WO2021043423A1

WO2021043423A1 - Method and apparatus for operating an automation facility and automation facility

Info

Publication number: WO2021043423A1
Application number: PCT/EP2019/073872
Authority: WO
Inventors: Christian Bauer; Steffen Lamparter; Fabio Perna; Klaus SCHAUFLER
Original assignee: Siemens Aktiengesellschaft
Priority date: 2019-09-06
Filing date: 2019-09-06
Publication date: 2021-03-11

Abstract

A method for operating an automation facility for an automated production of a verified product is proposed. The method comprising: a) receiving a plurality of possible execution graphs, b) receiving a plurality of production orders, c) providing a production plan comprising the plurality of production orders and the plurality of possible execution graphs, d) matching, within the production plan, each of the received production orders to a respective one of the plurality of possible execution graphs, e) executing the possible execution graphs in order to produce a plurality of produced specific products, and f) for each of the produced specific products, verifying whether the respective produced specific product corresponds to verification parameters, and, if positive, transmitting the respective executed execution graph as a successful executed execution graph to a production database.

Description

Method and apparatus for operating an automation facility and automation facility

The present invention relates to a method and to a computer program for operating an automation facility. Further, the present invention relates to an apparatus for operating an au- tomation facility and to an automation facility.

There is a desire in the market to have individualized or spe- cialized products, which have to be produced as individual units or in small numbers. For such products, traditional pro- duction schemes which require individual tools and individual production planning are too expensive. Thus, to meet the de- mand at competitive cost, a higher flexibility of the produc- tion facilities or automation facilities is required.

The flexibility can be increased by using computer-controlled production apparatuses, which can be used to perform certain production steps within a range of parameter values. But, the flexibility in today's automation facilities is limited. This is because, amongst others, the flexibility of conventional production lines is limited due to manual effort and/or manual intervention of an operator in the production design and engi- neering phase as well as due to fixed production apparatuses.

However, all these above-described approaches of today's auto- mation facilities rely on explicitly given and validated flex- ibilities by humans.

Skill-based production planning approaches address this prob- lem by enabling autonomous production lines within automation facilities that automatically apply the best production pro- cess. It introduces descriptions of specific products as well as production apparatuses and an algorithm for deriving feasi- ble production workflows. Due to the high, almost unlimited variance enabled by autonomous production apparatuses, a cer- tification of all feasible workflows becomes very hard or even impossible.

It is one object of the present invention to improve the oper- ation of an automation facility.

According to a first aspect, a method for operating an automa- tion facility including a number of production apparatuses for an automated production of at least a verified product is pro- posed. The method comprising the steps: a) receiving a plurality E of possible execution graphs, wherein each of the possible execution graphs E is indicative of a sequence of production apparatus steps being executable by at least a subset of the number of production apparatuses for producing a respective specific product, with E ³ 2, wherein each of the specific products is allocated to a deter- mined product type, b) receiving a plurality P of production orders, wherein each of the production orders P is indicative of an order in- formation allocated to the determined product type for produc- ing one specific product of the determined product type, with P > 2, c) providing a production plan comprising at least the plurality P of production orders and the plurality E of possi- ble execution graphs, d) matching, within the production plan, each of the re- ceived production orders P to a respective one of the plurali- ty E of possible execution graphs, e) executing the plurality E of possible execution graphs in order to produce a plurality S of produced specific prod- ucts, with S ³ 2, and f) for each of the produced specific products S, verifying whether the respective produced specific product corresponds to verification parameters indicative of a verified produced product, and, if positive, transmitting the respective execut- ed execution graph as a successful execution graph to a pro- duction database.

With the above-described method, it is possible to transmit a successful execution graph to the production database. The successful execution graph indicates that the specific prod- uct, which has been produced by executing said execution graph, is a verified product. Thus, the production database comprises at least a successful executed execution graph. Af- ter executing the plurality E of possible execution graphs, a plurality or subset of successful execution graphs is trans- mitted to and stored in the production database. As a result, it can be avoided that the transmitted successful execution graphs stored the production database have to be tested or stored again after the respective executed execution graph is defined as a successful execution graph.

