Method and apparatus for providing recommendations for completion of engineering projects

Abstract

The invention relates to a recommendation engine (1) for automatically providing recommendations for completion of an engineering project, the recommendation engine (1) comprising a first artificial intelligence, AI, module (1A) adapted to provide a latent representation of a sequence of selected items, and a second AI module (1B) adapted to process the latent representation of the sequence of selected items provided by the first AI module (1A) to generate at least one sequence of complementary items required to complement the sequence of selected items, thereby providing a complete sequence of items output via an interface (1C) as a recommendation to complete the engineering project.

Description

[0001] The present invention relates to a method and apparatus for providing recommendations for the completion of a complex engineering project, in particular an automation system.

[0002] Engineering projects, such as automation systems, can be complex and comprise a large number of different components. The configuration of a complex engineering project can comprise an iterative process in which a user selects components incrementally. The combination of these selected components can fulfill the functional requirements of the engineering project while also being compatible with each other. The configuration of a complex engineering process is not an easy task and requires time, effort, experience, and a certain amount of domain-specific knowledge in order to be completed correctly by a user.

[0003] It is therefore an object of the present invention to provide a method and apparatus for providing recommendations for the completion of an engineering project.

[0004] According to a first aspect of the invention, the object is achieved by a recommendation engine comprising the features of claim 1.

[0005] According to the first aspect, the invention provides a recommendation engine for providing recommendations for the completion of an engineering project, the recommendation engine comprising:

[0006] a first artificial intelligence module adapted to provide a latent representation of a set of items; and

[0007] a second artificial intelligence module adapted to process the latent representation of the set of items provided by the first artificial intelligence module to generate at least one sequence of complementary items required to complement the set of items, thereby providing a complete sequence of items outputted via an interface as a recommendation to complete the engineering project.

[0008] Providing recommendations for the completion of a partially configured engineering project reduces the time required for selecting components. Furthermore, the process of selecting items associated with components of an engineering process can be performed by a less experienced user with less domain-specific knowledge. The recommendation engine can be used for any kind of engineering project, in particular for different kinds of automation complex systems comprising a plurality of different components, i.e. hardware and / or software components.

[0009] In a possible embodiment of the recommendation engine according to the first aspect of the invention, the items are selected from a set of available items corresponding to hardware and / or software components available for the respective engineering project.

[0010] Each item can correspond to an associated hardware component, such as a controller, or to a software component, such as an application. Thus, the recommendation engine according to the present application can be used for a wide range of different engineering scenarios, which not only encompass hardware components, but also software components.

[0011] The information about the order of the selected items in the sequence provides additional context information, which supports the completion of the required items of the respective engineering scenario.

[0012] In a further possible embodiment of the recommendation engine according to the first aspect of the present application, the set of selected items is at least temporarily stored in a memory connected to the recommendation engine. Thus, a loss of selected items can be avoided.

[0013] In a further possible embodiment of the recommendation engine according to the first aspect of the present application, the first artificial intelligence module comprises a trained feature learning module, which is adapted to compute a latent representation of the set of items.

[0014] The latent representation computed by the first artificial intelligence module can encode technical information about the components of the engineering scenario.

[0015] In a further possible embodiment of the recommendation engine according to the first aspect of the present application, the second artificial intelligence module comprises a trained sequential model, which is adapted to compute at least one sequence of supplementary items as output for the recommendation to complete the engineering scenario.

[0016] The trained sequential model can exploit temporal dependencies between items selected during the engineering scenario.

[0017] In a further possible embodiment of the recommendation engine according to the first aspect of the present application, the items are selected by a user via a user interface, which has a screen adapted to output available items to the user. This facilitates the selection of available items.

[0018] In a further possible embodiment of the recommendation engine according to the first aspect of the present application, the one or more sequences of supplementary items generated by the second artificial intelligence module are output on the screen of the user interface for selecting the next item from one of the sequences of supplementary items, or for selecting one or more items from one of the sequences of supplementary items (which items do not necessarily appear one after the other), or for selecting the entire sequence of supplementary items by the user.

