DE102019100450A1 - Method and device for planning an assembly and / or manufacturing process - Google Patents

Abstract

A method (300) for planning an assembly and / or manufacturing process for a product is described. The method (300) comprises providing (301) a process description (100) of the assembly and / or manufacturing process with one or more process modules (101), a process module (101) comprising one or more action elements (103) and wherein an action element (103) is described by basic data (111) and by detailed data (112, 113, 114). The method 300 further comprises determining (302), on the basis of the basic data (111) an action element (103) of the process description (100), a basic code (221) of a process description language. In addition, the method (300) comprises adapting (303) the basic code (221) as a function of the detail data (112, 113, 114) in order to provide an action code (120) for the action element (103).

Description

The invention relates to a (computer-implemented) method and a corresponding (computer-based) device for the automated determination of an assembly and / or production sequence for the manufacture of a product, in particular a motor vehicle.

As part of the development of a vehicle, in particular a road vehicle, the planning and construction of an assembly line for the assembly of the newly developed vehicle are also carried out. In this context, it is desirable to determine processes for assembling and / or manufacturing the vehicle at an early stage of development. The assembly and / or manufacturing processes can be used to build a suitable assembly line. Furthermore, time values for individual assembly steps of a vehicle can be determined on the basis of an assembly and / or production sequence, wherein the time values can be taken into account when designing an assembly line, in particular in order to reduce the costs for the manufacture of a vehicle. Furthermore, decisions for an assembly-efficient construction of a vehicle can be made on the basis of precise time values for the assembly of a vehicle.

The present document deals with the technical problem of providing an apparatus and a method which make it possible to plan an assembly or manufacturing process for a product, in particular for a vehicle, in an efficient and reliable manner.

The task is solved in each case by one of the independent claims. Advantageous embodiments include described in the dependent claims. It is pointed out that additional features of a claim dependent on an independent claim without the features of the independent claim or only in combination with a subset of the features of the independent claim can form a separate invention and independent of the combination of all features of the independent claim can be made the subject of an independent claim, a divisional application or a late application. This applies in the same way to the technical teachings described in the description, which can form an invention which is independent of the features of the independent claims.

According to one aspect, a computer-implemented method for planning an assembly and / or manufacturing process for a product is described. The method comprises the provision of a process description of the assembly and / or manufacturing process with one or more process modules, a process module comprising one or more action elements, and an action element being described by basic data and detailed data. The basic data shows a component with which an action is carried out as part of the action element. The basic data also shows where the action is carried out on the component. The detail data indicates one or more additional parameters relating to the action that is carried out as part of the action element.

Furthermore, the method comprises determining, based on the basic data of an action element of the process description, a basic code in a process description language for describing the action of the action element. In addition, the method comprises adapting the basic code depending on the detail data in order to provide an action code for the action element.

According to a further aspect, a software (SW) program is described. The SW program can be set up to run on a processor and thereby perform the method described in this document.

According to a further aspect, a storage medium is described. The storage medium can comprise a software program which is set up to be executed on a processor and thereby to carry out the method described in this document.

According to a further aspect, a device (e.g. with at least one server) is described which is set up to carry out the method described in this document.

It should be noted that the methods, devices and systems described in this document can be used both alone and in combination with other methods, devices and systems described in this document. Furthermore, any aspect of the methods, devices and systems described in this document can be combined with one another in a variety of ways. In particular, the features of the claims can be combined with one another in a variety of ways.

• 1a beispielhafte Prozess-Bausteine eines Montage- und/oder Fertigungsprozesses;
• 1b beispielhafte Aktionselemente zur Beschreibung eines Prozess-Bausteins;
• 2a ein beispielhaftes System zur automatischen Generierung eines Montage- und/oder Fertigungsprozesses;
• 2b eine sequentielle bzw. iterative Ermittlung eines Aktions-Kodes für einen Prozess-Baustein; und
• 3 ein Ablaufdiagramm eines beispielhaften Verfahrens zur Planung eines Montage- und/oder Fertigungsprozesses.

The invention is described in more detail below on the basis of exemplary embodiments. Show

• 1a exemplary process components of an assembly and / or manufacturing process;
• 1b exemplary action elements for the description of a process module;
• 2a an exemplary system for the automatic generation of an assembly and / or manufacturing process;
• 2 B a sequential or iterative determination of an action code for a process module; and
• 3rd a flow chart of an exemplary method for planning an assembly and / or manufacturing process.

