Production process, and production plant for manufacturing a product using used parts

Abstract

In order to manufacture a product (P) using used parts (UP), a statistical dispersion (DUP) of a production-relevant property (PUP) of the used parts (UP) is determined. Furthermore, a plurality of different process plans (PP) is generated, each of which is used to specify a combination (S) of process steps (PS1, PS2, ...) that includes processing of the used parts (UP). Each process step combination (S) is simulated, wherein an associated process quality (PQ) and its dispersion (DQ), caused by the dispersion (DUP) of the production-relevant property (PUP), are determined. Furthermore, for each process plan (PP), a quality value (QV) is derived from the relevant determined process quality (PQ) and its dispersion (DQ), wherein an increase in the dispersion (DQ) of the process quality (PQ) adversely affects the quality value (QV). Depending on the derived quality values (QV), a process plan (PPO) optimised with regard to the quality value is then selected from the generated process plans (PP). Finally, the product (P) is manufactured according to the process step combination (SO) specified by the selected process plan (PPO).

Description

[0001] Description

[0002] Production process and production plant for manufacturing a product using used parts

[0003] Regardless of the grammatical gender of a particular term, persons with male, female or other gender identities are included.

[0004] Modern production processes increasingly aim to recycle products, their components, or their materials at the end of their life cycle to manufacture new products, in line with the principles of a circular economy. Ideally, the reused resources should constitute the largest possible share of the total resources required.

[0005] However, the reuse of used parts often presents the problem that the parts are worn, damaged, or dirty, or at least show signs of use. These circumstances often make direct use of used parts difficult. As a result, additional resources are frequently required for pre-processing the used parts to ensure the specified properties of the product manufactured with them.

[0006] Typically, separate production lines are used for manufacturing a product from used parts and for manufacturing it from new parts. In practice, remanufacturing using used parts is often outsourced to specialized third-party manufacturers. However, in such scenarios, some production and / or logistics resources become redundant.

[0007] The object of the present invention is to specify a production process and a production plant that allow for more efficient recycling of used parts.

[0008] This problem is solved by a production process with the features of claim 1, by a production plant with the features of claim 12, by a computer program product with the features of claim 13, and by a computer-readable storage medium with the features of claim 14. In a production process according to the invention, carried out by a production plant for manufacturing a product incorporating used parts, a statistical variation of a production-relevant property of the used parts is determined. Furthermore, a multitude of different process plans are generated, each specifying a combination, in particular a sequence, of process steps for manufacturing the product that includes processing the used parts.The process step combination specified by each process plan is simulated, determining the respective process quality and its variation caused by the variance of the production-relevant property. Furthermore, a quality score for each process plan is derived from the determined process quality and its variance, with increasing variance in process quality negatively impacting the quality score. Based on the derived quality scores, a quality-optimizing process plan is then selected from the generated process plans. Finally, the product is manufactured according to the process step combination specified by the selected process plan.

[0009] To carry out the production process according to the invention, a production plant, a computer program product and a computer-readable, preferably non-volatile storage medium are provided.

[0010] The production process, the production plant and the computer program product according to the invention can be implemented or executed in particular by means of one or more computers, one or more processors, application-specific integrated circuits (ASICs), digital signal processors (DSPs), a cloud infrastructure and / or so-called "Field Programmable Gate Arrays" (FPGAs).

[0011] A particular advantage of the invention lies in the fact that used parts, whose properties often vary significantly as a result of their previous use, can be efficiently reused for the production of new products. In particular, the invention allows for the automated identification and application of process step combinations that are especially robust against fluctuations in the production-relevant properties of the used parts.

[0012] Advantageous embodiments and further developments of the invention are specified in the dependent claims. According to an advantageous embodiment of the invention, the production-relevant property can relate to the geometry, weight, homogeneity, flexibility, wear, soiling, contamination, damage, completeness, corrosion, functionality, quantity, feature, condition, and / or traces of use of the used parts. The aforementioned production-relevant properties and their respective variations can generally be determined in a simple manner.

