CN111356353B - Electronic control of pick-and-place machines in the production of printed circuit boards - Google Patents

Abstract

The invention relates to a method for controlling a mounting process for mounting printed circuit boards, which is carried out in a production line by means of a plurality of pick-and-place machines (BA). The lines are controlled by a Line Optimizer (LO). Furthermore, a local optimization unit and a control module (SM) are provided, which converts the assembly task (ba) into a matching assembly task (ba') according to an adaptation algorithm. The invention also relates to a computer program, a control module (SM), a pick-and-place machine (BA) having a control module (SM) and a system.

Description

Electronic control of pick-and-place machines in the production of printed circuit boards

Technical Field

The invention relates to the field of assembling electronic components for printed circuit boards by means of a plurality of automatic assembling machines, which are arranged in a production line. The invention relates in particular to a software-based control device for an automatic placement machine.

Background

In assembly technology, electronic components are mounted on the surface of printed circuit boards, in particular in the so-called Surface Mount Technology (SMT), by means of so-called pick-and-place machines. The components to be assembled are supplied to the pick-up position by means of a supply device for the assembly process. For positioning the electronic components, a movable mounting head is inserted into the pick-and-place machine by means of a positioning system.

A plurality of pick-and-place robots are usually arranged in a production line, wherein transfer devices transport the circuit boards to be mounted from the pick-and-place robots to the pick-and-place robots, wherein different circuit boards are mounted for all pick-and-place robots in parallel operation in a respective time period.

In order to mount all circuit boards, it must now be determined at which point in time which components are to be positioned by which pick-and-place machines. This is done according to a digital assembly plan, which can be created, for example, by means of a computer program. Such an assembly plan is created for each circuit board type to be assembled before assembly, wherein the circuit board types differ in particular with regard to the design of the circuit to be assembled and with regard to the predefined conductor track structure. The assembly plan is preferably generated such that the total processing time for producing the printed circuit board is as minimal as possible. The assembly plan can thus be optimized. For this purpose, so-called line optimizers can be used, which collectively generate an optimized assembly plan for all automatic assembly machines of the production line.

In principle, an efficient assembly process with high quality requirements is of great importance in the production of printed circuit boards, in particular in SMT production. For example, if the components required for assembly are not available or are available too late in a specific pick-and-place machine of the circuit, the pick-and-place machine cannot carry out the assembly planning subtasks assigned to it, or can carry out them only with a delay. This has an adverse effect on the overall production process. The production line is either interrupted or delayed. This in turn has the disadvantage of incurring additional costs.

The cycle time for the further switching on of the conveyor belt for the transfer of circuit boards from one pick-and-place machine to the respective next pick-and-place machine is determined by the slowest mechanism in the chain, i.e. by the slowest robot. It is therefore of great importance to optimize the priority of the subtasks to be carried out by the pick-and-place machines in the production line. In this respect, it is known in the prior art to use a line optimizer, which breaks the entire assembly process down into individual subtasks, so that the entire assembly process can be carried out as efficiently as possible by the available automatic assembly machines. The purpose of the optimizer used in applicant's SIPLACE Pro program is that all automatic assembly machines of the line require the same processing time in order to complete their assembly process, so that no latency occurs in some of the machines. If no errors occur, a maximum performance or overall installation performance of the line can be achieved.

However, if temporary faults occur in the individual pick-and-place machines or machines (e.g. welding errors, reduced pick-up rates, torn films), this has an adverse effect on the production rate of the circuit boards for all machines. The non-failed machine is shut down after completing its assembly process until the machine has failed to end its assembly process. Thus, in addition to the assembly performance of a faulty machine, the utilization of non-faulty machines and the assembly performance achieved are also reduced and the overall assembly performance is reduced.

A drawback of today's systems is that temporary failures cause accumulations. If another fault occurs in the continued progress of the line after the fault is eliminated, the delays add up and the overall assembly performance of the line further degrades.

Temporary faults are unpredictable and occur at different machines. Other methods are too slow for this.

There is also a need in SMT production to increase flexibility in assembly. On the one hand, different printed circuit board types are usually to be mounted in any desired sequence on a pick-and-place machine of a production line. During production, for example, it can happen that different circuit boards are grouped in each pick-and-place machine. In principle, only components provided for the respective pick-and-place machine can be mounted in a pick-and-place machine. For example, components of type a and type B can be mounted on the circuit board in a first pick-and-place machine and components of type a and type C can be mounted on the circuit board in a second pick-and-place machine. For example, the problem arises that the individual assembly positions are either assembled several times or not assembled at all, since the individual assembly positions are handled by different automatic assembly machines, so that other automatic assembly machines are not known. In order to avoid this problem, a central management of the upper levels of the individual pick-and-place machines is important for controlling the entire placement process. In addition, the central control management can also assume the task of central line optimization.

However, in such central control units, it is not possible or appropriate to react to local changes at the individual pick-and-place machines (for example, in the event of local errors). This can adversely affect the overall production process.

Disclosure of Invention

The object of the present invention is therefore to provide a solution by means of which printed circuit boards can be mounted more efficiently in a production line by means of a plurality of pick-and-place machines without impairing the flexibility in the processing by the individual pick-and-place machines.

The inventors have recognized that temporary faults in the course of the assembly line can only be reacted insufficiently by means of fixed or statically preset assembly tasks. A flexible adjustment of the respective assembly tasks is therefore desirable, which makes reference to the current processing state or possible errors of the machine.

This object is achieved by the features of the independent claims, in particular by a method, a computer program, a control module, a pick-and-place machine with a control module and a system. Advantageous embodiments of the invention are set forth in the dependent claims and in the following description.

Thus, according to a first aspect, the invention relates to a method for controlling a mounting process performed by a pick-and-place machine in a production line having a plurality of pick-and-place machines for mounting printed circuit boards. The automatic placement machines in this production line perform the subtasks of the overall placement plan accordingly. The method comprises the following method steps:

reading assembly tasks (including control commands for assembly) generated respectively for the respective pick-and-place machines, which assembly tasks are preset for which positions on the circuit board are to be assembled for which component types, wherein the assembly tasks are made flexible and, in addition to each position, adjustment parameters are provided which comprise calculated SHOULD, CAN or MUST data sets in order to determine accordingly which assembly position SHOULD, CAN or MUST be carried out by the pick-and-place machine,

checking whether the processing state is received by the preceding automatic assembly machine or another management unit and in this case: detecting the processing state of a printed circuit board to be assembled;

executing an adaptation algorithm to flexibly adapt the read assembly tasks based on the adjustment parameters and, if necessary, the detected machining state to calculate matched assembly tasks that determine which assembly positions differ from the read assembly tasks for assembly and which assembly positions differ from the read assembly tasks for non-assembly;

-operating the pick-and-place machine with the matching pick-and-place task;

after the matching assembly task is performed, the processing state of the printed circuit board is detected (wherein the processing state indicates which position is assembled and which position is not assembled).