[0019] This provides the advantage that the user has the choice whether to select a single next item or to select an entire sequence of complementary items in order to complete the selection immediately. Thus, there is an automated mechanism for automatically completing a partially configured engineering project.

[0020] In a further possible embodiment of the recommendation engine according to the first aspect of the application, the first and second artificial intelligence modules comprise an artificial neural network trained on the technical features or properties of components and on a plurality of sequences of previously selected items. The artificial intelligence modules can be trained on item features and historical clickstream data.

[0021] In a further possible embodiment of the recommendation engine according to the first aspect of the application, the first artificial intelligence module comprises a trained autoencoder.

[0022] In an alternative embodiment of the recommendation engine according to the first aspect of the application, the first artificial intelligence module comprises a tensor decomposition model.

[0023] Other artificial intelligence modules can be used, which comprise models capable of generating a latent representation of an item.

[0024] In a further possible embodiment of the recommendation engine according to the first aspect of the application, the second artificial intelligence module comprises a trained recurrent neural network.

[0025] In a further alternative embodiment of the recommendation engine according to the first aspect of the application, the second artificial intelligence module comprises a trained convolutional neural network.

[0026] According to a further aspect, the application further provides a computer- implemented method comprising the features of claim 11.

[0027] According to a second aspect, the application provides a computer-implemented method for automatically providing recommendations for completion of an engineering project, the method comprising the steps of:

[0028] computing, by a first artificial intelligence module, a latent representation of a set of items;

[0029] processing, by a second artificial intelligence module, the latent representation of the set of items to generate at least one sequence of complementary items required for completion of the set of items; and

[0030] outputting, via an interface, the at least one sequence of complementary items as a recommendation to complete the engineering project.

[0031] In a possible embodiment of the computer-implemented method according to the second aspect of the application, the one or more sequences of complementary items generated by the second artificial intelligence module are output on a screen of the user interface for selecting the next item from one of said sequences of complementary items, or for selecting one or more items from one of said sequences of complementary items (which items do not necessarily appear one after the other), or for selecting an entire sequence of complementary items by the user.

[0032] In a further possible embodiment of the computer-implemented method according to the second aspect of the application, the selection of one of said complementary items, or the selection of an entire sequence of complementary items, by the user via said user interface automatically triggers an ordering command in order to order the associated components for said engineering project.

[0033] This facilitates the supply of components required for said engineering project.

[0034] According to a further aspect, the application further provides a software tool comprising the features of claim 14.

[0035] According to this aspect, the application provides a software tool comprising program code executable to carry out the computer-implemented method according to the second aspect of the application.

[0036] According to a further aspect, the application further provides a platform comprising the features of claim 15.

[0037] According to this aspect, the application provides a platform comprising the recommendation engine according to the first aspect of the application.

[0038] The platform can comprise a cloud platform.

[0039] In the following, possible embodiments of the different aspects of the application are described in more detail with reference to the accompanying drawings.

[0040] Figure 1 a block diagram is shown of a possible exemplary embodiment of a system comprising the recommendation engine according to the first aspect of the application;

[0041] Figure 2 a possible sequence of complementary items output as recommendations by the recommendation engine as illustrated in Figure 1

[0042] Figure 3 a flow chart is shown of a possible exemplary embodiment of a computer-implemented method for automatically providing recommendations for the completion of an engineering project according to a further aspect of the application.