[0013] In particular, the production facility can determine the production-relevant properties and their variations based on information about the respective condition of previously manufactured, used products. For this purpose, data collected during the life cycle of these previously manufactured products can be used. Such data is frequently collected as part of a life cycle analysis (LCA) or life cycle costing (LCC).

[0014] Alternatively or additionally, the production plant can measure the production-relevant properties of the used parts intended for manufacturing the product and, based on this, determine the variation in these properties. In this way, product manufacturing can be automatically adapted to changing properties of the used parts.

[0015] Furthermore, the production plant can be configured to capture a plant-specific rule for combining process steps. The generation of process plans can then be restricted to combinations of process steps corresponding to this rule. Since different process steps often cannot be combined arbitrarily, the above approach allows for the specification of requirements, constraints, or other framework conditions for production- or processing-related combinations of process steps, or for their adaptation to changing production situations or technical requirements. In many existing production plants, certain sub-combinations of process steps are predefined and cannot be changed.These specifications can be defined in machine-readable form using plant-specific rules for combining process steps and fed into a generator for generating process plans. The generator can then be instructed to create process plans in which one or more predefined sub-combinations remain constant, while only the remaining combinations are varied. Advantageously, the respective process quality can be used to quantify resource consumption, performance, waste volume, pollutant emissions, environmental impact, and / or compliance with production requirements for each process step combination. Performance can refer to production speed, yield, quantity, product quality, throughput time, and / or efficiency of the respective process step combination.Furthermore, resource consumption can affect production resources such as energy, production materials, computing resources, and / or tool wear. The above-mentioned characteristics of a production process can be determined with sufficient accuracy through simulation using process plans. A variety of efficient simulation methods are available for this purpose.

[0016] Alternatively or additionally, the respective process quality can be used to quantify the performance, product quality, environmental impact, recycling rate, lifespan, service life, and / or compliance with product requirements of the product simulated through the respective process step combination. Performance can refer to the product's power, efficiency, lifespan, susceptibility to wear, and / or functionality.

[0017] In particular, the respective process quality can be determined based on data about the life cycle of previously manufactured products. In this way, environmental impacts resulting from the use or transport of the products, such as CO2 emissions, recycling rates, lifespan, ecological footprint, and / or resource consumption due to product use, can be easily recorded and taken into account when optimizing the production process.

[0018] Process quality can preferably be determined as a weighted sum of several of the above properties of the production process, properties of the product, and properties of previously manufactured products.

[0019] According to a further advantageous embodiment of the invention, the process plans can be generated within the framework of a numerical optimization procedure, wherein one optimization goal of the procedure is to optimize the performance score. For this purpose, generation parameters of the generation process can be varied such that process plans with a high performance score are preferably generated. In addition, the optimization procedure can vary a respective process step parameter influencing the execution of a respective process step in such a way as to optimize the performance score.

[0020] In particular, the optimization method can include a genetic optimization method, a gradient-free optimization method, a discrete optimization method, a gradient descent method, and / or particle swarm optimization. In a genetic optimization method, the performance value calculation can be used as a genetic fitness function.

[0021] According to an advantageous further development of the invention, the process plans can comprise first process plans that include processing the used parts, and second process plans in which the processing of the used parts is replaced by processing new parts. In a first process step, the production-relevant property of an incoming used part can then be measured, and depending on this, either one of the first process plans or one of the second process plans can be selected as the quality-optimizing process plan for manufacturing the product. This allows for a quality-optimizing decision as to whether including a used part received during the ongoing manufacturing process is advantageous, or whether it is more advantageous to use a new part instead. In this way, less suitable used parts or...Used parts whose recycling in the current production process would require a disproportionate effort are efficiently sorted out or put to other uses.

[0022] An embodiment of the invention is explained in more detail below with reference to the drawing. The drawings illustrate, in schematic form:

[0023] Figure 1 shows a production process for manufacturing products using used parts,

[0024] Figure 2 shows different combinations of process steps for manufacturing products.

[0025] Figure 3 shows a process step, and Figure 4 shows a production process for the optimized manufacture of products using used parts.