As stated at the beginning, this document deals with the reliable, precise and automated planning of an assembly and / or manufacturing process. Such a process 100 can, as in 1a shown, in a variety of (assembly or manufacturing) steps 101 be divided. Every step can be done 101 as a building block of a process description 100 of a process. For each of these building blocks 101 Data can be stored in a storage unit, the data being a time value for the execution of the module 101 Show. The building blocks or steps 101 A planned process can also be called advanced stages. Exemplary building blocks 101 one process are: gluing an adhesive part; screwing a screw part; clipping a part; applying a medium; Etc.

To provide a time value (ie an expected time period) for a block 101 In a process, one or more methods can be considered, which are used to carry out the module or step 101 can be used. For example, a block or step 101 can be realized by four different methods. Each of the different methods can contribute to a certain extent in the implementation of the module 101 be used. Furthermore, each method can be associated with a certain time value, which indicates the time required to carry out the respective method. The time value of the block 101 can then be determined as a weighted average of the time values of the methods by which the block 101 can be implemented. The usage shares of the different methods can be used as weights.

The current value of a method can be determined using a "methods-time measurement" (MTM) procedure. To this end, the method can be broken down into a variety of action items 103 can be divided and it can be used for any action item 103 a time value can be determined (e.g. by measurement). The individual action elements 103 can be described by a respective code (or code). Exemplary action items 103 are: stretching, aligning, gripping, moving, positioning, letting go, etc. The time value of the method then results from the sum of the time values for the individual action elements 103 the method. A code can contain several letters and / or numbers.

Thus, a process building block 101 into a plurality of action items 103 be subdivided (if necessary using several possible methods). The individual action elements 103 can each be associated with time values, which then (if necessary taking into account different possible methods) at a time value for the process module 101 can be agglomerated.

• • welches Bauteil (wie z.B. eine Schraube, eine Zündspule, ein Karosserieteil, etc.) bewegt bzw. verarbeitet wird; und
• • örtliche Information, in Bezug auf eine Bewegung bzw. eine Verarbeitung des Bauteils.

An action item 103 includes, as in 1b shown, typically basic data 111 that display

• • which component (such as a screw, an ignition coil, a body part, etc.) is moved or processed; and
• • Local information regarding a movement or processing of the component.

For example, the basic data 111 for a basic element "pick up and place" indicate which component has to be picked up and where the component has to be picked up and / or placed.

• • die Anzahl 112 von Bauteilen anzeigen, die bewegt und/oder verarbeitet werden soll;
• • das Gewicht 113 eines Bauteils anzeigen, das bewegt und/oder verarbeitet werden soll;
• • die Entfernung 114 anzeigen, die im Rahmen des Aktionselements 101 von einem Monteur zurückgelegt werden muss; und/oder
• • die räumlichen Maße eines Bauteils anzeigen, das bewegt und/oder verarbeitet werden soll.

Furthermore, an action element 103 Detailed data 112 , 113 , 114 include, for example

• • the number 112 display of components that are to be moved and / or processed;
• • the weight 113 display a component that is to be moved and / or processed;
• • the distance 114 view that as part of the promotional item 101 must be covered by a mechanic; and or
• • display the spatial dimensions of a component that is to be moved and / or processed.

• • den Typ des Aktionselements 103 (z.B. „Aufnehmen und Platzieren“, „Platzieren“, „Betätigen“, „Durchführen einer Körperbewegung“, „Lesen“, „Visuelle Kontrolle“, „Verwenden eines Hilfsmittels“, etc.); und/oder
• • einen Zeitwert, der die erforderliche Zeit zur Ausführung des Aktionselements 103 anzeigt.

The basic data 111 and / or the detailed data 112 , 113 , 114 can be used by a human planner considering a set of rules to create an action code 120 for the action item 101 to determine, with the action code 120 the time value for the execution of the action element 103 displays. In particular, the promotion code 120 for an action item 103 Show,

• • the type of action item 103 (eg "picking up and placing", "placing", "actuating", "performing a body movement", "reading", "visual control", "using an aid", etc.); and or
• • a time value representing the time required to execute the action item 103 displays.