[0043] from Figure 1 ​As can be seen from the block diagram, the recommendation engine 1 according to the first aspect of the invention can form part of a system for configuring engineering schemes (such as automation systems). The automation system may include multiple hardware and / or software components. In the illustrated embodiment, as... Figure 1 The recommendation engine 1 illustrated in the diagram includes a first artificial intelligence module 1A and a second artificial intelligence module 1B. The first artificial intelligence module 1A is adapted to provide an implicit representation of a set of items. The second artificial intelligence module 1B is adapted to process the implicit representation of the set of items provided by the first artificial intelligence module 1A to generate at least one supplementary item sequence. This supplementary item sequence is needed to complete the set of items. To provide a complete item sequence, it can be output via interface 1C of the recommendation engine 1 to complete the corresponding engineering scheme. In a possible embodiment, these items are selected from a set of available items, where each available item corresponds to a hardware component and / or software component that can be used in the engineering scheme. Figure 1 In the embodiment illustrated, the configuration system includes a memory 2 connected to a user interface 3. The user interface 3 can be integrated into a user terminal or mobile user equipment. Figure 1 As shown, the collection of selected items is temporarily stored in memory 2. Recommendation engine 1 has access to... Figure 1 The memory 2 of the configuration system is shown. Figure 1 In the illustrated embodiment, user U selects three optional items I1, I2, and I3 one after another at times t1, t2, and t3 via user interface 3. Therefore, in this example, the set of selected items includes a sequence of item I1, followed by item I2, and then item I3. Each item I corresponds to a component of the engineering plan. Each item I may include one or more item features. In a possible embodiment, recommendation engine 1 has access to database 4 storing the item features of different items. Furthermore, recommendation engine 1 has access to another database 5, which stores multiple sequences of completed items. Each item I corresponds to a hardware component, such as a controller or display panel. Each component may include one or more features or attributes. For example, a controller may include power supply voltage, fail-safe compatibility, or its power consumption as technical features. A display panel may include power supply voltage and its screen resolution as technical features.

[0044] In a possible embodiment, the first artificial intelligence module 1A of the recommendation engine 1 may include a trained feature learning module adapted to: calculate as... Figure 1The implicit representation of the set of items I stored in the selection basket of the memory 2 as illustrated in Fig. 1. Furthermore, the second artificial intelligence module 1 B of the recommendation engine 1 can comprise a trained sequential model adapted to compute at least one sequence of complementary items to be output to the user interface 3 of the user U via the data interface 1 C of the recommendation engine 1. In a possible embodiment, the items I are selected by the user U via the user interface 3 having a screen adapted to output to the user the available items. The sequence(s) of complementary items generated by the second artificial intelligence module 1 B can be displayed on the screen of the user interface 3 for selecting the next item from one of the sequence of complementary items or for selecting the whole sequence of complementary items by the user U. Furthermore, it is possible for the user to select one or more items I from one of the sequence of complementary items without appearing one after the other.

[0045] The first artificial intelligence module 1 A and the second artificial intelligence module 1 B can comprise artificial neural networks trained on the technical features of the components and on a plurality of sequences of previously selected items. In a possible embodiment, the first artificial intelligence module 1 A can comprise a trained autoencoder. In an alternative embodiment, the first artificial intelligence module 1 A comprises a tensor decomposition model. In a preferred embodiment, the second artificial intelligence module 1 B comprises a trained recurrent neural network RNN. The recurrent neural network RNN is designed to exploit the temporal correlation between the selected items within the engineering project. Further artificial neural networks can be used for the second artificial intelligence module 1 B. In a possible embodiment, the second artificial intelligence module 1 B comprises a trained convolutional neural network.

[0046] As Figure 1 The recommendation system as illustrated in Fig. 1 comprises a recommendation engine 1 that can be used to complete an engineering project partially configured by the user U. The recommendation engine 1 reduces the time needed for the configuration of the completion of the engineering project. The recommendation engine 1 exploits the sequential nature of the configuration process of the engineering project. It can also exploit the complex relationships between the underlying components to generate project completion suggestions or recommendations. The information about the order in which the components or items have been introduced into the recommendation system 1 comprises contextual information that is used to complete the respective engineering project based on the historical examples of previously configured engineering projects for which the same information is available. The introduction of technical information about the components can additionally ensure their compatibility. Furthermore, this information can be used as a basis for the recommendations when not enough historical data is available. The recommendation engine 1 according to the present invention can suggest to the user U the next item to be added to its engineering project. The recommendation engine 1 can also be used to provide an automatic project completion, i.e. it can provide a sequence of items associated with the components still to be added in order to meet all the functional requirements of the engineering project. An automation mechanism can be installed to automatically complete the partially configured engineering project.

