Method and computer program product for providing and determining a parts list for a life-cycle assessment of a product

Abstract

The invention relates to a computer-implemented method (100) for determining and providing a parts list for a life-cycle assessment of a product, wherein the parts list comprises at least one or more components and / or sub-components of the product. The method is characterized by the following steps: - (101) generating a plurality of prompts for an artificial neural network, the plurality of prompts comprising at least one prompt which prompts the artificial neural network to determine the components and / or sub-components of the product having product-specific technical information relating to the components and / or sub-components, the determined technical information being used for the life-cycle assessment of the product; - (102) controlling the artificial neural network using the prompts in order to determine the parts list; and - (103) providing the parts list for the life-cycle assessment of the product, the parts list comprising the determined components and / or sub-components of the product together with the respective determined technical information relating thereto. The invention further relates to a corresponding computer-implemented method for the life-cycle assessment of a product and to a computer program product.

Description

[0001] 202501871

[0002] 1

[0003] Description

[0004] Method and computer program product for providing and determining a bill of materials for a product life cycle analysis

[0005] The present invention relates to the provision and determination of bills of materials for life cycle analyses of products, in particular to a computer-implemented method for providing and determining a bill of materials for a life cycle analysis of a product according to claim 1, a computer-implemented method for a life cycle analysis of a product according to claim 14 and a computer program product according to claim 15.

[0006] A life cycle assessment (LCA) is a method for evaluating and analyzing the environmental impacts associated with all phases of a product's life cycle, from raw material extraction through manufacturing, distribution, and use to disposal and / or recycling. An LCA is therefore crucial for analyzing and minimizing the ecological footprint of a product and / or process.

[0007] A key component of a life cycle assessment (LCA) is the bill of materials (BOM). The BOM typically lists all materials, components, and subcomponents. It serves as the basis for determining the environmental impacts associated with each element of the product throughout its entire life cycle. Therefore, a BOM is technically necessary for conducting an LCA. An LCA is technically necessary for determining the product's environmental impacts.

[0008] Typically, compiling a comprehensive bill of materials for LCA purposes is a time-consuming and resource-intensive process. For example, challenges arise in obtaining detailed technical information about materials and processes, particularly regarding complex supply chains and / or proprietary components from suppliers. This lack of available data can lead to incomplete and / or inaccurate LCAs and potentially reduce the effectiveness of sustainability initiatives.

[0009] Furthermore, the manual method of data collection and compilation for bills of materials (BOMs) carries the risk of human error and inconsistencies. Typically, significant time is invested in reviewing and analyzing scientific literature, industry reports, and expert knowledge to fill gaps in the BOM data for life cycle assessment (LCA).

[0010] 2

[0011] This process is inefficient and prone to fluctuations in the quality and completeness of the BOM or LCA.

[0012] Another technical challenge in creating bills of materials for LCAs is ensuring that all relevant environmental aspects are captured without double-counting impacts. This requires a careful balance between detail and precision, as well as a thorough understanding of the product lifecycle and the interrelationships between different components and processes.

[0013] With the increasing complexity of products and ever more complicated global supply chains, the technical task of determining comprehensive and accurate bills of materials for LCAs has become significantly more difficult.

[0014] The present invention is based on the technical problem of providing a more efficient method for determining a bill of materials for a life cycle analysis of a product.

[0015] The problem is solved by a computer-implemented method with the features of independent claim 1, by a computer-implemented method for life cycle analysis with the features of independent claim 14, and by a computer program product with the features of independent claim 15. The dependent claims specify advantageous embodiments and further developments of the invention.

[0016] The computer-implemented method according to the invention for determining and providing a bill of materials for a life cycle analysis of a product, wherein the bill of materials comprises at least one or more components and / or subcomponents of the product, is characterized by at least the following steps:

[0017] - Generating multiple prompts for an artificial neural network, wherein the multiple prompts include at least one prompt that instructs the artificial neural network to identify the components and / or subcomponents of the product with product-specific technical information of the components and / or subcomponents, wherein the identified technical information is used for the life cycle analysis of the product;

[0018] - Controlling the artificial neural network using the prompts to determine the bill of materials; and

[0019] - Providing the bill of materials for the product life cycle analysis, wherein the bill of materials includes the identified components and / or subcomponents of the product with their respective identified technical information. 202501871

[0020] 3

[0021] In particular, the bill of materials and / or the technical information will be used exclusively for a life cycle analysis of the product.