This has the advantage that the flexibility of the automation facility is increased due to the plurality of successful exe- cution graphs which are stored in the production database.

That is, because for a determined product type, a plurality of different successful execution graphs may be available in the production database. Thus, there are multiple possibilities, depending on the successful execution graphs, for producing a verified product. This is particularly important for cases, when a production apparatus, included in one of the successful execution graphs, is defective and unavailable. By means of the production database the automation facility is capable of choosing another successful executing graph for producing a verified product corresponding to the determined product type.

The above-described method has the further advantage that in the automation facility, with every executed execution graph out of the plurality E of possible execution graphs, the knowledge of the production database increases, in particular the number of successful execution graphs increases, and thus, the resilience of the automation facility is increased. The automation facility is preferably embodied as a fully au- tomated production facility or robotic plant, which is config- ured for fully automated production of specific products, starting from raw materials. The automation facility includes a number of production apparatuses, each of which is capable of performing certain production apparatus steps or production steps.

The production apparatuses may be referred to as autonomous machines, robots or automation devices, and are preferably computer-controlled. There may be a production apparatus for each production step that can be automated. Examples of pro- duction apparatuses are drills, mills, pumps, saws, grinders, mixers, welding robots, transporters, packagers, and so on. Preferably, at least two of the production apparatuses of the number of production apparatuses are different to each other, for example, one production apparatus is a mill while another production apparatus is a drill.

Each one of the production apparatuses is defined by its skillset, which includes the skills the respective production apparatus is capable of performing. For example, a production apparatus which is a mill may be capable of milling and drill- ing, wherein the precise skills may depend on the exact con- figuration of the mill and the tools which are provided for the mill. For example, the milling production apparatus may be equipped with an electric engine that has a maximum revolution per minute (RPM) of 5000, a milling head that may be moved in an x-y-plane measuring 50 cm by 50 cm, a milling tool for milling aluminum and a drill with a diameter of 5 mm providing a maximum drilling depth of 50 mm. Then, the skillset of this mill may include the two mentioned skills, which are defined by the mentioned parameters.

In particular, the determined product type is one type of a specific product. For example, product type A identifies a circuit board with two drilling holes, while product type B identifies a circuit board with four drilling holes.

Specifically, the specific product is a produced specific product. But it may be undetermined if the produced specific product corresponds to verification parameters indicative of a verified product. In other words, the produced specific prod- uct may correspond to an unverified product. In particular, the specific product is produced by executing one possible ex- ecution graph of the plurality E of execution graphs. It is desired that the produced specific product corresponds to ver- ification parameters indicative of a verified product. Prefer- ably, the specific product to be produced includes discrete specific products, such as a solid object made from wood, met- al, plastics, or the like, and/or electronic articles, and further includes formulated specific products that are ob- tained by a process (product-by-process products), such as chemical substances, cream, food, and the like. The specific product may include complex specific products which comprise a plurality of parts, each part being a specific product itself.

Preferably, the verified product is a produced specific prod- uct which fulfills all requirements according to successful product tests, product verifications and/or certification re- quirements. In particular, a produced specific product is a verified product, if the produced specific product corresponds to verification parameters indicative of a verified product.

In particular, a possible execution graph out of the plurality E of possible execution graph is one possibility to produce a specific product. Each possible execution graph is indicative of a sequence of production apparatus steps, executable by at least a subset of the number of production apparatuses. Thus, the produced specific product corresponds to the verified product or not. A production order is an instruction for producing a specific product. Preferably, the production order comprises an order information allocated to the determined product type for pro- ducing one specific product of the determined product type. Specifically, the order information comprises relevant parame- ters used for producing a specific product, in detail, at least the determined product type and the possible execution graph which includes the sequence of production apparatus steps for producing the specific product allocated to the de- termined product type.

The sequence of production apparatus steps may include a bill of materials, that is, the raw materials or intermediate prod- ucts required to produce the specific product allocated to the determined product type. The sequence of production apparatus steps, when fully conducted by the automation facility, leads to the specific product to be produced. Further, the sequence of production apparatus steps may be provided, for a given product type, by a customer of the specific product or product type. It may also be provided by an operator based on a de- scription or specification of the specific product to be pro- duced.