The inventors have the idea of making it possible to flexibly preset assembly plans centrally for the individual pick-and-place machines in the line, which have a fixed processing task and can be adapted to the local machine conditions (e.g. delays).

In principle, after a respective assembly cycle (or a period of time during which the transfer mechanism has a constant position and is not in motion), the printed circuit boards are transported on the transfer mechanism from the pick-and-place machine to the next pick-and-place machine in the production line. The transfer, i.e. the continued switching-on, is usually not carried out synchronously for all machines. The individual conveyor belts are synchronized with the predecessors/successors via a raster, so that it is known when the transfer can continue.

In a preferred embodiment of the invention, the steps "executing the adaptation algorithm" (and optionally "providing the matching assembly tasks as result data sets") and "handling the automatic assembly machines with matching assembly tasks" are iteratively repeated until the interrupt condition is met. Only if the interruption condition is met, the process continues with the step "detect the machining state". The interrupt condition may be preset. The interrupt condition is preferably:

"subsequent robotic assembly machines have not completed" and ("other CAN locations available" or "SHOULD/MUST locations as well").

Otherwise, the iteration may be performed until the repetition condition applies:

"other pick-and-place machines of the production line are not yet completed" and ("other CAN locations are available on the current pick-and-place machines" or "SHOULD or MUST locations as well").

In a preferred embodiment of the present invention, the matching assembly tasks include DO MORE and/or DO LESS commands.

In a preferred embodiment of the invention, the processing state is transmitted electronically (as an electronic data set) directly to the subsequent pick-and-place machines in the production line. This is done through M2M communication via the provided network. In particular, the whisper protocol (disclosure of WO 2000015017 for this reference) and/or the protocol via IPC-letters-9852 can be used here. It is possible to provide for this that the machine M1 is assembled less, but not by the immediately following machine M2, but more can be assembled later, for example, by the fourth machine M4. The respective processing states are then transferred iteratively from M1, to M2, to M3, and to M4. The data exchange from the pick-and-place machine to the pick-and-place machine preferably takes place in each case during further movement by means of the transfer mechanism.

In another preferred embodiment of the invention, the processing state may additionally be sent to a control module for controlling the production line. The control module may be configured with line optimization functionality. This has the advantage that the current processing state is not only communicated locally from machine to machine, but can also be made centrally, so that, for example, further analyses can be carried out.

According to an advantageous further development of the invention, the adaptation algorithm detects whether a delay or an error state is detected at least one of the automatic assembly machines of the production line, so that a matching assembly task is calculated in connection therewith or on the basis thereof. The delay may be detected locally on the respective machine. But this delay may also occur on another machine of the line and cause the adaptation algorithm to dynamically match or change the assembly task.

According to a further advantageous development of the invention, the process state comprises an electronic data set (which can be configured, in particular, as OMIT or DONE parameters or flags). From the processing state, commands for the modified control of the robot can be automatically calculated, in particular MORE (DO MORE command) or LESS (DO LESS command) by the respective automatic placement machine controlled according to the matched placement tasks. In addition, the processing state may also include other data, such as metadata, time stamps, and the like. When the subsequent automaton is not complete, the DO MORE command is output and executed. When the subsequent automaton is completed, the DO LESS command is output and executed.

In a further preferred embodiment of the invention, in the case of MORE assembly (DO MORE) to be carried out, corresponding MORE assembly tasks (as matching assembly tasks) can be assigned to the individual pick-and-place machines of the production line. The allocation can advantageously be adapted to the workload and to the current resources of the pick-and-place machine. The assignment can be performed by a central management unit, for example, a central line optimizer or mounter. Preferably, the dispensing is performed autonomously by the respective pick-and-place machine.

Preferably, the SHOULD data record is defined such that, in the absence of delays or errors, the corresponding position is to be assembled by the corresponding automatic assembly machine, and in the presence of delays or errors, as far as possible no assembly takes place. It may be provided here that not all locations with SHOULD data sets are "ignored" (i.e. not assembled); a part may already be assembled at the wrong point in time. The assembly/disassembly is autonomously determined by the respective pick-and-place machine by using an adaptation algorithm.

The CAN data record is preferably defined such that, in the event of a delay or error, the respective position CAN be assembled by a corresponding automatic assembly machine. A delay or error in the position occupied by the CAN (or the reason for this) CAN only be a delay on another line. If the machine itself fails, it is LESS assembled (DO LESS) and no longer assembled.

Preferably, the MUST data sets define the respective positions that MUST be assembled by the respective pick-and-place machine. This is the case in particular if the last pick-and-place machine in the line has a corresponding functionality.

In a further preferred embodiment of the invention, the effect-increasing value can be calculated for the implementation of an assembly plan which, using the adaptation algorithm, represents a time saving compared to a method without the application of the adaptation algorithm. Performance gains can thus be expressed numerically.

In a further preferred embodiment of the invention, a prediction data set can be calculated which, taking into account the adaptation algorithm used, provides a prediction for better equipping the pick-and-place machine from the analysis of the already executed assembly plan.

In a further preferred embodiment of the invention, an estimate can be derived for the assembly delay caused by the error in the case of a delay or error. The use of an adaptation algorithm can be activated, for example, only when the resulting estimated value exceeds a predefinable threshold value. This saves local resources of the pick-and-place machine. In an advantageous development of the invention, it can be provided that the evaluation value is also transmitted to the central line optimization unit, for example for evaluation.

In a further aspect, the invention relates to a computer program having a program code or a program medium, wherein the computer program can be stored on a computer-readable data carrier, wherein the program code or the program medium, when the computer program is executed on a computer or on a computer-based processing unit, causes the computer to carry out the method as described above.

In a further embodiment, the invention relates to a control module for a pick-and-place machine which, together with other pick-and-place machines, is used for mounting printed circuit boards in a production line. The control module is designed to control the pick-and-place machine by means of a method as described above.

In a preferred embodiment of the invention, the control module can be configured with ports to other control modules of other pick-and-place machines. In this case, a local network port, such as ethernet or SMEMA, may be provided.