[0026] Insofar as the same or corresponding reference symbols are used in different figures, these reference symbols denote the same or corresponding entities, which may be described, implemented or designed in particular as in connection with the figure in question.

[0027] Figure 1 illustrates a production process for manufacturing products P, in which used parts UP are used in addition to new parts and materials NP. The products P, for example, motors, their parts, intermediate products, electrical or mechanical assemblies, circuits, computers, robots, turbines, or other machines, are manufactured by a production plant PA from the new parts and materials NP, the used parts UP, and, if applicable, other raw materials.

[0028] Depending on the product being manufactured, used parts (UP) can include components, assemblies, and especially previously manufactured, used products of the same or similar type as the products P, or parts of these used products. Such used parts for production are often also referred to as secondary materials. In contrast, new or new raw materials are often referred to as primary materials.

[0029] In this exemplary embodiment, the production plant PA comprises several production machines PM1, PM2, PM3, PM4, and PM5, which, optionally together with human workers, each perform one or more process steps PS1, PS2, PS3, PS4, and PS5. For the sake of clarity, only five production machines, PM1 to PM5, are shown in Figure 1 as examples, each performing one of the associated production steps PS1, PS2, PS3, PS4, and PS5, respectively. Of course, the production plant PA can have a larger number of production machines, which can be arranged and networked in a considerably more complex manner. Accordingly, the process steps to be performed can be combined in a much more complex way than illustrated in Figure 1.

[0030] Production machines PM1 to PM5 can include, for example, machine tools, robots, milling machines, welding robots, conveyor belts, industrial furnaces, and / or shredders. Process steps PS1 to PS5 can include, in particular, machining, processing, assembling, mixing, separating, disassembling, heating, melting, shredding, and / or transporting parts, materials, or workpieces.

[0031] According to the present embodiment, the used parts UP are fed to production machine PM1 and processed by it in process step PS1, for example, disassembled. Similarly, the new parts and materials NP are fed to production machine PM2 and processed by it in process step PS2. Process steps PS1 and PS2 each output intermediate products (not explicitly shown), which are fed to production machine PM3 for further processing in process step PS3. The additional intermediate products output by process step PS3 are fed to production machine PM4 and processed there in process step PS4 into further intermediate products, which in turn are fed to production machine PM5 for processing into the final products P in process step PS5.

[0032] The finished products P are ultimately dispensed by the production plant PA and put into use by users U. The products P then go through their life cycle with the users U.

[0033] Preferably, data on the products P used are collected throughout their life cycle, in particular regarding service life, reliability, susceptibility to failure, wear and tear, CO2 balance, environmental impact, resource consumption, ecological footprint, and / or recycling rate. This data can then be used for a life cycle analysis (LCA) or life cycle costing (LCC). In the present embodiment, this data is used, as explained in more detail below, particularly for optimizing the production process.

[0034] As mentioned above, the aim, in line with the principles of a circular economy, is to recycle as large a proportion as possible of used products P at the end of their life cycle. To this end, as many used products P as possible are reintroduced into the production process as secondary material in the form of used parts UP after their use, in order to be processed as described above. In this way, a closed material cycle or recycling loop is implemented.

[0035] Furthermore, the aim is to ensure that the production process is as robust as possible against the often unavoidable fluctuations in production-relevant properties of the recycled used parts (UP). In particular, it should be ensured that the products P manufactured in this way meet specified requirements. In addition, it may be desirable for the production process and / or the life cycle of the manufactured products to have the smallest possible ecological footprint.

[0036] For this purpose, various combinations of process steps executable by the PA production plant are simulated within an optimization process to find a combination that is particularly robust against fluctuations in production-relevant properties of the used parts and simultaneously exhibits the highest possible process quality. Process quality can refer to performance, quality, and / or the ecological footprint of the production process and / or the product.

[0037] Figure 2 illustrates some such different process step combinations S1, S2, and S3 for the production of a product P. For clarity, Figure 2 explicitly shows only process step combinations that start with used parts UP and have no branches or junctions. In practice, such process step combinations can, of course, be more complex.