For example, a specific product allocated to a product type A is produced by executing at least one possible execution graph which is capable of producing said specific product. Specifi- cally, it is also possible that the specific product allocated to the product type A is produced by executing another possi- ble execution graph. For example, for one product type A, it is possible that multiple possible execution graphs are pro- vided for respectively producing a specific product allocated to the product type A.

The production plan comprises the plurality P of production orders, the determined product type and the plurality E of possible execution graphs. In particular, within the produc- tion plan, each of the possible execution graphs comprises a sequence of production apparatus steps. Each production appa- ratus step may require one or more skills of a respective pro- duction apparatus which is capable of performing this skill or the skills.

According to an embodiment, in step f), the successful execut- ed execution graph is additionally transmitted to a production planning unit which comprises the plurality E of possible exe- cution graphs and the plurality P of production orders.

Thus, by repeating the above-described method for operating an automation facility and transmitting every successful graph to the production planning unit, the number of possible execution graphs in the production planning unit decreases over the time because an increasing number of possible execution graphs in the production planning unit correspond to successful executed execution graphs. Therefore, it can be avoided that the trans- mitted successful executed execution graphs have to be tested again after the respective executed execution graph is defined as a successful execution graph. Thus, it is unnecessary that the production planning unit has to provide all possible exe- cution graphs and production orders in order to verify if an executed possible execution graphs results in a verified prod- uct. This has the advantage that the computing time for test- ing all possible execution graphs at every producing of a spe- cific product is decreased and therefore the hardware costs can be decreased.

The production planning unit may be implemented in hardware and/or in software.

According to a further embodiment, the matching in step d) is a bijective matching of the received production orders P to a respective one of the plurality E of possible execution graphs. This increases the usability for an operator of the automation facility.

For example, in the production plan, a possible execution graph El is bijectively matched to a production order PI. Preferably, bijective matching is the possibility to conclude an allocation from the first matching object to the second matching object and vice versa. This means for the present case, that it is possible to conclude from the possible execu- tion graph El to the production order PI and from the produc- tion order PI to the possible execution graph El.

According to a further embodiment, in step e), the plurality E of possible execution graphs are executed depending on the production apparatus steps being indicated by the respective possible execution graph in series or in parallel.

By executing the plurality E of possible execution graphs in parallel, the time for producing a plurality S of specific products which are then later verified, can be decreased. This has the advantage that, due to the parallelization, the amount of produced specific products may be increased.

For example, the term "depending on the production apparatus steps" of this further embodiment is described by the follow- ing example: It is possible that one possible execution graph El requires for the production of a specific product the se- quence of production apparatus steps according to the produc- tion apparatuses (PAx): PA1

PA3. At the same time, another possible execution graph E2 requires the following se- quence of production apparatuses in the automation facility: PA1 PA4. Every production apparatus PA1 to PA4 is ca- pable of performing a skill like milling, drilling, or the like. Thus, no parallel executing of the execution graphs El and E2 is possible because each of them needs the same produc- tion apparatuses (PA1 and PA2) at the same time. Therefore, in this case, only a serial execution of El and E2 is possible. According to a further embodiment, in step c), the providing of the production plan comprises at least a calculation, in which at least a target value function is maximized, wherein the target value is at least one type of requirement according to which the specific product is produced depending on the provided production plan.

By calculating the maximum of the target value or the target value function, it is possible to determine types of require- ments according to which the specific product is produced de- pending on the provided production plan. For example, if one type of requirement is to produce the specific product quick- ly, the successful execution graph, which is capable to pro- duce the specific product quickly, can be chosen by the pro- duction plan and can then be executed. This has the advantage that an operator of the automation facility can determine dif- ferent requirements according to which at least one specific product can be produced in dependence of the chosen execution graph. This increases the flexibility of the automation facil- ity.

For example, a target value comprises at least one type of re- quirement according to which the specific product is produced depending on the provided production plan. Preferably, one type of requirement is to produce the specific product quick- ly, another type is to produce it efficiently (low power con- sumption) and/or to produce it with a high quality.

