Dynamically deploy MOM modules across multiple tiers

Abstract

The present invention provides a method and system for dynamically deploying modules of a MOM application across multiple layers, comprising: providing a MOM application including multiple modules (21; 31, 32) organized in a layered architecture, wherein each module can be independently deployed in at least two layers of the multiple layers (11, 12, 13); for each module and for each layer (11, 12, 13) on which the module (21; 31, 32) can be deployed, defining a set of production parameters related to a layer (11, 12, 13) recommended for module deployment; defining a layer deployment function for each module, the layer deployment function having as input variables the values ​​of the production parameter set for the module and having as output variables the recommended layer (11, 12, 13) on which the module (21; 31, 32) is to be deployed; and, upon request, for a specific module (21; 31, 32) and for a specific point in time, using the values ​​of the input variables at the requested specific point in time, calculating (42) the recommended layer (11, 12, 13) for deployment using the layer deployment function.

Description

[0001] The present invention relates to a method for dynamically deploying modules of a MOM / MES application across multiple layers. Recently, the term MOM (Manufacturing Operations Management) is increasingly used to replace the term MES (Manufacturing Execution System). MES / MOM applications and systems can generally include software that acts as a component, or acts as a component in the sense explained below.

[0002] As defined by the Manufacturing Enterprise Solutions Association (MESA International), an MES / MOM system “is a dynamic information system that drives the efficient execution of manufacturing operations by managing “production operations from the point an order is released to manufacturing to the point the product is delivered as a finished product” and by providing “mission-critical information about production activities to other parties in the organization and the supply chain via two-way communications”.

[0003] In order to improve the quality and process performance of manufacturing plants, MES / MOM systems usually include functions such as resource allocation and status, dispatching production orders, data collection / acquisition, quality management, maintenance management, performance analysis, operation / detail scheduling, document control, labor management, process management and product tracking.

[0004] For example, Siemens AG A wide range of MES / MOM products are available under the IT product line.

[0005] Today, the typical MOM system consists of a "monolithic" application, or, in better cases, a group of applications that need to be integrated to provide a complete set of MOM functions required by the target industry.

[0006] MOM systems are typically deployed and used locally, such as in a data center or dedicated server at a production site, and require a dedicated IT team to manage the software and hardware infrastructure.

[0007] Manufacturing companies require MOM systems with the capabilities to adapt to the latest manufacturing trends.

[0008] The first manufacturing trend is the “globalization” of manufacturing companies and the resulting wider geographic distribution of their manufacturing facilities.

[0009] The second manufacturing trend is the demand for “mass customization” and the resulting need for new production models focused on smaller production batches and flexible production volumes.

[0010] These and other manufacturing trends, as well as new challenges and innovative technologies driven by global initiatives such as Industry 4.0, are calling for a new generation of MOM systems that can enable a more flexible and distributed production paradigm.

[0011] In fact, currently, for a specific MOM project, a complex engineering design phase is required to find the most suitable distribution of MOM functions. In addition, if the production process changes, the specific MOM system will not adapt to these changes with the required flexibility.

[0012] Figure 1 is a diagram schematically illustrating a typical physical layer pyramid for manufacturing IT systems.

[0013] The physical layer pyramid includes three physical layers: the field layer 11, the intra-field layer 12 and the extra-field layer 13.

[0014] From an IT perspective, manufacturing companies typically adopt such a three physical layer 11, 12, 13 approach for the distribution of production execution software, such as MOM / MES systems, quality systems, planning systems, SCADA (Supervisory Control and Data Acquisition) systems and other production execution applications.

[0015] Operations in a public cloud are typically performed on an off-premises layer 13 .

[0016] Operations that require a mix of dedicated servers, industrial edge devices, and hybrid or private clouds are typically performed on the on-premises layer 12 .

[0017] The field layer 11 includes HMI (Human Machine Interface) devices, such as panels and industrial edge devices, such as Microbox or Nanobox, SCADA, automation components of equipment, PLC (Programmable Logic Controller).

[0018] Thus, operations that are critical to the facility and that are network dependent are typically performed at layer 11. Non-critical operations, such as printing reports, are typically performed at layer 13.