[0022] Life cycle analysis can relate to or constitute a part of a comprehensive life cycle analysis. For example, the life cycle analysis within the meaning of the invention relates only to the manufacture of the product and / or its end of life. A full life cycle analysis is not required according to the invention, but can be provided for. In other words, it is sufficient if the bill of materials is used for at least part of a life cycle analysis of the product.

[0023] In the context of the present invention, the following terms are defined:

[0024] "Bill of materials" refers to a comprehensive data structure or data on components, subcomponents, materials, and / or related technical information required or used for the production and / or assembly of a product. The bill of materials, as defined in the invention, serves as the basis for a life cycle assessment (LCA) of the product, which is technically necessary for the analysis of the product's environmental impact throughout its entire life cycle. It is therefore not to be understood in a purely commercial, conceptual, or purely informational sense, but rather as a technical requirement for determining the LCA. The bill of materials is thus the fundamental technical prerequisite for a product LCA.

[0025] Life cycle assessment (LCA) is a systematic, technical process for determining the environmental impacts typically associated with all phases of a product's life cycle, from raw material extraction through manufacturing, distribution, and use to disposal and / or recycling. LCA is therefore an essential technical tool for professionals to ensure the sustainable production of a product.

[0026] An "artificial neural network" (also known as artificial intelligence or AI) is a computer system capable of executing complex tasks based on prompts. In this context, the artificial neural network (Kl or Kl model) is specifically a large language model used to process prompts and generate comprehensive bill of materials information.

[0027] "Prompts" refer to specially designed instructions, inputs, and / or queries intended to elicit output from an artificial neural network. In the context of the invention, these prompts are structured to guide the neural network in the 202501871

[0028] 4

[0029] The creation of the bill of materials is controlled for the purposes of a life cycle assessment (LCA). In this sense, they represent – ​​comparable to a classical control system – the control inputs for the artificial neural network.

[0030] The method according to the invention has several advantages:

[0031] Efficiency and time savings: The use of artificial intelligence prompts (AI prompts) and an AI model reduces the time required to generate a comprehensive bill of materials (BOM). Traditional methods typically involve time-consuming manual research, data collection, and compilation. The AI-powered approach automates and standardizes these processes, making them more technical. This enables more efficient, faster, and improved BOM generation for life cycle assessment (LCA).

[0032] Improved accuracy and consistency: By using a computer logic model controlled by specific, well-defined prompts, the invention minimizes human error and inconsistencies that occur with manual data collection and compilation. The method ensures that all necessary technical information is systematically queried and included in the bill of materials, leading to more reliable and consistent results across different products and life cycle assessments (LCAs).

[0033] Improved data completeness: The KL prompts are designed to capture comprehensive information about components and / or subcomponents, as well as their specific technical details, required for LCA. This systematic approach ensures that no critical or LCA-essential technical information is overlooked.

[0034] Improved data accessibility: The invention offers a solution to the problem of limited access to proprietary information from suppliers and / or complex supply chains. By generating synthetic, yet technically realistic data based on available knowledge and patterns, the AI ​​model can fill information gaps that would otherwise be difficult or impossible to obtain using conventional methods.

[0035] Scalability and adaptability: The process can be effectively scaled to handle products of varying complexity and can be adapted to product types by adjusting the KL prompts. This flexibility enables consistent bill of materials generation across a wide range of products and sectors.

[0036] 5

[0037] Reproducibility: The use of a defined set of logic prompts and a controlled, managed logic model ensures that the bill of materials (BOM) determination process is reproducible. This is a significant advantage over known methods that rely on individual expert knowledge and / or inconsistent manual research techniques.

[0038] Integration of current information: AI models, especially large language models, can be trained on vast amounts of current data. This allows the generated bills of materials to potentially contain the latest information about materials, processes, and environmental impacts, which may not be readily available through traditional research methods.