According to a further embodiment, the calculation of the max- imum of the target value function is performed by using, as an input for the calculation, actual measurement values and/or historic measurement values obtained by the production data- base.

Specifically, actual measurement values and historic measure- ment values may comprise measurement values in which a mean value over all process durations for a service, a mean value for all estimated processing times of each executed execution graphs and/or a mean value for all estimated energy consump- tion of each executed execution graphs, is formed.

According to a further embodiment, the step c) of providing the production plan is performed by using a machine learning algorithm, wherein the machine learning algorithm is at least applied to the plurality E of possible execution graphs and the plurality P of production orders.

By using a machine learning algorithm, it is possible that the production plan gets new learned matches at least between the plurality E of execution graphs and the plurality P of produc- tion orders. Thus, the production plan includes more infor- mation or a better knowledge about new matches between execu- tion graphs and production orders. This has the advantage that the resilience of the automation facility is increased because more possibilities are available, i.e., more execution graphs for producing a specific product.

Preferably, the machine learning algorithm is a reinforcement learning algorithm.

According to a further embodiment, the providing of the pro- duction plan comprises adding a flexibility value component to the calculated target value function, in order to add further key-performance-indicators to the provided production plan.

According to a further embodiment, the further key- performance-indicators comprise at least details about pro- cessing time of the executed execution graphs, estimated power consumption of the production apparatus steps of the executed execution graphs and/or duration of service of the production apparatuses of the executed execution graph. In particular, a key-performance-indicator or a further key performance indicator comprises details about an estimated processing time and/or an estimated energy consumption of an executed execution graph for producing the specific product.

Specifically, additional key-performance-indicators are a uti- lization of a production apparatus, an availability of a pro- duction apparatus and/or the certified/test measurements of an executed execution graph.

According to a further embodiment, in step f), additionally the respective further key-performance-indicators, which are allocated to the successful executed execution graphs, are transmitted to the production database.

Thus, the production database has a better knowledge of the combination or interaction between the executed execution graphs, produced specific products and the key-performance- indicators. Thus, the machine learning algorithm has an in- creased database knowledge available which can be used for a better training and learning of new matches within the produc- tion plan. As a result, the flexibility of the automation fa- cility is increased.

According to a further embodiment, the automation facility comprises a plurality T of determined product types, with T ³ 2, wherein a first produced specific product is allocated to a first determined product type, while a second produced specif- ic product is allocated to a second determined product type.

The ability to produce a plurality T of determined product types has the advantage that the flexibility of the automation facility is increased.

In this further embodiment, the automation facility is capable to produce a plurality T of determined product types, with T ³ 2. In particular, the plurality T comprises determined product types where the first determined product type is different to the second determined product type.

According to a further embodiment, the production database is formed as a relational production database.

In particular, in the relational production database, the da- tabase is managed or organized line by line in tables.

According to a second aspect, a computer program is proposed which comprises program code for executing the above-described method according to the first aspect or according to an embod- iment of the first aspect when run on at least one computer.

A computer program product, such as a computer program means, may be embodied as a memory card, USB stick, CD-ROM, DVD or as a file which may be downloaded from a server in a network. For example, such a file may be provided by transferring the file comprising the computer program product from a wireless commu- nication network.

According to a third aspect, an apparatus for operating an au- tomation facility including a number of production apparatuses for an automated production of at least a verified product is proposed. The apparatus comprises: a first receiving unit, configured to receive a plurality E of possible execution graphs, wherein each of the possible execution graphs E is indicative of a sequence of production apparatus steps being executable by at least a subset of the number of production apparatuses for producing a respective specific product, with E ³ 2, wherein each of the specific products is allocated to a determined product type, a second receiving unit, configured to receive a plurality P of production orders, wherein each of the production orders P is indicative of an order information allocated to the de- termined product type for producing one specific product of the determined product type, with P ³ 2, a providing unit, configured to provide a production plan comprising at least the plurality P of production orders and the plurality E of possible execution graphs, a matching unit, configured to match, within the produc- tion plan, each of the received production orders P to a re- spective one of the plurality E of possible execution graphs, an executing unit, configured to execute the plurality E of possible execution graphs in order to produce a plurality S of produced specific products, with S ³ 2, and a verifying unit, configured to verify, for each of the produced specific products S, whether the respective produced specific product corresponds to verification parameters indic- ative of a verified produced product, and, if positive, the verifying unit is further configured to transmit the respec- tive executed execution graph as a successful executed execu- tion graph to a production database.