[0047] Figure 2 An example illustrating the operation of the recommendation engine 1 according to the present invention is shown. In this example, user U has performed a partial configuration of an engineering scheme (such as an automation system) by successively selecting three items I1, I2, and I3 from the available items. Each item I corresponds to an associated component C available in the engineering scheme. This partial configuration (i.e., the set of selected items I1, I2, and I3) can be stored in... Figure 1 In the memory 2 of the recommendation system illustrated in the diagram, the set of selected items I1, I2, and I3 is applied to the first artificial intelligence module 1A of the recommendation engine 1. The first artificial intelligence module 1A is adapted to provide an implicit representation of the set of selected items I1, I2, and I3. The implicit representation of the selected items can be automatically computed by the trained first artificial intelligence module 1A. The second artificial intelligence module 1B receives the computed implicit representation of the set of selected items I1, I2, and I3 provided by the first artificial intelligence module 1A. The second artificial intelligence module 1B can process the received implicit representation of the set of selected items provided by the first artificial intelligence module 1A to compute at least one supplementary item sequence required to complement the set of selected items I1, I2, and I3, thereby providing the complete item sequence required to complete the entire project solution. Figure 2 In the illustrated example, the second AI module 1B generates different sequences of supplementary items to complete the engineering plan. These different sequences of supplementary items form different completion scenarios CompSc, which can be output to user U as recommendations. Figure 2 In the example illustrated, the first completed scenario recommendation CompScl includes item I. 11 I 12 I 13 I 14 I 15 I 16 Therefore, the first supplementary item sequence includes six items I. 11 to I 16 User U can select these six items I 11 to I 16 This provides the complete project sequence required to complete a partially configured (p config) project. The second completion scenario, CompSc2, includes the following supplementary project sequence: This sequence includes project I... 21 I 22 I 23 I 24 In the illustrated example, the third completed scene, CompSc3, includes supplementary item I. 31 I 32 I 33of the project. Thus, after the third item I3 has been selected at time t3 to provide the partial configuration (p config) of the project, three different completion scenarios CompSc can be output to the user U via the user interface 3 upon request as possible recommendations to complete the project for the partial configuration. The user U can now select either a single item I from one of the recommended completion item sequences CompSc or select one of the completion item sequences CompSc entirely to complete the entire project. The different completion item sequences CompSc generated by the second artificial intelligence module IB can be displayed on the screen of the user interface 3 for selection by the user U. The user U can use a first type of selection command (e.g. a click on an item) to select the next single item I from one of the completion item sequences or can utilize a second type of selection command (e.g. a click on CompSc) to select the entire completion item sequence. For example, the user U can select the first completion scenario CompSc1 entirely to automatically select all remaining completion items I 11 to I 16 , thereby completing the configuration of the project. Alternatively, the user U can select the completion scenario CompSc2 comprising the four items I 21 , I 22 , I 23 , I 24 or the completion scenario CompSc3 comprising the items I 31 , I 32 , I 33 . Alternatively, the user U can select only a single item with the associated selection command to continue the selection process iteratively. The selection of a single item can be performed, for example, by a click on the displayed item. For example, the user U can select only the item I 31 from the third displayed completion item sequence CompSc3 by a click on the item I 31 . In the illustrated example of the project Figure 2 , the user U can then select in a next step the item I 32 to obtain a further completion scenario CompSc4 displayed on the screen of the user interface 3 comprising the items I 41 , I 42 , I 43 . In a further step, the user U can then select the entire completed scenario CompSc4 comprising the three items I 41 , I 42 , I 43 to complete the project or can select only the next single item of the third completion scenario CompSc3 (i.e. I 33), or selecting a new next single item of the completion scenario (i.e. a single I 41 ).