[0038] Through the process step combination S1, the used parts UP, as shown in Figure 1, are first fed to process step PS1. The resulting intermediate products are then further processed into product P through process steps PS3, PS4, and PS5.

[0039] In contrast, with process step combination S2, the used parts UP are fed into process step PS6. The resulting intermediate products are then further processed into product P via process steps PS3, PS4, and PS5, as with process step combination S1. Process step PS6, which is new compared to process step combination S1, can, for example, include a cleaning process. In such a case, process step combination S2 is preferable if a high or highly fluctuating degree of contamination of the used parts UP is expected.

[0040] Finally, through process step combination S3, the used parts UP are fed to process step PS7, in order to be further processed into product P through process steps PS8, PS9, and PS5. Process step PS7 could, for example, include shredding, process step PS8 a melting process, and process step PS9 a material separation. In such a case, process step combination S3 is preferable if structure-preserving processing of the used parts UP proves inefficient in the simulation.

[0041] Figure 3 illustrates a general process step PS, representing the process steps PS1, ..., PS9 mentioned above. As mentioned above, process step PS can include, in particular, machining, processing, assembling, mixing, separating, disassembling, heating, melting, shredding, and / or transporting parts, materials, or workpieces.

[0042] The PS process step receives input products IP and processes them. The output products OP are then generated as a result of this processing. If the PS process step is executed at the beginning of a process step combination, then, in this exemplary embodiment, used parts UP or new parts NP are fed into this PS process step as input products. Similarly, if the PS process step is executed at the end of a process step combination, then in this exemplary embodiment, finished products P are generated as output products OP.

[0043] A specific execution of process step PS is parameterized by process step parameters PSP and controlled based on these process step parameters PSP. The process step parameters PSP can specify, in particular, a design, a milling geometry, a placement pattern, an oven temperature, a mixing ratio, a conveyor belt speed, and / or other processing-relevant parameters of process step PS.

[0044] During the execution of process step PS, production resources PR continue to be consumed, such as energy, auxiliary materials, tools or computing resources.

[0045] Furthermore, the execution of the PS process step generates waste products AE, such as chips, contaminants, wastewater and / or emissions.

[0046] As a rule, the successful execution of a process step (PS) requires that certain properties of the input products (IP) lie within predefined first limits. Similarly, the successful execution of process step PS should generally ensure that certain properties of the output products (OP) lie within predefined second limits. Such production-relevant properties can relate, for example, to geometry, weight, homogeneity, flexibility, wear, soiling, contamination, damage, completeness, corrosion, functionality, condition, and / or traces of use of the input products (IP) or the output products (OP). Such production-relevant properties can usually be easily measured or otherwise determined by the production system (PA).

[0047] In the present embodiment, such a production-relevant property of the starting products IP is denoted by PI. The production-relevant property PI fluctuates around a mean value with a range of fluctuations quantified by a statistical dispersion DI. It is assumed that the fluctuating production-relevant property PI generally lies within the predefined first limits.

[0048] Similarly, let PO be a production-relevant property of the output products OP. This production-relevant property PO fluctuates around a mean value with a range of fluctuations quantified by a statistical dispersion DO.

[0049] The statistical variations DI and DO can each be specified by a variance, a standard deviation, a confidence interval, a probability distribution, and / or a distribution type. The statistical variations DI and DO can usually be measured or determined easily.

[0050] As already partially indicated above, the aim is to find a combination of process steps whose process steps, in total, require as few production resources (PR) as possible, generate as few waste products (AE) as possible, have the highest possible performance, and are as robust as possible against fluctuations in the product-relevant properties of the used parts (UP).

[0051] For this purpose, a large number of different process step combinations are generated within an optimization procedure and evaluated by means of a simulation. Furthermore, process step parameters of the process steps within the generated process step combinations are also varied during the optimization. The above procedure is explained in more detail below with reference to Figure 4. Figure 4 illustrates a production process carried out by a production plant PA for the optimized manufacture of products P, incorporating used parts UP as well as new parts and materials NP.