The automation facility preferably includes an apparatus or the like, which controls operation of each of the production apparatuses. The apparatus may be implemented in hardware and/or in software. If the apparatus is implemented in hard- ware, it may be embodied as a device, e.g. as a computer or as a processor or as a part of a system, e.g. a computer system.

If the apparatus is implemented in software, it may be embod- ied as a computer program product, as a function, as a rou- tine, as a program code or as an executable object. For exam- ple, the apparatus creates a production plan based on the se- quence of production apparatus steps provided.

The respective unit, e.g. the verifying unit or the executing unit, may be implemented in hardware and/or in software. If said unit is implemented in hardware, it may be embodied as a device, e.g. as a computer or as a processor or as a part of a system, e.g. a computer system. If said unit is implemented in software it may be embodied as a computer program product, as a function, as a routine, as a program code or as an executa- ble object. According to a fourth aspect, an automation facility is pro- posed which comprises a number of production apparatuses for an automated production of at least the verified product, the above-described apparatus according to the third aspect and the production database.

Any embodiment of the first aspect may be combined with any embodiment of the first aspect to obtain another embodiment of the first aspect.

The embodiments and features described with reference to the method of the first aspect apply mutatis mutandis to the appa- ratus.

Further possible implementations or alternative solutions of the invention also encompass combinations - that are not ex- plicitly mentioned herein - of features described above or be- low with regard to the embodiments. The person skilled in the art may also add individual or isolated aspects and features to the most basic form of the invention.

Further embodiments, features and advantages of the present invention will become apparent from the subsequent description and dependent claims, taken in conjunction with the accompany- ing drawings, in which:

Fig. 1 shows a flow chart illustrating steps of a method for operating an automation facility according to an em- bodiment;

Fig. 2 shows a block diagram of an apparatus for operating an automation facility according to an embodiment;

Fig. 3 shows a block diagram of an automation facility ac- cording to an embodiment comprising the apparatus of Fig. 2; and Fig. 4 shows a table of a production plan according to an embodiment.

In the Figures, like reference numerals designate like or functionally equivalent elements, unless otherwise indicated. In the following, by referring to Figs. 1, 2 and 3, an automa- tion facility 1 are well as a method and an apparatus 100 for operating the automation facility 1 is described. This in- cludes the method for operating the automation facility 1 as illustrated in Fig. 1, the apparatus 100 for operating the au- tomation facility 1 as shown in Fig. 2 and the environment of the automation facility 1 comprising the apparatus 100 and a production database 20 as shown in Fig. 3.

In Fig. 2, an apparatus 100 for operating an automation facil- ity 1 according to an embodiment is shown. The automation fa- cility 1 is shown in Fig. 3 and comprises at least the appa- ratus 100 for operating the automation facility 1, the produc- tion database 20 and a number of production apparatuses 41 for an automated production of at least one verified product.

In Fig. 3, each of the illustrated production apparatuses 41 may be different to each other. One production apparatus 41 can be embodied as a drill, while another production apparatus 41 can be embodied as a mill. A connector 40 connects the pro- duction apparatuses 41 and the production database 20 to the apparatus 100 via communication links 42. The connector 40 and the communication links 42 may also be arranged in a different way. For example, the production apparatuses 41 can be con- nected directly to the apparatus 100 and/or the production da- tabase 20 may be connected directly via the communication links 42 to the apparatus 100 without using the connector 40. Thereby, in this embodiment, the production database 20 is formed as a relational production database. In particular, the automation facility 1 of Fig. 3 comprises a plurality T of determined product types 22, with T ³ 2. In this regard, a first produced specific product 29 is allocated to a first determined product type 22, while a second produced specific product 29 is allocated to a second determined prod- uct type 22.

In Fig. 1, a flow chart illustrating steps of the method for operating an automation facility 1 according to an embodiment is shown. The method comprises the steps S101 to S106. The ap- paratus 100 of Fig. 2 is configured to execute the steps S101 to S106 of Fig. 1.