In a further aspect, the invention relates to a pick-and-place machine with an (electronic) control module as described above.

In another aspect, the present invention is directed to a system for controlling a production line for producing printed circuit boards, the system having:

a plurality of pick-and-place machines for mounting printed circuit boards, wherein each pick-and-place machine executes a respective subtask of a mounting plan and wherein each pick-and-place machine has an electronic control module;

a line optimizer which determines for this purpose the division of the assembly plan into subtasks for assignment to the individual pick-and-place machines of the production line and also determines for this purpose the flexible assembly tasks which are respectively generated for the respective pick-and-place machines and which are preset in which positions on the printed circuit board which component types are to be assembled (and preferably also the line optimizer is preset in which time and in which sequence the assembly is to be carried out), wherein the assembly tasks are made flexible and, in addition to each position, adjustment parameters are also determined which comprise calculated SHOULD, CAN or MUST data sets in order to respectively determine which assembly position is to be, CAN or MUST be carried out by the pick-and-place machine;

data connections between the individual pick-and-place machines and the line optimizer and between the individual pick-and-place machines.

In the assembly process according to the prior art, assembly plans have been created by a central line optimizer and divided into "task packages" and distributed in the form of unchangeable, static assembly tasks specifically to the respective automatic assembly machines for implementation. The assembly task can no longer be changed after. If a fault occurs in an automation unit at this time, it is not possible to react on this basis with sufficient flexibility. This fault usually affects other automata in the line, which can continue its operation, for example, only when the fault is cleared. This interrupts the line. Local and dynamic changes of the state of the assembly line, which are created centrally and require assembly tasks to be carried out locally, and which match the assembly line, are not possible. Furthermore, the individual robots of the prior art cannot process information about the processing state of the other robots of the line when they are available. It is therefore not possible for the pick-and-place machine to react accordingly to this information. This disadvantage is eliminated according to one embodiment of the implementation method proposed here, in which the processing state is communicated directly and immediately from one robot to the respective next robot, so that local assembly decisions can be made and corresponding machine control signals can be generated as a reaction (by means of a matching assembly task) on the basis thereof.

The invention thus offers the important advantage that the assembly plan can be adapted more flexibly to the current operating situation and that, in particular in the event of a temporary error, better utilization of the line load is possible. The adjustment of the initial (original and centrally created) assembly tasks can be carried out locally on the pick-and-place machine. The adjustment is therefore carried out locally and can be carried out by means of an adaptation algorithm, in particular at runtime. Whereby the line can react very quickly to a temporary change. The production duration can be reduced and the idle times of the individual pick-and-place machines of the machine which has to "wait" for a malfunction can be avoided. Thereby saving costs. In addition, more data is available that can be used to analyze and improve the production process. The invention thus provides for the detection of the processing state at each pick-and-place machine after each placement cycle or after the execution of the associated placement task. This processing state yields information about the quality and freedom from errors of the assembly up to this point. Determining the machining state provides a technical basis for measures that can be automatically started to compensate for possible delays or errors (e.g. due to non-assembly). The processing state can be sent to the pick-and-place machines and/or the central line optimizer directly connected in the line and/or to further computer-based processing units connected, for example, via a network for further analysis.

The terms of the present application are defined below.

The term "pick-and-place machine" is understood to mean a machine device which is suitable for placing components on component carriers, for example electronic circuit boards. As already mentioned, the pick-and-place machine (also referred to below simply as a robot or machine) can have a pick-and-place head which picks up a component from a component supply, transfers it into a placement region of the pick-and-place machine and places it on a component carrier. The automatic placement machine may have a carrier with a support arm, on which the placement head is movably arranged. The pick-and-place machine or its components can preferably be controlled electronically. The pick-and-place machine preferably comprises an electronic processing unit, which may comprise, in particular, a control module. Here computer-based electronic means comprising a digital processing unit for processing digital data, such as a CPU, microprocessor, microcontroller or integrated circuit, for example in the form of an FPGA or the like. Automatic assembly machines are common components of production lines and are arranged in series.

The term "serial" refers here to the situation in which a conveyor belt or other transfer device guides the circuit boards to be assembled from the machine to the machine. The corresponding automaton processes the assembly tasks specifically assigned to it almost in parallel. One circuit board is processed in the machine per cycle or time period. The circuit boards are conveyed further from cycle to the respective next cycle by means of a conveyor belt or another conveyor. It is desirable to continue the transmission as synchronously as possible. If a fault occurs in the prior art, all automatic assembly machines have to wait for the faulty machine, which means that the performance is degraded. The adaptation of the assembly tasks according to the invention makes it possible to eliminate the fixed assignment of the assembly tasks for each machine and to make it flexible.

The term "production line" may particularly denote a production system in which a plurality of stations, in particular processing stations, are connected within a process line. The process line has a direction of propagation here. The processing stations can have, for example, automatic assembly machines, and can be arranged one after the other and in the transfer direction along the process line, wherein the processing stations can be connected by means of a conveyor, in particular a conveyor belt. The conveying direction corresponds here to the main direction of extension of the production line. In other words, the conveying direction denotes the direction in which the product to be manufactured (e.g. printed circuit board) passes through the production line.

In addition to the automatic assembly machine, the production line can have, for example, the following "processing stations": a solder stamping press, an oven, a warehouse unloader, a marker, a solder inspection device, a buffer, a CPI device, an automated optical inspection device (AOI), and/or a warehouse loader. For example, component carriers, in particular printed circuit boards, can be picked up by a storage unloader into a production line and then transported along a conveyor, for example a conveyor belt, in a main direction of extension and the components can be mounted there by a pick-and-place machine. The electronic components, in particular the assembled printed circuit boards, can then be removed from the production line at the end of the production line by means of a storage unloader. Multiple production lines may be positioned substantially parallel to each other. Furthermore, the production apparatus may have other components, such as a material library, from which the component storage parts can be transported to the production line. Furthermore, the production plant may also have a control unit which is required for coordinating the process steps of the production line, for example a line optimizer.

The line optimizer may include an electronic processing unit. This relates to computer-based electronic components comprising a digital processing unit, such as a CPU, microprocessor, microcontroller or integrated circuit, for example in the form of an FPGA or the like, for processing digital data. The memory is typically provided locally or centrally and is in data communication with the processing unit. The line optimization mechanism results in centralized control and optimization of the entire line. The circuit optimization machine can preferably be designed as all or selected automatic assembly machines for the circuits. The circuit optimization machine can be designed to generate an assembly plan for a component carrier, for example for a printed circuit board. The assembly plan can be divided into a plurality of assembly tasks by an algorithm, and the assembly tasks are performed on the individual automatic assembly machines. The first assembly task is assigned to a first pick-and-place machine, for example, and the second assembly task is assigned to a second pick-and-place machine, for example.