[0019] The examples provided above are for illustration only, and those skilled in the art will readily appreciate that other deployments of operations and devices are possible for these three layers, and other different layers may be employed.

[0020] Due to the monolithic structure of most MOM systems, the commonly used current deployment methods are mainly a compromise between centralized and distributed. Centralized methods are usually used for enterprise MOM functions, such as overall scheduling. In contrast, distributed methods are usually used for MOM functions related to the shop floor, such as production order execution or data collection.

[0021] It is therefore an object of the present invention to overcome the above-mentioned disadvantages by providing a method and system for deploying modules of a MOM application across multiple tiers in a dynamic manner.

[0022] The aforementioned object is achieved by a method and system for dynamically deploying modules of a MOM application across multiple layers, the method and system comprising the following contents:

[0023] a) providing a MOM / MES application comprising a plurality of modules organized in a layered architecture, wherein each module can be independently deployed in at least two layers of the plurality of layers;

[0024] b) for each module and for each layer on which the module can be deployed, defining a set of production parameters associated with the layer recommended for module deployment;

[0025] c) defining for each module a layer deployment function having as input variables the values ​​of the production parameter set for the module and having as output variables the recommended layers at which to deploy the module;

[0026] d) Upon request, for a specific module and for a specific time point, the recommended deployed layers are calculated using the layer deployment function by using the values ​​of the input variables at the specific time point requested.

[0027] In an embodiment of the present invention, at the time of selection, if the recommended layer is different from the actually deployed layer of the specific module, the specific module may be moved to the recommended layer.

[0028] In an embodiment of the present invention, the layer deployment function optimizes a performance parameter selected from a cost parameter, a production quality parameter, a production speed parameter, an environmental impact parameter and an energy consumption parameter.

[0029] Furthermore, a computer program element may be provided, comprising computer program code for performing the steps according to the above method when loaded in a digital processor of a computing device.

[0030] Additionally, a computer program product stored on a computer usable medium may be provided, the computer program product comprising a computer readable program code for causing a computing device to perform the method mentioned.

[0031] Embodiments of the present invention enable the core functionality of a MOM application to be distributed over multiple physical layers in a flexible manner in an optimized manner.

[0032] Embodiments of the present invention provide a highly flexible, scalable and / or dynamic approach.

[0033] Embodiments of the present invention allow for automatic deployment of MOM functional modules at different manufacturing levels in order to optimize production management.

[0034] Embodiments of the present invention enable the selection of the most efficient deployment level.

[0035] Embodiments of the present invention enable support for the global “as a service” transformation trend consistent with the digital factory cloud strategy.

[0036] Embodiments of the present invention allow MOM functionality to be flexibly distributed horizontally and vertically across multiple layers without requiring MOM project re-engineering.

[0037] Embodiments of the present invention allow modules to be deployed across multiple physical infrastructures.

[0038] Embodiments of the present invention allow for a time-varying distribution of MOM modules.

[0039] The present invention will now be described in preferred but non-exclusive embodiments with reference to the accompanying drawings, in which:

[0040] Figure 1 is a diagram schematically illustrating a typical physical layer pyramid for manufacturing IT systems.

[0041] Figure 2 is a block diagram schematically illustrating a modular architecture of a MOM application according to a first exemplary embodiment of the present invention.

[0042] Figure 3 is a block diagram schematically illustrating a modular architecture of a MOM application according to a second exemplary embodiment of the present invention.

[0043] Figure 4 is a flow chart that schematically illustrates an example implementation of dynamically deploying one or more modules across three tiers.

[0044] At least some embodiments of the present invention address the above-mentioned problems in which modules of a MOM application are deployed across multiple tiers.

[0045] A MOM application is provided, the MOM application comprising a plurality of modules organized in a layered architecture. Each module can be independently deployed in at least two layers of the plurality of layers. Figure 2 21, wherein each parent module has direct interaction with a subset of related child modules. The independent modules can advantageously group one or more MOM functions. The modules are defined so that they can conveniently support the horizontal and vertical distribution of MOM functions in a flexible manner.

[0046] In an embodiment, the appropriate distribution of MOM functions for a particular MOM project is statically defined at engineering time.

[0047] In an embodiment, the MOM functions required for the MOM solution are appropriately distributed across the tiers at engineering time.