[0048] Figure 3 A flowchart showing a possible exemplary embodiment of a computer-implemented method according to a further aspect of the present application is shown. In the illustrated embodiment, the computer-implemented method is used for providing recommendations for the completion of an engineering project, in particular an automation system comprising a plurality of hardware and / or software components. In the illustrated embodiment, the computer-implemented method comprises three main steps.

[0049] In a first step S1, a latent representation of the set of items I is computed by a first artificial intelligence module 1A.

[0050] In a further step S2, the latent representation of the item I is processed by a second artificial intelligence module 1B to generate at least one sequence of complementary items required for completing the selected sequence of items.

[0051] In a further step S3, the at least one sequence of complementary items is output as a recommendation to complete the engineering project. In the example illustrated in Fig. Figure 2 , the four sequences of complementary items forming the different completion scenarios CompSc1 to CompSc4 can be output to the user U via the user interface 3 for further selection. In the illustrated example of Fig. Figure 2 , the first three completion scenarios CompSc1 to CompSc3 are displayed for selection for the partial configuration at time t3. After the user U has selected the item I 31 and the item I 32 , the fourth completion scenario CompSc4 is displayed.

[0052] The first artificial intelligence module 1A computes a latent representation for each selected item I, which comprises a vector v of different features of the associated component. The recommendation engine 1 of the recommendation system according to the present application has the advantage of having less dependency on manually defined rules. When provided with sufficiently rich contextual information and enough training examples, the recommendation system can discover more complex correlations between components than those that can be specified by domain experts. The performance of the recommendation system as illustrated in Fig. Figure 1 improves over time based on the collected training data.

[0053] A further advantage of the recommendation system according to the present application is that the system not only explicitly provides recommendations to the user U, but can also suggest to the user U how to complete the entire engineering project, instead of just selecting the next item or component. As illustrated in Fig. Figure 3The computer implementation method illustrated in the flowchart can be implemented in a software tool that includes executable program code to perform... Figure 3 The steps are shown in the diagram.

[0054] In possible embodiments, such as Figure 1 The recommendation engine 1 illustrated in the system diagram can be implemented as a desktop solution. In this embodiment, the computer implementation method can be executed by the processor of a user terminal. In an alternative embodiment, recommendation engine 1 can also be implemented on a web server of a cloud platform connected to multiple different user terminals via a data network. In this embodiment, the computer implementation method can be executed by one or more processors of the cloud platform server.

[0055] Further embodiments of the computer implementation method according to the present invention are possible. For example, in Figure 2 In the example, user U can also select items not suggested in the completion scenario CompSc (i.e., the supplementary item sequence). The system can then automatically check whether the selected item leads to the completion of the project. If the selected item leads to the completion of the project, the item can be accepted, and upon request, a new completion scenario CompSc can be calculated on an expanded partial configuration including the added item. In another exemplary embodiment, after selecting item I suggested by the calculated completion scenario CompSc, user U can also step back to the initial partial configuration, i.e., cancel the selected suggested supplementary item. For example, in Figure 2 In the example, item I has already been selected from the first completed scenario. 11 Afterwards, user U can delete the selection and fall back to the partial configuration including items I1, I2, and I3, and request a new calculation for the completed scenario and / or select another item, such as item I in the second completed scenario CompSc2. 21 After selecting the completion scenario CompSc, in one of the possible implementations, user U can confirm the selection to trigger an automatic ordering process for the associated components. For example, if item I has already been selected... 21 I 22 I 23 I 24 After completing scenario CompSc2, the system can ask the user if they want the project to be completed and trigger the selected item I in the selected completion scenario 2. 21 I 22 I 23 I 24Automatic ordering of associated physical components (hardware and / or software components). Thus, in a possible implementation, user U's selection of the next item or the entire sequence of supplementary items via user interface 3 can automatically trigger an ordering command to order associated components for the completion of the engineering project.

