wherein the selectable linkage elements comprise arithmetic and/or logical linkage elements.

8. The method as claimed in claim 7, wherein one of the selectable linkage elements is formed by a combination linkage element, which generates, for each possible combination of operating states of upstream installation modules and/or sub-installations, a state time proportion distribution indicating, for each combined state, a productivity with regard to a product output by the upstream installation modules and/or sub-installations.

9. The method as claimed in claim 7, wherein one of the selectable linkage elements is formed by a minimization linkage element, which generates, for each possible combination of operating states of upstream installation modules and/or sub-installations, a state time proportion distribution indicating, for each combined state, the minimum of the productivities associated with the operating states with regard to a product output by the upstream installation modules and/or sub-installations.

10. The method as claimed in claim 1, wherein the installation modules and the linkage elements are linked with one another via linkage lines for generating the layout model of the industrial installation by a graphical editor.

11. The method as claimed in claim 1, wherein, for each end product of the industrial installation, a result state time proportion distribution is calculated which indicates, for each operating state of the entire industrial installation, a productivity with regard to the end product output by the industrial installation.

12. The method as claimed in claim 11, wherein the result state time proportion distributions of the different end products output by the industrial installation are weighted with the respective allocated value per produced quantity of the respective end product, wherein the total value of all the end products output by the industrial installation maximized.

13. The method as claimed in claim 1, wherein the end products of the industrial installation comprise chemical and/or biochemical and/or pharmaceutical products and/or sensor data and/or electrical power.

14. A planning software tool for planning an industrial installation having program commands for carrying out the method as claimed in claim 1.

15. A system for calculating a productivity of an industrial installation including installation modules, comprising

a database, in which a Markov model is stored for each installation module, said Markov model indicating, for operating states of the relevant installation module with a different productivity, probabilities of the relevant installation module being in the different operating states and transition rates of state changes between the operating states of the relevant installation module;

and comprising a calculation unit, which calculates a Cartesian product from a first Markov model of a first installation module and from a second Markov model of a second installation module linked with the first installation module within the industrial installation or of a sub-installation linked with the first installation module within the industrial installation, said Cartesian product indicating, for operating states of a combined sub-installation including the two linked installation modules or of the sub-installation linked with the first installation module, probabilities of the combined sub-installation being in different operating states and transition rates of state changes between the operating states of the combined sub-installation, wherein the calculated Cartesian product is subsequently reduced to a Markov model of the combined sub-installation by those operating states of the calculated Cartesian product which have the same productivity being combined to form one operating state, wherein the calculation of the Cartesian product and the subsequent reduction thereof are carried out progressively until the combined sub-installation forms the entire industrial installations.

16. An industrial installation comprising an installation controller which controls installation components of the industrial installation depending on productivities which are calculated with regard to end products of the industrial installation by a system as claimed in claim 15.