[0039] Reduced resource requirements: By automating the determination of the bill of materials, the invention reduces the need for extensive personnel resources that are typically required for manual data entry and compilation.

[0040] Standardization potential: The structured approach to determining bills of materials (BOMs) using KL prompts creates the opportunity to standardize BOM determination for life cycle assessments (LCAs). This leads to more comparable and consistent LCA results, enabling a comparable assessment of environmental impacts.

[0041] Facilitating iterative improvements: The speed and consistency of AI-supported methods enable rapid iterations in bill of materials (BOM) creation. This allows for quick refinements and updates to the BOM as new information becomes available and / or the product design evolves, thus supporting a more agile and responsive LCA process.

[0042] The computer-implemented method according to the invention for a life cycle analysis of a product, wherein the life cycle analysis is based on a bill of materials of the product, is characterized in that the bill of materials of the life cycle analysis is created using a method according to one of claims 1 to 13.

[0043] The inventive method for determining the list of parts offers similar, equivalent and equivalent advantages.

[0044] The computer program product according to the invention is characterized in that it comprises instructions which, when executed by a computing unit, in particular a computer, cause it to execute a method and / or steps of the method according to any one of claims 1 to 14.

[0045] 6

[0046] The inventive method for determining the list of parts offers similar, equivalent and equivalent advantages.

[0047] In a preferred embodiment of the invention, the prompts include a prompt for creating an initial bill of materials structure.

[0048] This provides a technical basis for generating the bill of materials (BOM). The structured template allows the KL model to determine the necessary information more efficiently and accurately. The initial structure serves as a guideline, ensuring that all essential information is included from the outset. This reduces errors and omissions in the generated BOM. Furthermore, the initial structure can be adapted to different product types and / or industry-specific requirements, increasing the flexibility and applicability of the process.

[0049] In a preferred embodiment of the invention, the initial bill of materials structure comprises columns for a name of the components and / or subcomponents, their hierarchy, quantity, weight, volume, energy and / or resource consumption and / or their physical unit.

[0050] This specific structure ensures that all information beneficial for a life cycle assessment (LCA), particularly technically relevant information, is systematically captured and determined. Considering hierarchical information provides insights into the product composition, from main components to subcomponents. Quantity, weight, and / or volume data are advantageous for accurately calculating environmental impacts. Information on energy and resource consumption can be directly incorporated into the life cycle analysis, providing valuable insights into the product's ecological footprint. Alternatively or additionally, columns could be used for procurement information, manufacturing processes, and / or material specifications. These can be designed according to the specific requirements of the life cycle analysis.

[0051] In a preferred embodiment of the invention, the prompts include a prompt that ensures that the bill of materials includes only one material for each component and / or subcomponent.

[0052] This improves the accuracy and clarity of the bill of materials. This is because it ensures that each component or subcomponent is identified with a 202501871

[0053] 7

[0054] The material is linked to a single component, thus avoiding ambiguities and potential errors in material allocation. This is particularly advantageous for life cycle assessments, where precise material information is essential for calculating environmental impacts. Furthermore, the bill of materials-based LCA is improved, as this design facilitates the unambiguous tracking of each component's environmental footprint. Alternatively or additionally, a hierarchical material breakdown could be included for complex components, listing the main material at the component level and providing more detailed material compositions at the sub-component level.

[0055] In a preferred embodiment of the invention, prompts include a prompt for adding information on on-site manufacturing and / or assembly to the bill of materials.

[0056] Including information on on-site manufacturing and assembly provides a more comprehensive view of the product's life cycle. This data is particularly beneficial for accurately assessing the environmental impact of the production phase, which can be a significant contribution. Include this information in the bill of materials (BOM) to ensure that as many relevant processes as possible are considered in the life cycle analysis. This can include energy consumption during assembly, on-site waste, and / or specific manufacturing techniques that may influence the product's environmental footprint.

[0057] In a preferred embodiment of the invention, the prompts include a prompt for adding material and / or processing step information to the bill of materials.