The assembly tasks are electronic data sets and contain commands for the respective assembly of the printed circuit boards by the respective automation (e.g. by means of the respective functionalities, equipment, etc.). The assembly task may include a plurality of components and in particular data defining which component types are assembled on which positions on the printed circuit board and, if necessary, also in which sequence. The assembly is in principle carried out in series. This means that the printed circuit board is usually equipped with components in sequence, i.e. for example a first component a, then B, then C. But this order is not mandatory, but only the result of path optimization (traversing Salesman Algorithmus). The assembly task also includes at least one calculated tuning parameter. The tuning parameters themselves comprise SHOULD, CAN or MUST data sets at each assembly location. The control parameters are used to determine which mounting position (SHOULD), CAN or MUST (MUST) is to be executed by the pick-and-place machine. Thus, for example, an assembly task can be a set of the following data sets:

element type 1, to be assembled in position 1, MUST

Component type 2, to be mounted in position 2, SHOULD

Component type 2 to be mounted in position 3, CAN

The "matching assembly task" differs from the "assembly task" in that an adaptation algorithm is used on the (read) assembly task.

The adaptation algorithm evaluates the provided control parameters for each position and carries out the adaptation on the basis of the current state (error-free or faulty/faulty) of the existing resources and/or of the pick-and-place machine. The matching algorithm is implemented in particular after each de novo loop. De novo circulation represents, for example, about 12-20 positions, for example, depending on the number of pipettes. In principle, the assembly head picks up the 12-20 components from the conveyor before it runs onto the assembled circuit board and assembles it. The process consisting of pick-up and assembly and travel path is called a de novo cycle.

For example, if there is some free space to signal at the time of assembly (e.g., CAN, SHOULD) and there is a delay or failure to implement the assembly, then a command "DO LESS" (LESS assembly) may be generated which is added to the matching assembly task. In this case, the machine actually assumes "fewer" tasks than in the original assembly task, so that the line is assembled without interruption. For example, if the tuning parameters limit the free space (e.g., MUST) when assembling the corresponding location and there is no failure to perform the assembly, but the assembly task does not set the assembly location, a command "DO MORE" may be generated at this point, which is added to the matching assembly task. In this case, the machine performs the assembly more than actually takes on the "more" task in the original assembly task. The matching assembly tasks are calculated locally on the respective pick-and-place machine. The original assembly task is adapted to the current production situation in the automatic assembly machine on a case-by-case and dynamic basis and is overwritten with the command DO MORE or DO LESS, in particular as a function of the location, in order to generate an adapted assembly task which is optimized locally as a function of the existing production situation (e.g. error cases, delays, etc.). Decisions (DO MORE or DO LESS) are made for each location, respectively. The decision is preferably made in one adjustment step and ideally for the entire de novo loop. If neither DO MORE nor DO LESS is generated for at least one location, the (original) assembly tasks can also be reconciled with the assembly tasks matched for the respective case.

The following rules (described as pseudo-code) may be employed for generating the DO MORE command:

LOOP ((until the automatic assembling machine is completed) AND (the next automatic assembling machine is not completed))

THEN performs other assembly with de novo cycle (═ 12-20 positions);

the following rules (described as pseudo-code) may be employed for generating the DO LESS command:

IF ((incomplete automatic assembly machine) AND (complete automatic assembly machine after that))

THEN did not assemble (neglect) PERENTAGE (percent) of the remaining SHOULD assembly positions;

PERENTAGE (percent) is an optional parameter to optimize for this.

The "readiness of the subsequent pick-and-place machine" of the data set or information can be calculated from the preparation notice of the transport mechanism. This information gives a signal whether a subsequent pick-and-place machine is ready in the production line. The prepared message may be detected, for example, from a state in which the output strip is empty. This state gives a signal and the subsequent automatic assembly machine is completed. In other cases, it may be concluded that the machine is on a critical section while the input tape is used and while the output tape is empty.

The term "process status" is an electronic data set and represents or represents the current assembly or condition of the circuit board. The processing state can, for example, indicate that the respective assembly task has been completely or not yet completely carried out. The sum of all processing states is referred to as the assembled state of the printed circuit board. The processing state can detect each position (mounting position on the printed circuit board). From the machining state, an error can be deduced. The processing state is transmitted from the machine to the machine, in particular to the machine following in the respective line, by means of an electronic data transmission device. Communication takes place according to the whisper protocol and/or according to a protocol of the IPC Hermes9852 type. The processing state may also be transmitted to a central line optimizer or another central management for further analysis. Each local pick-and-place machine determines its processing state after the completion of the placement or after the "completion" of the work according to the associated placement task. In principle, two options are available for determining the processing state: first, and preferably, the processing state can be derived from the result of the adaptation algorithm (it is then assumed that the mounting machine can always correctly obtain all control commands from the matching mounting task). The second method can be implemented cumulatively or alternatively, for example, optically. In the latter case, for example, deviations between the actual and the theoretical assembly can be detected.

The adaptation algorithm is an automatically implemented computer program which can be executed locally on the pick-and-place machine. The adaptation algorithm is used to adapt the original assembly task in respect of possible more or less assemblies, depending on the detected state of the pick-and-place machine or the production line. In this case, temporary delays should be taken into account, in particular simultaneously. Advantageously, the method exerts an advantageous effect at a plurality of delays and at high-speed operation of the production line. The adaptation algorithm overwrites the original assembly task with a matching assembly task that contains for each location a DO MORE or DO LESS command or may also be the same as the original assembly task (e.g., if no errors or failures have occurred).

The foregoing describes an implementation in accordance with the objectives of the method. Features, advantages or alternative embodiments mentioned herein may also be applied to other claimed objects and vice versa. In other words, particular claims (e.g., claims directed to a system or directed to a computer program product) may also be presented with method-related described and/or claimed features. The respective functional features of the method are formed by the respective specific modules, in particular by hardware modules or microprocessor modules, systems or products, and vice versa.

Another implementation is a computer program with computer code to implement all the method steps of the above-described method when the computer program is executed on a computer. For this purpose, the computer program can also be stored on a computer-readable medium.