A process description can thus be used manually to produce a product 100 a process with a multitude of process steps 101 are created, the individual process steps 101 into individual action elements 103 can be divided into each action item 103 through an action code 120 can be described. The work processes for the manufacture of a product are thus typically planned manually by a planner and are then analyzed manually using a process description language, such as MTM, in order to determine a current value for the planned work process.

Manual planning is relatively expensive and has a relatively high susceptibility to errors.

To reduce the effort and susceptibility to errors, planning can be automated based on a standardized process description language using artificial intelligence (AI) to determine an efficient workflow.

This in 2nd AI-based system shown 200 receives an electronic description as input 201 (eg a CAD drawing or a CAD model) of a component or an assembly (generally a product). Furthermore, the system 200 an electronic description as input 202 take into account the construction of the manufacturing facility in which the product is to be manufactured. In particular, the electronic description 202 show where the individual parts or components for producing a product are located relative to each other.

The AI-based system 200 can have been trained on the basis of training data with a large number of training data records. On the one hand, a training data record shows a product (in particular, a training data record can be an electronic description 201 a product and an electronic description 202 of the construction of the production facility), and on the other hand the training data record shows the workflow for the product described in the standardized process description language. The workflow displayed in a training data record may have been planned manually by a planner.

By teaching an AI-based system 200 (which includes, for example, one or more neural networks) based on (manually created) training data, the system can 200 be empowered to automate a workflow 100 to create.

This automatically derived workflow 100 can be checked by a planner and confirmed if necessary.

The automatically derived workflow 100 can be adjusted if necessary (e.g. if there is actual data relating to the actual manufacture of the product). The process description of the adapted workflow can then be used as a further training data set for adapting the AI-based system 200 be used. So the system can 200 be continuously improved.

• • den Typ des Aktionselements 103 an; und
• • einen Default-Zeitwert zur Ausführung des Aktionselements 103 an.

Alternatively or additionally, the determination of action codes 120 for the one or more action items 103 of the individual process components 101 a process description 100 of a process can be determined automatically, in particular using AI. For this purpose, in a first step based on the basic data 111 an action item 103 a basic code 221 be determined. The reason code 221 shows

• • the type of action item 103 on; and
• • a default time value for the execution of the action element 103 on.

In one or more further steps, the one or more detailed data can then 112 , 113 , 114 of the action item 103 be taken into account to the basic code 221 gradually adjust, and one or more intermediate codes 222 , 223 and finally the promotion code 120 for the action item 103 to investigate. In the individual steps, the time value can also be adjusted, in particular increased or possibly reduced, based on the default time value.

An action item 103 for a particularly complex action, for example for an action in which a particularly large distance has to be covered, can possibly be split up and / or by means of several action codes 120 to be discribed. For example, covering a particularly large distance can result in an additional action code 120 for an action item 103 "Body movement" is introduced.

It is thus a cascaded or iterative determination of action codes 120 for promotional items 103 described. The cascaded determination can be made using AI-based systems 211 , 212 , 213 , 214 automated for the individual steps. The AI-based systems 211 , 212 , 213 , 214 can each be taught in using training data with a large number of training data records. A training data set for teaching the system shows 211 actual basic data 111 and an actual reason code 221 (which was determined manually by a planner, for example). A training data set for teaching a system 212 can change the code 221 , the detail data relevant for the adjustment 112 , and the code actually adjusted (eg by a planner) 222 include. This applies accordingly to the systems 213 , 214 .

In this way, AI-based systems can be taught and provided, the AI-based system 211 is set up from the basic data 111 for an action item 103 automates a basic code 221 to investigate. The AI-based system 221 is set up the basic code 221 automated taking into account the detailed data 112 to change to an intermediate code 222 to investigate. Through the system 214 can finally the final action code 120 for the action item 103 to be provided.

3rd shows a flowchart of an exemplary computer-implemented method 300 for planning an assembly and / or manufacturing process for a product, for example for a motor vehicle. The procedure 300 can be executed by one or more servers or computers.

The procedure 300 includes providing 301 a process description 100 of the assembly and / or manufacturing process, the process description 100 one or more process modules 101 includes. The process description 100 can, for example, be read in via a data interface. As an alternative or in addition, the process description 100 (as related to 2a described) on the basis of an electronic description 201 of the product (e.g. based on a CAD drawing or based on a CAD model) using a machine-learned system 200 be determined.