[0058] This adds further, technically beneficial data / information to the bill of materials for life cycle assessment (LCA). Considering material information enables a detailed analysis of the environmental impacts of raw material extraction and processing. Information on processing steps provides insights into the energy consumption, emissions, and resource use associated with transforming raw materials into finished components. This level of detail allows for a more accurate modeling of the product's environmental footprint across its entire life cycle.

[0059] In a preferred embodiment of the invention, the prompts include a prompt for checking and eliminating double counting of production and

[0060] Processing steps. 202501871

[0061] 8

[0062] This advantageously improves the accuracy of the life cycle analysis.

[0063] Double counting can significantly distort results, leading to an overestimation of environmental impacts and thus technically incorrect results. By implementing a specific check for this problem, the method increases the reliability of the generated or determined bill of materials. The aforementioned prompt could further include cross-referencing manufacturing steps across various components and subcomponents to identify and eliminate redundancies. Alternatively or additionally, an automated reconciliation process could be included that not only identifies potential double counting but also suggests corrections and / or consolidations of manufacturing steps. This would further optimize or improve the generated bill of materials.

[0064] In a preferred embodiment of the invention, the artificial neural network is designed as a large language model.

[0065] In other words, a large language model is preferably used as an artificial neural network.

[0066] Using a large language model offers several advantages for creating a technically comprehensive bill of materials. The AI ​​models mentioned have been trained on massive datasets, enabling them to analyze, understand, and generate various text and text formats across a wide range of topics, including technical and industrial fields. This allows the AI ​​model to derive and generate detailed information about components, materials, and / or processes, even in the absence of specific data. Large language models also excel at understanding context and nuance, which is crucial for interpreting and responding to complex prompts related to product composition and manufacturing processes.Alternatively or additionally, specialized domain-specific models trained on engineering and manufacturing data could be used, and / or a hybrid approach combining a large language model with more focused, industry-specific AI models.

[0067] In a preferred embodiment of the invention, the prompts include a prompt for adding information on the disposal and / or recycling of the product to the bill of materials.

[0068] Including end-of-life information in the bill of materials is advantageous for a comprehensive life cycle analysis. This data enables the evaluation of the 202501871

[0069] 9

[0070] The environmental impact of the product beyond its service life, particularly considering technical characteristics such as recyclability, biodegradability, and / or reuse potential. By incorporating this information at the component level, the method enables more accurate modeling of various end-of-life scenarios and their associated environmental impacts. This can influence design decisions aimed at improving the overall sustainability of the product.

[0071] In a preferred embodiment of the invention, the method includes validating the determined bill of materials with regard to standards for life cycle analysis.

[0072] This validation step ensures that the generated bill of materials (BOM) meets the stringent requirements of standardized life cycle analyses. By comparing the generated BOM with established standards, such as ISO 14040 and / or ISO 14044, the method increases the credibility and comparability of the resulting life cycle analysis. This can include checks for completeness, data quality, and compliance with specific reporting formats. The validation process helps identify gaps and / or inconsistencies in the BOM that could negatively impact the accuracy of the life cycle analysis.

[0073] In a preferred embodiment of the invention, the method comprises an iterative refinement of the bill of materials based on feedback from an automated validation system, wherein the automated validation system validates the bill of materials with regard to the life cycle analysis of the product.

[0074] This iterative refinement process improves the accuracy and completeness of the bill of materials (BOM). By incorporating automated feedback, the method can quickly identify and correct inconsistencies, missing information, and / or potential errors. This dynamic approach enables continuous improvement of the BOM and ensures that it remains up-to-date and aligned with the specific requirements of lifecycle analysis. The automated validation system could check factors such as data consistency, completeness of material and process information, and compliance with lifecycle analysis guidelines. Alternatively or additionally, a machine learning component could be used that learns from past validations to improve future BOM generations.

[0075] In a preferred embodiment of the invention, the prompts are dynamically generated based on a product type and specific requirements of the life cycle analysis.