[0052] To determine production-relevant properties of the used parts (UP) to be processed, as well as the respective statistical variation of these properties, the PA production plant is equipped with an AM analysis module. As mentioned above, the production-relevant properties can include, for example, geometry, weight, homogeneity, flexibility, wear, soiling, contamination, damage, completeness, corrosion, functionality, condition, and / or signs of use of the used parts (UP). For the sake of clarity, only a single production-relevant property (PUP) and its statistical variation (DUP) will be explicitly considered below.

[0053] In this embodiment, the AM analysis module measures the production-relevant property PUP of a large number of used parts UP. Alternatively or additionally, the AM analysis module can acquire data LCD collected during the respective life cycle of previously manufactured products regarding the production-relevant property PUP. From the measurement results and / or the acquired life cycle data LDC, the AM analysis module then determines the statistical variation DUP of the production-relevant property PUP. Elementary statistical methods are available for determining such a statistical variation DUP.

[0054] The statistical variance DUP, along with a mean value of the production-relevant property PUP, is transmitted to a simulator SIM of the production plant PA. For clarity, this mean value is also labelled PUP in Figure 4.

[0055] The SIM simulator is used to simulate production processes of the PA production plant. For this purpose, the SIM simulator has an assigned process-step-specific simulation model SM1, SM2, ... for each of the process steps PS1, PS2, ... described above. As already indicated in connection with Figure 3, the execution of each process step PS1 or PS2, ... is parameterized and controlled by a respective process step parameter PSP1 or PSP2, .... Using each process-step-specific simulation model, the SIM simulator can quantitatively simulate the respective process step PS1 or PS2, ... depending on its process step parameters PSP1 or PSP2, ... and the capabilities of the executing production machines.Such process-step-specific simulation models are available for a wide variety of industrial process steps and allow for efficient and realistic simulation of these process steps. In particular, the SIM simulator can include a so-called digital twin of the production plant PA.

[0056] The production plant PA also has a generator GEN for generating process plans PP, each of which specifies a combination S of process steps PS1, PS2, ... . Furthermore, each process plan PP specifies at least one process step-specific parameter for preferably each of the process steps PS1, PS2, ... . The set of process step parameters generated by the generator GEN is denoted by the reference symbol PSP in Figure 4.

[0057] In practice, only a few of all possible combinations of process steps are technically feasible or even practical from a production standpoint. Which process steps are meaningful or even combinable with which other process steps is generally determined by predefined, well-known rules. Accordingly, the generator GEN stores corresponding rules R in machine-readable form. These rules R cause the generator GEN to generate only process plans PP that specify combinations S of process steps PS1, PS2, ... corresponding to the rules R. In particular, the generator GEN can generate rule-compliant process plans PP using a so-called production plant planner.

[0058] The generator GEN produces a multitude of process plans PP for many variations of process step combinations S, whereby the process step parameters PSP1, PSP2, ... of each involved process step PS1, PS2, ... are also varied. The multitude of generated process plans PP are transmitted from the generator GEN to the simulator SIM.

[0059] The simulator SIM captures the process step combination S specified by a given process plan PP and assigns the corresponding simulation model SM1 or SM2, ... to each of its process steps PS1, PS2, ... . The assigned simulation models SM1, SM2, ... are then combined in the same way as the process steps PS1, PS2, ... of the respective process step combination S. Furthermore, the simulation models SM1, SM2, ... are parameterized by the process step parameters PSP1, PSP2, ... specified in the respective process plan PP. Using these combined and parameterized simulation models SM1, SM2, ..., the simulator SIM then quantitatively simulates the respective process step combination S based on the process step parameters PSP.In this process, used parts (S) are simulated to be supplied to the beginning of the respective process step combination, and products (S) are simulated to be output to the end of this process step combination.

[0060] The simulation takes into account the production-relevant property PUP of the used parts UP and the statistical variation DUP of this property PUP. Based on this, the simulator SIM determines a process quality PQ and a statistical variation DQ of this process quality PQ, caused by the variation DUP, as simulation results for a given process plan PP.