Drawings

Embodiments and features thereof, as well as other advantages, which are understood in a non-limiting manner, are set forth in the following detailed description of the drawings, which is to be read in connection with the accompanying drawings. Shown in the drawings are:

fig. 1 shows an overview of a plurality of pick-and-place machines, which are operated by a line optimizer.

Fig. 2 shows an overview of a plurality of pick-and-place machines, which are operated by a transfer device.

Fig. 3 shows an interaction diagram for exchanging messages between a control module of a pick-and-place machine and a line optimizer.

Fig. 4 shows a flow chart of a method according to a preferred embodiment of the invention.

Fig. 5 shows a further flowchart of a method according to a preferred embodiment of the invention.

Fig. 6 shows a schematic overview of a production line with four pick-and-place machines over a plurality of time periods.

Description of the figures

BA automatic assembly machine

SM control module

V processing unit

LP printed circuit board

NW network, in particular local network

LO line optimizer

TV transmission device

FL production line

Ba Assembly task

ba' matched Assembly tasks

bz processing state

t element type

S1 read Assembly task ba

S2 detecting the machining state bz

S3 application adaptation algorithm

S4 provides the matching assembly task ba' as a result data set

S5 actuates the automatic placement machine with a matching placement task ba'.

Detailed Description

The invention serves to locally adapt centrally generated assembly tasks and is described below in the example of printed circuit board production. The invention is also applicable to other components to be assembled.

The strict assignment of assembly tasks to assembly machines is relaxed. There are two types here:

1. MORE Assembly (DO MORE command)

When machine a completes its assembly process (synonymous with assembly task), but subsequent machine B has not signaled its readiness (e.g., due to a temporary failure), machine a takes over all subsequent machines C/D/… and the partial assembly process of the failed machine B if the components assembled in machine a are available for this. Information about more assembly is transmitted to the subsequent machine by means of the machining state (data indicating that additional assembly positions have been occupied). For which a network connection may be used. The (subsequent) machine B/C/D/… that should actually assemble these assembly positions ignores these assembly tasks that match them accordingly.

2. LESS assembly (DO LESS command)

When machine a stops the remaining line machines due to a temporary fault, the machine ignores the assembly of a portion of the remaining assembly locations. The assembly task matched at this time comprises a DO LESS command. Provided that the assembly location can be assembled in other production line processes. It must especially be ensured that the elements are equipped so that there is a correct pipette or the like. These assembly positions are assembled at the latest in the last machine Z, which can perform the make-up. The information about the omitted assemblies continues to be displayed in the processing state, so that the missing assemblies can be implemented in other processes of the production line (also referred to simply as "line").

The line optimizer (also referred to as "optimizer") generally specifies which assembly positions can be omitted or additionally assembled, since the line optimizer knows the equipment of the entire line. Furthermore, it is determined in which machine of the circuit a specific installation position must be occupied at the latest. This is done by adjusting parameters included in the assembly task.

If a temporary fault occurs, more or less equipment is autonomously and automatically determined in the respective machine. The optimizer does not participate in this decision.

As in the case of the workstation-based material data recording in the methods up to now, the machine reports, for example for retrospective applications, which components have already been assembled. Thereby also securing traceability of components while relaxing the assembly distribution relationship.

First, the optimization machine calculates which assembly locations may/must be ignored by the machine or additionally assembled. This is done by calculating the adjustment parameters for each position and is done specifically for each pick-and-place machine individually. The calculation is carried out centrally for all pick-and-place machines, since they know the equipment of the entire circuit. Furthermore, it is determined up to which machine of the line must occupy the particular installation location at the latest. The attributes of the mounting locations in the tuning parameter data set are:

-SHOULD：

if it is actually to be assembled, but not necessarily. This attribute is assigned to the assembly position according to which the optimization machine should be assembled on the machine in order to achieve the best overall assembly performance without failure. But allows the assembly to be ignored due to failure with fewer assemblies. Assembly is of course only carried out when no components have yet been assembled there.

-CAN：

Assembly may be performed at this point, but is not required, when no components are present. The component is assembled in the machine only if more assembly takes place due to a failure of another machine.

-MUST：

There are no components yet, which must be assembled at this point. The attribute contains a mounting location in the last machine of the line, which can be mounted. This may be the same machine that is assembled without failure.

Two types of more and less assembly are described below:

MORE assembly/DO MORE:

if a fault occurs on one machine, the remaining machines do not wait for the readiness signal of the faulty machine to be inoperative after completing their assembly process. But rather the assembly process continues as the assembled components may be used in other assembly locations. The control parameter "CAN" based on the assembly position transmitted by the optimization machine automatically and autonomously decides to carry out more assemblies in each machine. This does not affect, above all, the circuit board productivity or the overall assembly performance of the circuit.

This method is not advantageous if only one of the machines has a fault, since the possible reduction in the circuit board processing time in the subsequent circuit runs does not lead to a reduction in the overall assembly performance. Since the pick-and-place machine arranged after the temporarily defective machine only assembles a part of its actually planned components, the pick-and-place machine is then stopped. The reduced load is only moved to the rear of the line.

However, if another fault occurs in the subsequent line operation, the advantages of this method are evident: the second failure results in no or only a small reduction in the overall assembly performance, since the machine that now fails needs to be assembled with fewer components due to more assemblies through the preceding machine. Thus, the circuit board productivity of the wiring is not further lowered, thereby avoiding further lowering of the overall assembly performance.

Information about more installations and additional installation positions that have already been occupied is passed via the network ("whisper") and the installation position is ignored by the machine that should actually install it.

LESS assembly/DO LESS:

if a fault occurs on a machine A, the machine ignores a portion of the components to be assembled. This is only done for components that can still be assembled, i.e. prepared, in the subsequent course of the line. As a result of the neglect, the remaining machines, after their assembly tasks have been completed, do not wait for the readiness signals/start-up of the faulty machine on the fly, but can continue to operate with normal cycles. The omission of a part of the components to be assembled prevents the malfunction from affecting the cycle time in the first place.

The components that are ignored by machine a must be made up by the subsequent machine B/C/…, where more assembly must take place.

Machine a and its predecessor, which are no longer malfunctioning at this time, in machines B/C/… that are located in the line after the first malfunctioning machine a during more installations, may also be undergoing more installations (see the "more installations" paragraph above) because the successor machines have not yet signaled readiness. Thus pre-treating and prophylactically mitigating potential new failures.