A process building block 101 can have one or more action items 103 include. Each action element describes 103 one as part of the process building block 101 or the process to be carried out. The action of an action item 103 can have one action type from a variety of action types. Exemplary action types are: picking up a component, placing a component, performing a body movement, actuating a control element, performing a visual check, performing a reading activity, and / or using an aid.

An action item 103 can by basic data 111 and through detailed data 112 , 113 , 114 to be discribed. The basic data 111 display a component with which as part of the action element 103 an action is performed. Furthermore, the basic data 111 show where the action is performed on the component.

The detail data 112 , 113 , 114 can display one or more additional parameters related to the action that are part of the action item 103 is performed. Exemplary further parameters are: the weight of a component; the extent of a movement that is carried out as part of the action; the size of a component and / or the number of components that are processed as part of the action.

The procedure 300 includes identifying 302 , based on the basic data 111 , an action item 103 the process description 100 of a basic code 221 in a process description language. The reason code 221 can the action of the action item 103 describe. When determining the basic code 221 can the detailed data 112 , 113 , 114 to be disregarded. The reason code 221 can, for example, display the type of action from the plurality of predefined action types. Furthermore, the basic code 221 display a time value related to the amount of time required to complete the action. You can do this by using the basic code 221 displayed time value the detailed data 112 , 113 , 114 be disregarded.

The one for a code 221 , 222 , 223 , 120 Process description language used can be a standardized language for describing work processes. In particular, the process description language can include or be MTM-1, MTM-UAS, MTM-HWD and / or BMW SD.

The procedure also includes 300 the customization 303 of the basic code 221 depending on the detailed data 112 , 113 , 114 to an action code 120 for the action item 103 to provide. The promotion code 120 can be the type of action of the action item 103 display from the majority of predefined action types. Furthermore, the promotion code 120 display the time value relative to the amount of time it takes to complete the action. You can use the promotion code 120 both the basic data 111 as well as the detailed data 112 , 113 , 114 have been taken into account.

It can thus be an iterative determination of the action codes 120 for the one or more action items 103 a process description 100 respectively. The iterative determination enables a precise and reliable automated determination of action codes 120 . In particular, the basic code 221 using a machine-learned system 211 be determined. Alternatively or in addition, the promotion code 120 using a machine-learned system 212 , 213 , 214 be determined. The systems 211 , 212 , 213 , 214 may have been learned on the basis of training data, the training data being at least partially created by human planners. In particular, the machine-learned systems 211 , 212 , 213 , 214 each have been trained on the basis of training data with a large number of training data records, the individual training data records each having a code 221 , 120 (for example, a basic code, an intermediate code or an action code) by a human planner for the respective action element 103 was determined. In this way, the know-how of human planners and / or a set of rules for creating action codes can be transferred to a computer-implemented system in an efficient and reliable manner.

The detail data 112 , 113 , 114 an action item 103 can use several parameters to describe the action of the action element 103 include. It can then be part of the process 300 an iterative adjustment of the code 221 , 222 , 223 for the action item 103 depending on a single one of the parameters take place gradually or iteratively the action code 120 for the action item 103 to provide, which then all parameters of the detailed data 112 , 113 , 114 considered. A machine-learned system can be used in an efficient and precise manner for each of the iterative steps 211 , 212 , 213 , 214 learned and used.

As part of the process 300 can be based on the determined action codes 120 for the one or more action items 103 the process description 100 of the assembly and / or manufacturing process, an assembly line for the assembly and / or manufacturing process for manufacturing the product, in particular for manufacturing a large number of products, is provided. Alternatively, or in addition, can be used as part of the process 300 depending on the promotion codes 120 for the one or more action items 103 the process description 100 of the assembly and / or manufacturing process one or more products are manufactured. This can reduce the cost of manufacturing products.

Furthermore, a device for planning an assembly and / or manufacturing process for a product is described in this document. The device is set up, a process description 100 the assembly and / or manufacturing process with one or more process components 101 to provide. A process building block 101 includes one or more action items 103 , being an action item 103 through basic data 111 and through detailed data 112 , 113 , 114 is described. The basic data 111 display a component with which in the context of the action element 103 an action is performed. Furthermore, the basic data show 111 where the action is performed on the component. The detail data indicates one or more additional parameters relating to the action that is carried out as part of the action element.

