What materials are actually contained in a product with several thousand components?
To answer this question reliably, information from product data sheets, technical documentation and supplier records must be collated and assigned to the correct components. The required details are often already available, but are scattered across numerous documents and systems.
With large product structures, therefore, the main challenge lies in assigning information: Which information relates to which component, which materials are involved, and which details need to be taken into account for further assessment?
Making product data usable for the assessment of hazardous substances
Manually reviewing and categorising this information can be very time-consuming when dealing with complex product structures. With the help of AI, relevant details can be automatically extracted from existing documents and organised in a structured manner.
Product information, materials, fabric specifications and dependencies can be linked together and represented in a digital product model. This also makes it possible to identify missing or contradictory information more quickly.
This not only makes research easier. It also gives companies a better overview of what information is actually available for individual products and components.
RoHS compliance places high demands on product information
One area of application for such a solution is RoHS compliance. RoHS stands for Restriction of Hazardous Substances and refers to the EU directive on the restriction of certain hazardous substances in electrical and electronic equipment.
For the companies concerned, this means, amongst other things, being able to trace relevant substance information at product and component level. For products with a large number of components, simply collating and assigning the available information can involve a considerable amount of manual work.
Technology can support this part of the process. Documents are analysed automatically, relevant details are assigned to the respective components and made available for subsequent technical assessment. The regulatory assessment remains the responsibility of the company and the relevant technical experts.
Product Compliance in Practice
COCUS has already implemented a similar approach for a company in the heavy industrial sector.
The starting point: product information was available in product data sheets, technical documentation and other sources, and needed to be collated for the assessment of components and materials.
To this end, an AI-powered solution has been developed which automatically extracts and classifies information from existing documents and assigns it to a digital product model. It can also be used to map complex product structures comprising up to 10,000 components and dependencies.
Automated processing reduces the manual effort involved in gathering and categorising information. At the same time, product, material and substance information is made available in a structured format for compliance, documentation and other processes throughout the product life cycle.
Digitising product compliance step by step
To get started, it is not necessary to consider the entire product portfolio. A single selected product or assembly may be sufficient to analyse existing documents and data and to assess technical feasibility.
A proof of concept can then be used to test which information can be extracted automatically and how reliably it can be assigned to products, components and materials. Specific use cases, such as supporting RoHS processes, can be taken into account in this context.
On this basis, the solution can be extended to cover further products and, if required, integrated with existing supplier systems.
Using product information more effectively to ensure compliance
Whether for RoHS or other product compliance requirements, it is crucial to ensure that relevant information from existing documents is reliably accessible and correctly linked to the relevant products, components and materials.
COCUS supports companies with the technical implementation of this process. This ranges from the analysis of existing product information, through automated extraction and mapping, to integration with existing supplier systems. This enables even extensive product structures to be processed more efficiently and information to be made available for further technical evaluation.