Further, the apparatus 100 according to Fig. 2 comprises a first receiving unit 21a, a second receiving unit 21b, a providing unit 24a, a matching unit 24b, an executing unit 27 and a verifying unit 28.

The first receiving unit 21a is configured to receive a plu- rality E of possible execution graphs 23. This receiving cor- responds to step S101 in Fig. 1. Each of the possible execu- tion graphs E 23 is indicative of a sequence of production ap- paratus steps being executable by at least a subset of the number of production apparatuses 41 for producing a respective specific product 29, with E ³ 2. In addition, each of the spe- cific products 29 is allocated to a determined product type 22.

Moreover, the second receiving unit 21b is configured to re- ceive a plurality P of production orders 25. In particular, this receiving corresponds to step S102 in Fig. 1. Each of the production orders P 25 is indicative of an order information allocated to the determined product type 22 for producing one specific product 29 of the determined product type 22, with P ³ 2. The first and the second receiving units 21a, 21b are part of a production planning unit 21 in Fig. 2. The produc- tion planning unit 21 comprises at least the plurality E of possible execution graphs 23 and the plurality P of production orders 25. The second receiving unit 21b is may be configured to receive at least the determined product type 22. Further- more, a combined input/output of the production database 20 is connected to an input of the first and second receiving units 21a, 21b.

An output of the first and second receiving units 21a, 21b is connected to a first input of a scheduler 24 of the providing unit 24a and the matching unit 24b in order to provide the plurality E of possible execution graphs 23 to the providing unit 24a and to the matching unit 24b.

Further, the providing unit 24a of Fig. 2 is then configured to provide a production plan 26. The providing corresponds to step S103 of Fig. 1. This production plan 26 comprises the plurality P of production orders 25 and the plurality E of possible execution graphs 23. Furthermore, the providing of the production plan 26 comprises at least a calculation, in which at least a target value function is maximized. This tar- get value is at least one type of requirement according to which the specific product 29 is produced depending on the provided production plan 26. The input for the calculation of the maximum of the target value function are actual measure- ment values and/or historic measurement values which are ob- tained by the production database 20.

Furthermore, the providing unit 24a of Fig. 2 may be config- ured to provide the production plan 26 by using a machine learning algorithm. In particular, the machine learning algo- rithm is at least applied to the plurality E of possible exe- cution graphs 23 and the plurality P of production orders 25.

Next, the matching unit 24b is configured to match, within the production plan 26, each of the received production orders P 25 to a respective one of the plurality E of possible execu- tion graphs 23. This corresponds to step S104 of Fig. 1. The performed matching may be a bijective matching of the received production orders P 25 to a respective one of the plurality E of possible execution graphs 23. A second input of the sched- uler 24 of Fig. 2 is connected to a further output of the pro- duction database 20.

The output of the scheduler 24 is connected to an input of the executing unit 27.

After the matching, the executing unit 27 executes the plural- ity E of possible execution graphs 23 in order to produce a plurality S of specific products 29, with S ³ 2. This execu- tion corresponds to step S105 in Fig. 1. Additionally, it is possible that the plurality E of possible execution graphs 23 are executed depending on the production apparatus steps being indicated by the respective possible execution graph in series or in parallel. An output of the executing unit 27 is connect- ed to an input of the verifying unit 28.

The verifying unit 28 is configured to verify, for each of the produced specific products 29, whether the respective produced specific product 29 corresponds to verification parameters in- dicative of a verified produced product. If this verification is positive, that means that the respective produced specific product 29 corresponds to the verification parameters indica- tive of the verified produced product, the verifying unit 28 transmits the respective executed execution graph as a suc- cessful execution graph 30 over a connection between the veri- fying unit 28, the executing unit 27 and the production data- base 20 to an input to the production database 20. The verify- ing and the transmitting correspond to step S106 in Fig. 1.

In this embodiment, the successful execution graph 30 is addi- tionally transmitted, over the connection between the produc- tion planning unit 21 and the production database 20, to the production planning unit 21. Fig. 4 shows a table of a production plan 26 according to an embodiment. The production plan 26 may include further columns and lines with additional information.