The device is also set up on the basis of the basic data 111 an action item 103 a basic code in the process description 221 in a process description language (to determine the action element 103 (possibly alone) based on the basic data 111 to describe). In addition, the device is set up, the basic code 111 depending on the detailed data 112 , 113 , 114 to adapt to an action code 120 for the action item 103 to provide (of the action item 103 taking into account the basic data 111 and the detailed data 112 , 113 , 114 describes).

The present invention is not restricted to the exemplary embodiments shown. In particular, it should be noted that the description and the figures are only intended to illustrate the principle of the proposed methods, devices and systems.

Claims (10)

Computer-implemented method (300) for planning an assembly and / or manufacturing process for a product, the method (300) comprising - Providing (301) a process description (100) of the assembly and / or manufacturing process with one or more process modules (101); wherein a process module (101) comprises one or more action elements (103); an action element (103) being described by basic data (111) and by detailed data (112, 113, 114); wherein the basic data (111) indicate a component with which an action is carried out as part of the action element (103); the basic data (111) indicating where the action is performed on the component; and wherein the detail data (112, 113, 114) indicate one or more further parameters related to the action performed within the action element (103); - Determining (302), on the basis of the basic data (111) an action element (103) of the process description (100), a basic code (221) in a process description language, and - Adapting (303) the basic code (221) as a function of the detail data (112, 113, 114) in order to provide an action code (120) for the action element (103). Method (300) according to one of the preceding claims, wherein - the basic code (221) is determined by means of a machine-learned system (211); and / or - the action code (120) is determined by means of a machine-learned system (212, 213, 214). Method (300) according to Claim 2 , wherein - a machine-learned system (211, 212, 213, 214) was trained on the basis of training data with a large number of training data sets; and - a training data record comprises a code (221, 120) which was determined by a human planner for an action element (103). The method (300) according to any one of the preceding claims, wherein - The detailed data (112, 113, 114) of an action element (103) comprise several parameters for describing the action of the action element (103); - The parameters include in particular a weight of a component, an extent of movement, a size of a component and / or a number of components; and the method (300) comprises iteratively adapting the code (221, 222, 223) for the action element (103) depending on a respective one of the parameters in order to provide the action code (120) for the action element (103), which takes into account all parameters of the detailed data (112, 113, 114). Method (300) according to one of the preceding claims, wherein the method (300) comprises determining the process description (100) based on an electronic description (201) of the product by means of a machine-learned system (200). The method (300) according to any one of the preceding claims, wherein - the process description language is a standardized language for describing work processes; and or - includes the process description language MTM-1, MTM-UAS, MTM-HWD and / or BMW SD. The method (300) according to any one of the preceding claims, wherein - the action of an action element (103) has an action type from a plurality of action types; and - The multitude of action types includes: picking up a component, placing a component, performing a body movement, actuating an operating element, performing a visual check, performing a reading activity, and / or using an aid. Method (300) according to one of the preceding claims, wherein an action code (120) and / or a basic code (221) indicates, one type of action from a plurality of predefined action types; and or - a time value related to a period of time required to execute the action. The method (300) according to any one of the preceding claims, the method (300) comprising - Providing an assembly line for the assembly and / or manufacturing process based on the action codes (120) for the one or more action elements (103) of the process description (100) of the assembly and / or manufacturing process; and or - Production of a product depending on the action codes (120) for the one or more action elements (103) of the process description (100) of the assembly and / or manufacturing process. Device for planning an assembly and / or manufacturing process for a product, the device being set up, - to provide a process description (100) of the assembly and / or manufacturing process with one or more process components (101); wherein a process module (101) comprises one or more action elements (103); an action element (103) being described by basic data (111) and by detailed data (112, 113, 114); wherein the basic data (111) indicate a component with which an action is carried out as part of the action element (103); the basic data (111) indicating where the action is performed on the component; and wherein the detail data (112, 113, 114) indicate one or more further parameters related to the action performed within the action element (103); - on the basis of the basic data (111) of an action element (103) of the process description (100) to determine a basic code (221) in a process description language; and - To adapt the basic code (221) depending on the detail data (112, 113, 114) in order to provide an action code (120) for the action element (103).

Images