[0061] The variation DQ can be determined, for example, by randomly varying the production-relevant property PUP according to its variation and running a simulation for each variation, determining a value for the process quality PQ each time. The variation DQ can then be easily calculated from the resulting process quality values. This approach is often referred to as sampling.

[0062] The respective process quality PQ can relate in particular to production resource consumption, energy consumption, material consumption, tool wear, performance, product quality, production speed, yield, efficiency, waste quantity, pollutant emissions, environmental impact, compliance with production requirements and / or other production-relevant parameters of the respective process step combination S.

[0063] Alternatively or additionally, the respective process quality (PQ) can also relate to performance, product quality, power, efficiency, wear resistance, functionality, environmental impact, recycling rate, service life, usage duration, compliance with product requirements, and / or other product-relevant parameters of the product simulated by the respective process step combination (S). The aforementioned product-relevant parameters can be determined, in particular, using the life cycle data (LCD), which can be supplied to the simulator (SIM) for this purpose. Preferably, the respective process quality (PQ) is calculated by the simulator (SIM) as a weighted sum of at least some of the above production-relevant and / or product-relevant parameters.

[0064] The determined process quality (PQ) and its variance (DQ) are transmitted from the simulator (SIM) to a quality evaluator (QEV). The latter derives a corresponding quality value (QV) from the respective process quality (PQ) and its variance (DQ). Since—as already mentioned above—a production process is desired that is as robust as possible against fluctuations in production-relevant properties of the used parts (UP), the variance (DQ) of the process quality should be relatively low. For this reason, the quality value (QV) is calculated such that it increases with increasing process quality (PQ) or decreasing variance (DQ), and conversely decreases with decreasing process quality (PQ) or increasing variance (DQ).

[0065] In particular, the respective quality value QV can be calculated as a weighted sum of the process quality PQ and the variation DQ, where the process quality PQ is weighted positively, e.g. 1, and the variation DQ is weighted negatively, e.g. -1.

[0066] The respective quality value QV is assigned to the corresponding process plan PP for which it was calculated and transmitted by the quality evaluator QEV to an optimization module OPT of the production plant PA. Simultaneously, the respective process plan PP is transmitted to the optimization module OPT by the generator GEN.

[0067] The optimization module OPT executes a numerical optimization procedure aimed at maximizing the efficiency value QV. To this end, the OPT module instructs the generator GEN – as indicated by a dashed arrow in Figure 4 – to vary the process step combinations S and the process step parameters PSP in such a way that the resulting efficiency values ​​QV increase, at least on average. A variety of standard optimization procedures are available for such optimization procedures.

[0068] Preferably, a genetic optimization method can be used. In this method, a population of different variations of process step combinations S and process step parameters PSP is considered over several generations. Variations for which a comparatively high quality value QV is determined through simulation are preferentially allowed to reproduce. In this case, it can be expected that the quality value QV will increase, at least on average, in subsequent generations.

[0069] Alternatively or additionally to a genetic optimization method, particle swarm optimization or another gradient-free optimization method can also be used. Specifically for varying the process step parameters (PSP), a gradient descent method can also be employed.

[0070] The optimization procedure used can be terminated when the quality value QV reaches a threshold value specified for the production of products P.

[0071] After successful completion of the optimization process, the optimization module OPT selects from the generated process plans PP the process plan PPO for which the highest quality value QV was determined during the optimization process. The quality-value-optimizing process plan PPO specifies an optimized process step combination SO and optimized process step parameters PSPO for the process steps of the optimized process step combination SO.

[0072] After the selection of the quality-optimizing process plan PPO, the products P are finally manufactured by the production plant PA according to the process step combination SO parameterized by the process step parameters PSPO.

[0073] For the present embodiment, it is assumed that the optimized process step combination SO corresponds to the combination shown in Figure 1. In this case, used parts UP are processed into products P by the process step sequence PS1, PS3, PS4 and PS5, and new parts NP are processed by the process step sequence PS2, PS3, PS4 and PS5.