The production plan 26 in Fig. 4 comprises a column with dif- ferent production orders 25, a column with different deter- mined product types 22, a column with different execution graphs 23, a column with different verification results, a column with different processing times, a column with differ- ent energy consumption and a column with different further key-performance-indicators 31 (KPIs in Fig.4). The other ref- erence numerals are shown in Fig. 2.

With reference to line 2 of the table of the production plan 26, the table is explained in more detail: In line 2/1. Col- umn, the production order 25 defined as A-1254. This produc- tion order 25 is allocated to the product type A. Thus, with this production order 25, the product type A can be produced. Further, in line 2, for producing a specific product 29 which corresponds to product type A, the execution graph 23 with the sequence of production apparatus steps PA2 - PA4 - PA5 is used. PA is the acronym for production apparatus. PA2 is the production apparatus 2 which is, for example, configured to drill. Also, the execution graphs 25 in the lines 1 and 3 can be used to produce the specific product 29. But, the execution graph in line 3 leads later to an unverified specific product, see "no" in the fourth column

Next, after executing the execution graph in line 2, the spe- cific product 29 which corresponds to product type A, is pro- duced. Then, a verification of this specific product 29 is performed in order to verify, whether the produced specific product 29 corresponds to verification parameters indicative of the verified produced product. If positive, this is defined with the word "yes" in line 2. Moreover, during the execution of the steps S101 to S106 in Fig. 1, (further) key-performance-indicators 31 (KPIs in Fig.4) can be obtained and stored. For example, the providing of the production plan 26 in step S103 in Fig. 1 may comprise adding a flexibility value component to the calculated target value function, in order to add the further key-performance- indicators 31 to the provided production plan 26.

As shown in Fig. 4, a key performance indicator KPI comprises at least details about processing time (see line 2) of the ex- ecuted execution graph 23 and estimated power consumption (see line 2) of the production apparatus steps of the executed exe- cution graph 23. In addition, a further key-performance- indicator 31 may comprise details about duration of service of the production apparatuses 41 of the executed execution graph 23. For updating the production database 20, the respective further key-performance-indicators 31, which are allocated to successful execution graphs 30, for example in line 2 or line 1, are transmitted to the production database 20.

Although the present invention has been described in accord- ance with preferred embodiments, it is obvious for the person skilled in the art that modifications are possible in all em- bodiments.

Reference Numerals:

1 automation facility

20 production database

21 production planning unit

21a receiving unit

21b receiving unit

22 product type

23 execution graph

24 scheduler 24a providing unit 24b matching unit

25 production order

26 production plan

27 executing unit

28 verifying unit

29 specific product

30 successful execution graph

31 key-performance-indicators

40 connector

41 production apparatus

42 communication links 100 apparatus

S101 to S106 method steps

Claims

Patent claims

1. A method for operating an automation facility (1) includ- ing a number of production apparatuses (41) for an automated production of at least a verified product, the method compris- ing: a) receiving (S101) a plurality E of possible execution graphs (23), wherein each of the possible execution graphs E (23) is indicative of a sequence of production apparatus steps being executable by at least a subset of the number of produc- tion apparatuses (41) for producing a respective specific product (29), with E ³ 2, wherein each of the specific prod- ucts (29) is allocated to a determined product type (22), b) receiving (S102) a plurality P of production orders (25), wherein each of the production orders P (25) is indicative of an order information allocated to the determined product type (22) for producing one specific product (29) of the determined product type (22), with P ³ 2, c) providing (S103) a production plan (26) comprising at least the plurality P of production orders (25) and the plu- rality E of possible execution graphs (23), d) matching (S104), within the production plan (26), each of the received production orders P (25) to a respective one of the plurality E of possible execution graphs (23), e) executing (S105) the plurality E of possible execution graphs (23) in order to produce a plurality S of produced spe- cific products (29), with S ³ 2, and f) for each of the produced specific products S (29), verify- ing (S106) whether the respective produced specific product (29) corresponds to verification parameters indicative of a verified produced product, and, if positive, transmitting the respective executed execution graph as a successful execution graph (30) to a production database (20).