[0048] For each module and for each tier in which the module can be deployed, a set of relevant production parameters is defined that recommends the tier in which the module is deployed.

[0049] Examples of relevant production parameters for tier recommendation purposes include, but are not limited to, the following: production volume, production variability, number of product variants, and batch size.

[0050] For each module, a layer deployment function is defined that takes as input variables the values ​​of the module and the production parameter set of the relevant layers, and as output variables the recommended layers at which to deploy the module.

[0051] The layer deployment function can be expressed as the following deployment formula:

[0052] DeployLocation[Mod j ,Lev i ]=f(ProdPar 1 [Lev i ,Mod j ], ProdPar 2 [Lev i , Mod j ], .., ProdPar N [Lev i , Mod j ])

[0053] in:

[0054] -Mod j is a specific MOM module "j"

[0055] -Lev i Indicates the deployment layer of the module. For example, Lev 0 It can be the field level, Lev 1 Can be an inner layer of the field, and Lev 2 Can be an off-site layer.

[0056] - function f for each production parameter ProdPar associated with MOM module "j" when MOM module "j" is deployed on level "i" K to provide the corresponding weights.

[0057] Given a set of specific production parameters ProdPar1, ..., ProdParN considered relevant to a specific production process, the deployment formula results in i Deployment Module Mod j Provides a recommendation score value. Thus, for each module, the best recommended layer to deploy depends on the production parameters associated with the modules in each layer.

[0058] Upon request, for a specific module and for a specific point in time, the recommended deployed layers are calculated using the layer deployment function by using the values ​​of the input variables at the specific time point requested.

[0059] When selected, if the recommended layer is different from the actually deployed layer of the specific module, the specific module can be moved to the recommended layer.

[0060] Advantageously, using embodiments of the present invention, the deployment layer of functional modules can be switched in a flexible and dynamic manner.

[0061] Conveniently, the modular architecture enables the distribution of MOM functionality across multiple tiers.

[0062] For each MOM function, the corresponding deployment tier can be advantageously determined by considering the relevant production requirements of the specific MOM function. For example, for the production order execution function, low data latency is important, while for the analysis function, data storage scalability is important.

[0063] In an embodiment, the formulation of the layer deployment function may optimize one or more of the following performance parameters: cost, production quality, production speed, environmental impact, energy consumption, or any other production performance parameter.

[0064] Simple example implementation

[0065] Figure 3 is a block diagram schematically illustrating a modular architecture of a MOM application according to a second exemplary embodiment of the present invention.

[0066] For the purpose of explanation, it is assumed that the MOM application includes two independent and hierarchically connected MOM modules 31, 32, which are respectively referred to as a production order management ("POM") module 31 and a production order execution ("POE") module 32.

[0067] The functions of the POM module 31 include checking whether a product can be produced, creating a corresponding production order, optionally dividing the created order into sub-orders, and dispatching them to the POE module 32 .

[0068] The functions of the POE module 32 include executing the operations defined in the production order and sending "as built data" back to the POM module 31. Typically, the as built data includes all production data generated during the execution of the production order (e.g., start and end times, consumed materials, etc.).

[0069] The following production parameters are considered relevant to the two modules 31 , 32 : the production volume VOL, which represents the quantity of the final product to be produced, and the product variability VAR, which represents the number of product variants.

[0070] In this simple example implementation, the layer deployment function aims to optimize computational cost.

[0071] The tier deployment formula takes into account all relevant computational costs of the production parameters VOL, VAR and the variables used to manage the production parameters VOL, VAR. Examples of computational costs may include, but are not limited to: network workload, CPU power, storage requirements, and others.

[0072] The matrix in Table 1 shows the normalized calculated cost calculation of the POM module 31 in relation to its production parameters VOL, VAR.

[0073] For example, in this simple example, increased production volume VOL or even increased production variability VAR requires more computing resources, for example to effectively support increased scheduling tasks, which would then be better managed in layer L2 using a cloud-based system.

[0074] Therefore, in the case of high variability (high VAR), the POM module is better deployed in the cloud to provide more computing power for its scheduling calculations.