[0076] 10

[0077] This dynamic approach to prompt generation enables a highly customized and efficient bill of materials (BOM) creation process. By adapting prompts to the specific product type and life cycle analysis requirements, the method ensures that all technically relevant information is captured while minimizing unnecessary data collection. This can lead to more accurate and focused BOMs and improve the subsequent life cycle analysis process. Dynamic generation can take into account factors such as industry sector, product complexity, and specific environmental impact categories of interest.Alternatively or additionally, a library of predefined prompt sets for common product types could be used, which can be further adapted based on specific LCA requirements, and / or an AI-driven system that learns to optimize prompt generation based on previous LCAs.

[0078] In a preferred embodiment of the invention, additional external databases and / or systems regarding product and / or material information are integrated.

[0079] This integration capability improves the accuracy of the bill of materials. By connecting to external databases, the method can incorporate current and specialized information on materials, processes, and environmental impacts. This could include access to life cycle databases, material property databases, and / or industry-specific environmental impact factors. The integration enables a more comprehensive and accurate life cycle analysis by supplementing the AI-generated data with validated external information.

[0080] Further advantages, features, and details of the invention will become apparent from the embodiments described below and the drawing. The figure shows a flowchart of a method for determining a bill of materials for a life cycle analysis of a product according to one embodiment of the invention.

[0081] Similar, equivalent, or equivalent elements can be provided with the same reference symbols in the figure.

[0082] The figure shows a flowchart of procedure 100 for determining a bill of materials for a product life cycle analysis. The bill of materials is thus provided for the life cycle analysis, specifically exclusively for the life cycle analysis. Procedure 100 comprises steps 101, 102, and 103. 202501871

[0083] 11

[0084] In the first step of the procedure, several KL prompts are generated. These prompts are instructions designed to obtain specific information, particularly technical information, required for generating a detailed bill of materials. The prompts are structured to guide the KL in collecting comprehensive data about the product's components and subcomponents, along with their associated technical information.

[0085] In the second step (102) of the procedure, an AI model is controlled using the prompts generated in the first step (101) to determine the bill of materials (BOM). The AI ​​model, specifically a large language model, processes the prompts to generate the technical information required for the life cycle assessment (LCA). This step leverages the power of artificial intelligence to quickly generate detailed and accurate data for the BOM. Using an AI model allows for the incorporation of large amounts of knowledge and the ability to derive information that might not be readily available through conventional methods.

[0086] In the third step (103) of the procedure, the generated bill of materials (BOM) is provided for the life cycle assessment (LCA) based on the output of the Kl model. This BOM includes the product's components and / or subcomponents along with their respective technical information required for the life cycle analysis. The technical information may include material composition, weight, volume, energy consumption, and / or processing steps.

[0087] Alternative or supplementary embodiments of the procedure could include additional steps and / or feedback loops. For example, a validation step could be added after the third step 103. This ensures that the generated bill of materials meets specific criteria and / or standards for life cycle analysis. Another embodiment could include the consideration of external databases, particularly with regard to product and / or material information.

[0088] Method 100, as described, thus provides a technical basis for the efficient and accurate determination of bills of materials for life cycle analyses and addresses technical challenges such as data availability, time expenditure, and / or potential inaccuracies in manual assembly processes. By utilizing AI technology, this method offers a scalable and adaptable technical solution for various product types and industries, enabling more technically reliable product life cycle analyses.

[0089] 12

[0090] According to one embodiment of the invention, the following specific prompts can be used:

[0091] 1. Create a bill of materials (BOM) to be used for generating a life cycle assessment (LCA). List the components in as much detail as possible and include any subcomponents. Generate the BOM for [Quantity; or X kg (Weight)] [Product Name] ([Optionally, include a short product description if a known product weight is helpful]).

[0092] The results can be generated in a structured table with the following columns:

[0093] Component (Component name)

[0094] Hierarchy (Product=1, Components Hierarchy 2, Subcomponents Hierarchy 3, etc.)

[0095] Quantity (number of pieces / components)

[0096] Value (weight, volume or energy consumption of the component)

[0097] Unit of measurement (kg, kWh etc.)

[0098] 2. Add further inputs (e.g., energy and resource consumption and / or scrap rate). Try not to create new columns, but rather new rows, and implement the information in the existing columns.