As in the case of the workstation-based material data recording in the methods up to now, the machine reports, for example for retrospective applications, which components have already been assembled. Thereby also securing traceability of components while relaxing the assembly distribution relationship.

Further prediction and analysis can be achieved by detecting and storing the process state. In particular, the time period elapsed can be analyzed and advantageous effects can be achieved. The potential that has not been exploited in the past period of time can also be analyzed and conclusions can be drawn in the future. In order to work effectively, the invention presupposes that the same components are provided in as many different machines as possible. It is possible to analyze which performance improvements have been achieved by this method when the component Z has additionally been assembled in a particular machine Y in the past period of time. Analysis of the past can be used to predict the future. Informing the user which advantages the additional device has. The user may then choose to do additional provisioning assuming these will also be advantageous in the future.

The invention is elucidated below with reference to the drawings.

FIG. 1 shows a plurality of pick-and-place machines BA1, BA2, … BAn, which exchange data via a network. The pick-and-place robot BA also exchanges data with the line optimizer LO. Each pick-and-place machine comprises an electronic processing unit V. A control module SM may be provided on the respective processing unit V. A first processing unit V1, which includes the first control module SM1, can thus be arranged on the first pick-and-place robot BA1, and a second processing unit V2, which includes the second control module SM2, etc., can be arranged on the second pick-and-place robot BA 2.

Alternative architectures may also be provided, for example to pass the result of a locally implemented adaptation algorithm (which represents the decision DO MORE/DO LESS) to the central processing unit. It is important that the implementation of the adaptation algorithm is always carried out locally on or for the respective robot and takes into account its production and the processing state of the circuit board (up to and including this point).

Fig. 2 shows a schematic illustration of a pick-and-place machine BA1, BA2, BAn, in which the circuit boards to be mounted are transferred periodically and synchronously from the robot BAi to the robot BAi +1 via a transfer device TV (for example a conveyor belt). The automatic assembling machines BA (generally 3-12) form a production line FL. (but there is also a significantly longer production line FL). If a delay occurs in one of the automata BA, this represents in the machine state and this is to be reported as a message to the subsequent automata, which can execute an adaptation algorithm in response to the message. The transfer of messages and the data exchange associated therewith is described in detail below in connection with fig. 3.

If temporary faults occur in succession in time on the individual robots BA, the adverse effect of further faults on the overall assembly performance is reduced or avoided. Thus improving the practical performance of the line FL.

A sequence diagram is shown in fig. 3. The control modules SM arranged in the pick-and-place robot BA communicate with each other and with the line optimizer LO. At the beginning, the ith control module SMi receives an assembly task ba from the line optimizer LO. The assembly task includes adjustment parameters for the assembly positions to be assembled on the respective pick-and-place machine, which have specifications SHOULD, CAN or MUST. The control module SMi detects the processing state bz by its "former" in the production line, the control module SMi-1, and may then implement an adaptation algorithm on the control module SMi. The adaptation algorithm is used to decide locally which position to fit or not to fit, which may be different from the originally detected fitting task. The adaptation algorithm transforms the assembly task of the line optimizer LO into a locally optimized matching assembly task ba'. The assembly is then carried out in the automatic assembly machine BAi with a matching assembly task ba', which may include MORE or fewer assemblies (DO MORE or DO LESS command) if necessary. As shown in fig. 3, the method is performed iteratively. The length of the fault is unknown in advance. The CAN positions are thus processed cyclically from the beginning (i.e. 12-20 positions at a time) so long after the SHOULD and MUST positions (i.e. the positions occupied by the SHOULD or MUST data sets) have been completed, until the printed circuit board LP CAN be transferred further. The same applies to the execution of the DO LESS command. This can also be checked iteratively in a loop. It is thus reevaluated whether the DO LESS command has to be executed even after each de novo loop. The following repeating pattern is thus obtained:

de novo cycle (KZy) - > Adaptation Algorithm (AA) - > KZy- > AA- > KZy- > AA- > KZy- > AA- > KZy- > AA- > and the like.

After the assembly is carried out, the current machining state bz and the current assembly state are detected. The processing state bz is transferred to the next pick-and-place robot BAi + 1. The machining state bz is then optionally transmitted to the line optimizer LO. As this is only optional, this is shown in dotted lines in fig. 3.

Fig. 4 shows a flow chart. The line optimizer LO sends a first mounting task BA1 to the first pick-and-place machine BA1 at reference numeral 1 (it should be noted in the drawing that the case: BA denotes a mounting task and the reference numeral BA denotes a pick-and-place machine).

In this example, the assembly task ba1 illustratively includes the following data sets:

element type t1, Position _1 or POS _1 and adjustment parameter MUST

Component type t2, Position _2(POS _2) and tuning parameter SHOULD

Component type t2, Position _3(POS _3) and control parameter CAN

In this case, the respective given component type is mounted at a given mounting location and a manipulated variable (MUST, SHOULD or CAN) is assigned to the subtasks.

The pick-and-place machine BA comprises a processing unit V1, which itself comprises a control module SM 1. In this example, the robot mounter BA1 is configured to mount the component types t1 and t 2. The automatic placement machine BA can naturally place only the component types that it has prepared for this purpose. In this example, robot BA1 is equipped with component types t1 and t2 and robot BA2 is equipped with component types t2 and t 3.

In the first embodiment, the control module SM1 of the processing unit V1 decides at reference numeral 2a that Position _2 is not to be assembled, but should be omitted. Therefore, the DO LESS command is output for Position _ 2. This is possible because Position _2 has the tuning parameter SHOULD. This decision can be based, for example, on the fact that the circuit board has been removed from the following pick-and-place robot BA2 and the next pick-and-place robot BA2 has been completed. Therefore, the initial assembly task ba for the corresponding position is covered. This is schematically explained in the stored data set as follows:

accordingly, for Position _2, the machining state bz contains an instruction OMIT indicating that the Position is not assembled in order to inform BA 2.

In an alternative second embodiment, the control module SM1 of the processing unit V1 decides at reference numeral 2b that Position _3 should be additionally set up (as opposed to the initial set-up task ba of the line optimizer LO). Therefore, DO MORE is output for Position _ 3. This is possible because Position _3 has the tuning parameter CAN. This decision can be based, for example, on the fact that the mounting task for the circuit board may have been completed and the circuit board may not yet have been removed from the pick-and-place machine BA (for example because the next management unit or the next pick-and-place machine is not yet ready). This can be explained exemplarily in the stored data set as follows:

accordingly for Position _3, the machining state bz contains an instruction DONE indicating that the Position has been assembled in order to inform BA 2.