[0075] Low VOL Medium VOL High VOL Low VAR 0.10 0.20 0.40 VAR 0.35 0.60 0.65 High VAR 0.45 0.70 1.00

[0076] Table 1: Normalized computational cost

[0077] The values ​​in the cells of Table 1 indicate the normalized computational costs of the production parameter quantity VOL and variability VAR for a specific range (low, medium, high) of the module POM.

[0078] The results of the layer deployment function for optimizing computational cost are shown in exemplary Table 2, where the values ​​detailed in Table 1 are calculated based on the values ​​for a specific POM module Mod. j Deployment-level recommended Lev i is discretized.

[0079]

[0080] Table 2 - Recommended Deployment Levels

[0081] Similarly, a corresponding table of POE modules may be generated to determine the recommended deployment level.

[0082] Figure 4 is a flow chart that schematically illustrates an example implementation of dynamically deploying one or more modules across three tiers.

[0083] At step 41, a module Mod corresponding to a given MOM function is selected. j .

[0084] At step 42, for a specific time point, use the i Mod j The deployment formula is calculated based on the given parameter values.

[0085] At step 43, the deployment formula provides three layers Lev 1 ,Lev 2 and Lev 0 Mod to be deployed j The recommended layer is given as a result.

[0086] If the recommended layer is different from the actually deployed layer, then the module Mod is executed at steps 44, 45, and 46. j All necessary steps to move to the recommended tier.

[0087] Examples of such necessary switching steps include, but are not limited to, automatic operation (eg, automatic shutdown of a MOM module and restart at a different stage), requesting manual action from a plant supervisor, a mix of automatic and manual action.

[0088] In an embodiment, steps 41 to 46 may be conveniently repeated for each module. In other embodiments, steps 41 to 46 may be conveniently repeated for different time points or when notified, such as when certain production parameters or user requests are changed.

Claims

1. A method for dynamically deploying a module of a MOM application across multiple layers, the method comprising the following steps: a) providing a MOM / MES application comprising a plurality of modules (21; 31, 32) organized in a layered architecture, wherein each module can be independently deployed in at least two layers of the plurality of layers (11, 12, 13); b) defining, for each module and for each layer (11, 12, 13) on which said module (21; 31, 32) can be deployed, a set of production parameters associated with the layer (11, 12, 13) recommended for the deployment of the module; c) defining, for each module, a layer deployment function having as input variables the values ​​of the production parameter set for the module and having as output variables the recommended layers (11, 12, 13) at which the module (21; 31, 32) is to be deployed; d) upon request, for a specific module (21; 31, 32) and for a specific point in time, using the values ​​of the input variables at the requested specific point in time, calculating (42) the recommended layers (11, 12, 13) at which the module (21; 31, 32) is to be deployed using the layer deployment function; and f) upon selection, if the recommended layer is different from the actually deployed layer of the specific module, the specific module is moved to the recommended layer (44, 45, 46).

2. The method according to claim 1, wherein: The formula of the layer deployment function optimizes a performance parameter selected from the group consisting of: - cost parameters; -Production quality parameters; -Production speed parameters; - Environmental impact parameters; -Energy consumption.

3. A system for dynamically deploying modules of a MOM application across multiple layers, the system comprising: a) a MOM / MES application comprising a plurality of modules organized in a layered architecture, wherein each module can be independently deployed in at least two layers of the plurality of layers; b) for each module and for each layer on which said module can be deployed, means for defining a set of production parameters associated with the layer recommended for deployment of the module; c) means for defining, for each module, a layer deployment function having as input variables the values ​​of the production parameter set for said module and having as output variables the recommended layers at which said module is to be deployed; d) means for calculating, upon request, for a specific module and for a specific point in time, using the values ​​of the input variables at the requested specific point in time, using the layer deployment function to calculate the recommended layers at which the module is to be deployed; and f) means for moving the specific module to the recommended layer if the recommended layer is different from the actually deployed layer of the specific module when selected.

4. The system according to claim 3, wherein: The formula of the layer deployment function optimizes a performance parameter selected from the group consisting of: - cost parameters; -Production quality parameters; -Production speed parameters; - Environmental impact parameters; -Energy consumption.

5. A computer program product for executing the steps of the method according to any one of claims 1 and 2.

Images