[0056] The recommendation system according to the present invention can employ the temporal correlation between selected items associated with the components of the engineering scheme. After user U has completed the set of items to complete the scheme, the contents of database 5, which includes multiple historical sequences of completed items, can be updated using the selected completion scenarios. Therefore, the performance of the recommendation system can improve over time as the number of completed schemes increases. The recommendation system can be used by one or more users U.

[0057] The recommendation engine 1 and method according to the present invention can be used in a wide range of different applications and use cases, and are not limited to these. Figure 1 and Figure 2 The embodiment is illustrated in the figure. The recommendation engine 1 and computer implementation method according to the present invention can be used in any complex system that requires configuration of its functional components. The computer implementation method can be implemented in software tools such as TIA Selection Tool, TIA Portal, or NX Designer. The present invention provides a sequential recommendation system for completing any kind of complex engineering scheme.

Claims

1. A recommendation engine for automatically providing recommendations for the completion of engineering solutions, the recommendation engine comprising: The first artificial intelligence (AI) module (1A) is adapted to provide an implicit representation of a collection of items; as well as The second artificial intelligence (AI) module (1B) is adapted to: Process the implicit representation of the item set provided by the first artificial intelligence (AI) module (1A) to generate at least one supplementary item sequence required to supplement the item set to provide a complete item sequence, and The at least one supplementary item sequence is output via interface (IC) as a recommendation for completing the engineering solution, allowing the user to select the entire supplementary item sequence. The second artificial intelligence (AI) module (1B) includes a trained sequential model adapted to compute the at least one supplementary item sequence, and The recommendation engine is adapted to automatically complete the engineering scheme using the selected sequence of supplementary items.

2. The recommendation engine of claim 1, wherein the items are selected from a set of available items corresponding to hardware components and / or software components that can be used in the engineering scheme.

3. The recommendation engine according to claim 2, wherein the set of selected items is at least temporarily stored in a memory (2) connected to the recommendation engine (1).

4. The recommendation engine according to any one of claims 1 to 3, wherein the first artificial intelligence (AI) module (1A) includes a trained feature learning module adapted to compute a latent representation of the item set.

5. The recommendation engine according to claim 2 or 3, wherein the items are selected by the user via a user interface (3), the user interface (3) having a screen adapted to output available items to the user.

6. The recommendation engine according to any one of claims 1 to 3, wherein the first artificial intelligence (AI) module (1A) and the second artificial intelligence (AI) module (1B) include artificial neural networks trained on the technical features of the components and multiple sequences of previously selected items.

7. The recommendation engine according to any one of claims 1 to 3, wherein the first artificial intelligence (AI) module (1A) comprises a trained autoencoder or tensor decomposition model.

8. The recommendation engine according to any one of claims 1 to 3, wherein the second artificial intelligence (AI) module (1B) comprises a trained recurrent neural network or a trained convolutional neural network.

9. A computer-implemented method for automatically providing recommendations for the completion of an engineering project, the method comprising the following steps: The implicit representation of the item set (S1) is computed by the first artificial intelligence (AI) module (1A); The implicit representation of the item set is processed (S2) by a second artificial intelligence (AI) module (1B) to generate at least one supplementary item sequence required to complete the sequence of selected items, wherein the second artificial intelligence (AI) module (1B) includes a trained sequence model adapted to compute the at least one supplementary item sequence. The at least one supplementary item sequence is output (S3) via interface (1C) as a recommendation for completing the engineering scheme, allowing the user to select the entire supplementary item sequence; and The engineering plan is completed automatically using the selected sequence of supplementary items.

10. A software tool comprising program code, said program code being executable to perform the computer implementation method according to claim 9.

11. A platform comprising a recommendation engine according to any one of claims 1 to 8.

Images