The two examples mentioned above are to be understood as exclusive alternatives, since it is not meaningful to fit more on the one hand and to omit the position on the other hand. In principle, the DO MORE or DO LESS command for the overlay mounting task BA is generated locally on the pick-and-place machine BA and is implemented directly.

Thus, the processing state bz in the first example comprises an indication that the Position is written for Position _2 and that assembly is still missing or in the second example an indication that assembly has been performed for Position _ 3.

The processing state is transferred to the next robot mounter BA 2. The process state is an electronic data set and can be binary coded, for example, comprising at least two elements:

1. position indication on a circuit board and

2. indication of the current assembly status (assembled/unassembled).

The processing state bz1 for Position _2 is transmitted at 3a and the processing state bz2 for Position _3 is transmitted at 3b to the subsequent pick-and-place machine BA 2. It should be noted here that the reference numerals identified in the circles in fig. 4 do not have a time-enforced order. For example, the machining state bz2 and then bz1 can also be transmitted. Preferably, the transfer takes place in a common data packet in order to save network technology resources.

The second pick-and-place robot BA2 thus receives the current and actual processing states bz1, bz2 from the preceding pick-and-place robots.

The second pick-and-place machine BA2 also obtains a placement task BA2 from the line optimizer. This assembly task has been output at reference numeral 1 and includes, in this example, the following data sets by way of example:

element type t2, Position _3 and adjustment parameter MUST

Element type t3, Position _4 and adjustment parameter MUST

Element type t2, Position _2 and adjustment parameter MUST

At reference numeral 4a, the control module SM2 of the second pick-and-place machine BA2 decides locally that a DO MORE command should be output for Position _2 and executed, since this Position has not yet been picked up according to the processing state bz 1.

At reference numeral 4b, the control module SM2 of the second robot mounter BA2 locally decides that the DO LESS command should be output for Position _3 and executed because the Position has been mounted according to the transferred processing state bz 2.

If a delay then occurs again, time can be saved by means of a matching assembly task and the assembly can be carried out more efficiently overall.

Fig. 5 shows a flow chart of a method according to a preferred embodiment of the invention. After the method for local optimization has been started, assembly tasks each generated by the line optimizer LO for the pick-and-place robots BA are detected in step S1. In step S2, the processing state bz is detected by the respective preceding robot Bai-1 or read from the intermediate memory. In step S3, an adaptation algorithm is executed in order to provide the matching assembly task ba' as a result data set in step S4. In step S5, the robot mounter BA is operated with the matching mounting task BA'. In step S6, the current processing state bz is detected, which is sent to the subsequent automaton via direct machine-to-machine communication. The previous processing state can thus be received independently of the receiving of the assembly task. The corresponding automaton thus has more information and can therefore be optimized better locally. This results in improved performance.

Fig. 6 shows in a schematic representation the general sequence of the mounting process and in particular how the circuit boards (or printed circuit boards denoted by reference LP in fig. 6) are mounted in different pick-and-place machines BA. The time axis extends from top to bottom and is shown as a downward pointing arrow on the left. The conveyor belt TV conveys the printed circuit boards LP from the robot BA in a conveying direction TV (to the right in fig. 6) to the robot. The rows shown in succession represent successive time periods or time-disks t0, t1, t2, etc. For the sake of simplicity, the printed circuit board LP is divided into respective four quadrants for better understanding, which shall indicate the areas on the circuit board that shall be mounted by the four pick-and-place machines BA 1-4. In this case, the first robot mounter BA1 should mount the upper left quadrant, the second robot mounter BA2 mount the upper right quadrant, the third robot mounter BA3 mount the lower left quadrant and the fourth robot mounter BA4 mount the lower right quadrant. It should be noted that this is merely a very simplified schematic illustration and representation and that the principle process movements should be elucidated. In general, the positions to be assembled are not predefined in the form of quadrants, but rather by means of unambiguous position indications or position coordinates. Furthermore, the robots can also be equipped for the same or different component types. The first printed circuit board LP1 performs the process as the first one, followed by the other printed circuit boards LP2, LP3, and LP 4. The first printed circuit board is located at the fourth automatic motor BA4 at the period t 0. The last printed circuit board LP4 shown in the example in fig. 6 is located in the first cycle t0 at the first robot assembly BA1, in the second cycle t1 at the second robot BA2 and in the third cycle t2 at the third robot BA3 and in the fourth cycle t3 at the fourth robot BA 4.

In the example shown in fig. 6, the third robot BA3 now has a delay in the third period t2, so that the third robot cannot perform its initial assembly tasks in the required time (SOLL-Zeit). According to the invention, it is provided at this point that the control module SM3 of the third robot BA3 "decides" not to fit the position on the basis of the adaptation algorithm, so that the operation of the tape TV can be continued and the other robots of the line FL do not have to stop. This is possible because the location has CAN. In fig. 6, this is indicated by the unfilled circle. The fourth robot BA4 then receives information about the unassembled state via the transferred processing state bz3 and supplements this position on behalf of the former robot BA 3. This is possible because the fourth automatic assembly machine is accordingly ready for this purpose. This is represented in fig. 6 by a cross in a circle.

In principle, when a temporary delay occurs at a pick-and-place machine BA, the adaptation algorithm is then executed locally according to the invention in order to start MORE or LESS pick-ups by outputting a DO MORE or DO LESS command. This is possible because the respective subsequent machine state bz is informed to the respective subsequent machine. This is shown in fig. 6 at the top by the arrow with the reference bz. Therefore, according to the present invention, the waiting time (output DO MORE command) due to the delay of another pick-and-place machine BA in the line FL is advantageously utilized and the stop time can be avoided. According to the invention, if a delay occurs at one robot and the normal processing time (cycle) is not sufficient to mount all the positions, the other pick-and-place machines BA can also continue to mount. In this case, only the MUST locations are assembled and the SHOULD locations (for which the DO leave command is output) can be ignored for assembly by subsequent machines (so-called "make-up").

Finally it is pointed out that the description and embodiments of the invention are in principle not restricted to a particular physical solution of the invention. The features which are set forth and illustrated in combination with the individual embodiments of the invention can be combined differently in the object according to the invention in order to achieve their advantageous effects at the same time.