[0099] 3. Add the following section to the table: On-site manufacturing / assembly (All processes in the actual factory should be totaled, but excluding upstream processing to avoid double counting).

[0100] 4. Add columns: Material (general material information for the component), Specification Material (specific material for the component), Processing Steps (processing / manufacturing steps for the line)

[0101] 5. Is there double counting with regard to the manufacturing and processing steps?

[0102] 6. Please note that only one unique material is assigned to each entry. If more than one material is assigned to a component, split it into multiple subcomponents. Extend the bill of materials until only one material is assigned to each component.

[0103] 7. Check the bill of materials. Is the bill of materials correct and can it be used for an LCA? 202501871

[0104] 13

[0105] 8. Provide the calculated bill of materials for an LCA.

[0106] Although the invention has been further illustrated and described in detail by the preferred embodiments, the invention is not limited by the disclosed examples, nor can other variations be derived from them by a person skilled in the art without departing from the scope of protection of the invention. 202501871

[0107] Reference symbol list

[0108] 100 Procedures 101 First step 102 Second step 103 Third step

Claims

202501871 15 Patent claims 1. Computer-implemented method (100) for providing and determining a bill of materials for a life cycle analysis of a product, wherein the bill of materials includes at least one or more components and / or subcomponents of the product, wherein the method is characterized by the following steps: - (101) Generating multiple prompts for an artificial neural network, wherein the multiple prompts include at least one prompt that instructs the artificial neural network to identify the components and / or subcomponents of the product with product-specific technical information of the components and / or subcomponents, wherein the identified technical information is used for the life cycle analysis of the product; - (102) Controlling the artificial neural network using the prompts to determine the bill of materials; and - (103) Providing the bill of materials for the product life cycle analysis, wherein the bill of materials includes the identified components and / or subcomponents of the product with their respective identified technical information.

2. Computer-implemented method (100) according to claim 1, characterized in that the prompts include a prompt for creating an initial bill of materials structure.

3. Computer-implemented method (100) according to claim 2, characterized in that the initial bill of materials structure includes columns for a name of the components and / or subcomponents, their hierarchy, quantity, weight, volume, energy and / or resource consumption and / or their physical unit.

4. Computer-implemented method (100) according to one of the preceding claims, characterized in that the prompts include a prompt that ensures that the bill of materials includes only one material for each component and / or subcomponent.

5. Computer-implemented method (100) according to one of the preceding claims, characterized in that the prompts include a prompt for adding information for on-site manufacturing and / or assembly to the bill of materials.

6. Computer-implemented method (100) according to one of the preceding claims, characterized in that the prompts include a prompt for adding material and / or processing step information to the bill of materials. 202501871 16 7. Computer-implemented method (100) according to one of the preceding claims, characterized in that the prompts include a prompt for checking and eliminating double counting of manufacturing and processing steps.

8. Computer-implemented method (100) according to one of the preceding claims, characterized in that the artificial neural network is designed as a large language model.

9. Computer-implemented method (100) according to one of the preceding claims, characterized in that the prompts include a prompt for adding information on the disposal and / or recycling of the product to the bill of materials.

10. Computer-implemented method (100) according to one of the preceding claims, characterized in that the method includes validation of the determined bill of materials with regard to standards for life cycle analysis.

11. Computer-implemented method (100) according to one of the preceding claims, characterized in that the method comprises an iterative refinement of the bill of materials based on feedback from an automated validation system, wherein the automated validation system validates the bill of materials with regard to the life cycle analysis of the product.

12. Computer-implemented method (100) according to one of the preceding claims, characterized in that the prompts are dynamically generated based on a product type and specific requirements of the life cycle analysis.

13. Computer-implemented method (100) according to one of claims 1 to 12, characterized in that additionally external databases and / or systems for product and / or material information are integrated.

14. Computer-implemented method for a life cycle analysis of a product, wherein the life cycle analysis is based on a bill of materials of the product, characterized in that the bill of materials of the life cycle analysis is created using a method according to one of the preceding claims.

15. Computer program product comprising instructions which, when executed by a computing unit, in particular a computer, cause it to execute a method and / or steps of the method according to any one of claims 1 to 14.

Images