2. The method according to claim 1, characterized in that, in step f) (S106), the successful execution graph (30) is additionally transmitted to a production planning unit (21) which comprises the plurality E of possible execution graphs (23) and the plurality P of production orders (25).

3. The method according to claim 1 or 2, characterized in that the matching in step d) (S104) is a bijective matching of the received production orders P (25) to a respective one of the plurality E of possible execution graphs (23).

4. The method according to one of claims 1 - 3, characterized in that, in step e) (S105), the plurality E of possible execution graphs (23) are executed depending on the production apparatus steps being indicated by the respective possible execution graph (23) in series or in parallel.

5. The method according to one of claims 1 - 4, characterized in that, in step c), the providing (S103) of the production plan (26) comprises at least a calculation, in which at least a target value function is maximized, wherein the target value is at least one type of requirement according to which the specific product (29) is produced depending on the provided production plan (26).

6. The method according to claim 5, characterized in that the calculation of the maximum of the target value func- tion is performed by using, as an input for the calculation, actual measurement values and/or historic measurement values obtained by the production database (20).

7. The method according to claim 5 or 6, characterized in that the step c) of providing (S103) the production plan (26) is performed by using a machine learning algorithm, wherein the machine learning algorithm is at least applied to the plu- rality E of possible execution graphs (23) and the plurality P of production orders (25).

8. The method according to one of claims 5 - 7, characterized in that the providing (S103) of the production plan (26) compris- es adding a flexibility value component to the calculated tar- get value function, in order to add further key-performance- indicators (31) to the provided production plan (26).

9. The method according to claim 8, characterized in that the further key-performance-indicators (31) comprise at least details about processing time of the executed execution graphs (23), estimated power consumption of the production ap- paratus steps of the executed execution graphs (23) and/or du- ration of service of the production apparatuses (41) of the executed execution graph (23).

10. The method according to claim 8 or 9, characterized in that, in step f) (S106), additionally the respective further key-performance-indicators (31), which are allocated to the successful executed execution graphs (30), are transmitted to the production database (20).

11. The method according to one of claims 1 - 10, characterized in that the automation facility (1) comprises a plurality T of determined product types (22), with T ³ 2, wherein a first produced specific product (29) is allocated to a first deter- mined product type (22), while a second produced specific product (29) is allocated to a second determined product type (22).

12. The method according to one of claims 1 - 11, characterized in that the production database (20) is formed as a relational production database.

13. A computer program product comprising a program code for executing the method according to one of the claims 1 to 12 when run on at least one computer.

14. An apparatus (100) for operating an automation facility (1) including a number of production apparatuses (41) for an automated production of at least a verified product, wherein the apparatus (100) comprising: a first receiving unit (21a), configured to receive a plu- rality E of possible execution graphs (23), wherein each of the possible execution graphs E (23) is indicative of a se- quence of production apparatus steps being executable by at least a subset of the number of production apparatuses (41) for producing a respective specific product (29), with E ³ 2, wherein each of the specific products (29) is allocated to a determined product type (22), a second receiving unit (21b), configured to receive a plurality P of production orders (25), wherein each of the production orders P (25) is indicative of an order information allocated to the determined product type (22) for producing one specific product (29) of the determined product type (22), with P ³ 2, a providing unit (24a), configured to provide a production plan (26) comprising at least the plurality P of production orders (25) and the plurality E of possible execution graphs (23), a matching unit (24b), configured to match, within the production plan (26), each of the received production orders P (25) to a respective one of the plurality E of possible execu- tion graphs (23), an executing unit (27), configured to execute the plurali- ty E of possible execution graphs (23) in order to produce a plurality S of produced specific products (29), with S ³ 2, and a verifying unit (28), configured to verify, for each of the produced specific products S (29), whether the respective produced specific product (29) corresponds to verification pa- rameters indicative of a verified produced product, and, if positive, the verifying unit (28) is further configured to transmit the respective executed execution graph as a success- ful execution graph (30) to a production database (20).

15. An automation facility (1), comprising a number of produc- tion apparatuses (41) for an automated production of at least the verified product, the apparatus (100) according to claim 14 and the production database (20).