It is particularly clear to the person skilled in the art that the invention can be used not only for SMT technology but also for other assembly technologies that require an assembly process to be handled by a plurality of automatic assembly machines, for example machine-assisted THT technology (through hole technology). Furthermore, the control modules and elements of the system may be implemented distributed over a plurality of physical products. The present invention is also not limited to a particular circuit board mounting system style. The invention can thus also be applied in machines for assembling circuit boards that are moving but have a stationary head or machines without a feed unit and in which the components are located directly in the head. In principle, the invention can be used in all production lines with individual pick-and-place machines connected in series. The processing steps of a single machine must be able to be divided into smaller units. In this case, the smaller units must be able to be implemented by a plurality of individual machines of the line.

The scope of protection of the invention is not given by the following claims and is limited by the features set forth in the description or shown in the drawings.

Claims (19)

1. Method for controlling the mounting process of pick-and-place robots (BA) in a production line (FL), wherein printed circuit boards (LP) are mounted by means of a plurality of pick-and-place robots (BA 1, BA2, … BAn), wherein each pick-and-place robot (BA) performs a sub-task of a mounting plan, comprising the following method steps:

-reading (S1) the placement sequences (BA) generated for the respective pick-and-place machines (BA), which placement sequences (BA) are predetermined and which component types are to be placed at which locations on the printed circuit board (LP), wherein the placement sequences are made flexible by additionally determining for each location an adjustment parameter which contains a calculated SHOULD, CAN or MUST data set in order to specify which placement position SHOULD be performed by the respective pick-and-place machine (BA) in the case of the SHOULD data set, which placement position CAN be performed by the respective pick-and-place machine (BA) in the case of the CAN data set, and which placement position MUST be performed by the respective pick-and-place machine (BA) in the case of the MUST data set;

-detecting (S2) a processing state (bz) of the printed circuit board (LP) to be assembled;

-executing (S3) an adaptation algorithm to flexibly adapt the read assembly order (ba) based on the adjustment parameters and the detected machining state (bz) to calculate a matching assembly order (ba') to determine which assembly positions differ from the read assembly order (ba) for assembly and which assembly positions differ from the read assembly order (ba) for non-assembly;

-handling (S5) a pick-and-place machine (BA) with a matching placement sequence (BA');

-detecting (S6) the processing state (bz) of the printed circuit board (LP) after performing the matching assembly sequence (ba').

2. Method according to claim 1, wherein the processing state (bz) is directly transferred electronically to at least one subsequent pick-and-place machine (BA) in the production line.

3. The method according to claim 1 or 2, wherein the performing (S3) and the manipulating (S5) in the steps are repeatedly performed until a presettable interrupt condition is satisfied.

4. Method according to claim 1, wherein the adaptation algorithm is checked, whether a delay occurs on at least one of the pick-and-place machines (BA) of the production line (FL) and a matching placement sequence (BA') is calculated therefrom.

5. Method according to claim 1, wherein the processing state (bz) contains an electronic data set relating to more or less assembly by the respective automatic assembly machine (BA).

6. Method according to claim 1, wherein in the case of further assembly to be performed, a respective further assembly sequence is assigned to the respective pick-and-place machine (BA) following on the production line (FL).

7. Method according to claim 1, wherein the pick-and-place robots (BA) exchange data with one another via machine-to-machine communication and send the processing states (bz) to the respective subsequent pick-and-place robots (BA) in the production line.

8. The method of claim 7, wherein the machine-to-machine communication comprises whisper protocol and/or via IPC-letters-9852 protocol.

9. Method according to claim 1, wherein the SHOULD data set defines that the respective mounting location SHOULD be mounted by the respective pick-and-place machine (BA) in the absence of delays or in the absence of errors, and that no mounting SHOULD take place in the presence of delays or errors.

10. Method according to claim 1, wherein the CAN data set is defined such that the respective assembly position CAN be assembled by the respective pick-and-place machine in the event of a delay or an error.

11. Method according to claim 1, wherein the MUST data set defines that the respective mounting position has to be mounted by a respective pick-and-place machine (BA).

12. The method according to claim 1, wherein the effect-enhancing value is calculated for implementation of a fitting plan that appears to save time using the adaptation algorithm compared to a method without applying the adaptation algorithm.

13. Method according to claim 1, wherein a prediction data record is calculated which, taking into account the adaptation algorithm used, provides a prediction for better equipping the pick-and-place robots (BA) from an analysis of the already executed assembly plan.

14. Method according to claim 1, wherein in the case of delays or errors, the pick-and-place machine (BA) determines an estimate of the delay caused by the placement error and activates the use of the adaptation algorithm only if the determined estimate exceeds a threshold value which can be set in advance.

15. A computer-readable storage medium, wherein the computer-readable storage medium is configured to store a computer program executable by a processor to implement the steps in the method according to any one of claims 1-14.

16. Control module (SM) for a pick-and-place machine (BA), wherein the pick-and-place machine (BA) is used together with other pick-and-place machines (BA) for mounting printed circuit boards (LP) in a production line (FL), wherein the control module (SM) is configured to control the pick-and-place machine (BA) by means of a method according to any one of claims 1 to 14.

17. Control module (SM) according to claim 16, wherein the control module (SM) is configured with an interface with other control modules (SM) of other pick-and-place machines (BA).

18. Pick-and-place machine (BA) with a control module (SM), which is a control module (SM) according to claim 16 or 17.

19. System for controlling a production line (FL) for producing printed circuit boards (LP), the system having:

-a plurality of pick-and-place machines (BA) for mounting printed circuit boards (LP), wherein each pick-and-place machine (BA) executes a respective subtask of a mounting plan and wherein each pick-and-place machine (BA) has a control module (SM), which is a control module (SM) according to claim 16, which control module (SM) is electronic;

-a Line Optimizer (LO) for dividing the mounting plan into subtasks for assignment to the individual pick-and-place machines (BA) of the production line (FL) and for calculating in each case a flexible mounting sequence generated for the respective pick-and-place machine (BA), which mounting sequence has been set up in advance which component type (t) is to be mounted at which position on the printed circuit board (LP), wherein the mounting sequence (BA) is made more flexible by additionally determining for each position an adjustment parameter which contains a calculated SHOULD, CAN or MUST data set in order to specify which mounting position SHOULD be carried out by the respective pick-and-place machine (BA) in the case of a SHOULD data set, which mounting position CAN be carried out by the respective pick-and-place machine (BA) in the case of a MUST data set, specifying which assembly position must be performed by the respective pick-and-place machine (BA);

-data connections between the individual pick-and-place machines (BA) and the line optimization machine (LO).

Images