Claims

Patent claims 1. Production process for manufacturing a product (P) using used parts (UP), wherein a production plant (PA) a) determines a statistical variation (DUP) of a production-relevant property (PUP) of the used parts (UP), b) generates a large number of different process plans (PP), each of which comprises a combination (S) of process steps (PS1, PS2,...) encompassing the processing of the used parts (UP).) for the production of the product (P), c) the process step combination (S) specified by a respective process plan (PP) is simulated, whereby a respective process quality (PQ) and its variation (DQ) caused by the variation (DUP) of the production-relevant property (PUP) are determined, d) a quality value (QV) for the respective process plan (PP) is derived from the respective determined process quality (PQ) and its variation (DQ), whereby an increasing variation (DQ) of the process quality (PQ) negatively influences the quality value (QV), e) depending on the derived quality values ​​(QV), a quality value-optimizing process plan (PPO) is selected from the generated process plans (PP), and f) the product (P) is produced according to the process step combination (SO) specified by the selected process plan (PPO).

2. Method according to claim 1, characterized in that the production-relevant property (PUP) relates to a geometry, weight, homogeneity, flexibility, wear, soiling, impurity, damage, completeness, corrosion, functionality, number, feature, condition and / or traces of use of the used parts (UP).

3. Method according to one of the preceding claims, characterized in that the production plant - the production-relevant property (PUP) and its variation (DUP) are determined based on information (LCD) about the respective condition of previously manufactured, used products, and / or - the production-relevant property (PUP) of the used parts (UP) intended for the manufacture of the product (P) is measured and, depending on this, the variation (DUP) of the production-relevant property (PUP) is determined.

4. Method according to one of the preceding claims, characterized in that the production plant (PA) captures a rule (R) specific to the production plant (PA) for combining process steps (PS1 , PS2,...) and that the generation of the process plans (PP) is limited to combinations of process steps (PS1 , PS2, ...) corresponding to this rule (R).

5. Method according to one of the preceding claims, characterized in that the respective process quality (PQ) quantifies resource consumption, performance, waste quantity, pollutant emissions, environmental impact and / or compliance with production requirements of the respective process step combination (S).

6. Method according to one of the preceding claims, characterized in that the respective process quality (PQ) quantifies the performance, product quality, environmental impact, recycling rate, lifetime, service life and / or compliance with product requirements of the product (P) simulated by the respective process step combination (S).

7. Method according to one of the preceding claims, characterized in that the respective process quality (PQ) is determined on the basis of information (LCD) about a life cycle of previously manufactured products.

8. Method according to one of the preceding claims, characterized in that the process plans (PP) are generated within the framework of a numerical optimization procedure (OPT), wherein an optimization goal of the optimization procedure (OPT) is directed to optimize the quality value (QV).

9. Method according to claim 8, characterized in that the process step parameter (PSP1 , PSP2, ...) influencing the execution of a respective process step (PS1, PS2, ...) by the optimization method (OPT) is varied in such a way that the quality value (QV) is optimized.

10. Method according to claim 8 or 9, characterized in that the optimization method (OPT) comprises a genetic optimization method, a gradient-free optimization method, a discrete optimization method, a gradient descent method and / or a particle swarm optimization.

11. Method according to one of the preceding claims, characterized in that the process plans (PP) comprise first process plans comprising processing of the used parts (UP), and second process plans in which the processing of the used parts (UP) is replaced by processing of new parts (NT), and that in a first process step the production-relevant property (PUP) of an incoming used part (UP) is measured and, depending on this, either one of the first process plans or one of the second process plans is selected as a quality-optimizing process plan (PPG) for the production of the product (P).

12. Production plant (PA) for manufacturing a product (P) using used parts (UP), comprising means for carrying out all steps of a method according to any of the preceding claims.

13. Computer program product comprising instructions which, when the program is executed by a computer, cause a production plant (PA) according to claim 12 to execute a method according to any one of claims 1 to 11.

14. Computer-readable storage medium with a stored computer program product according to claim 13.

Images