case-studies

Automated Labeling System for OEM Technical Manuals

Home / Case Studies / Automated Labeling System for OEM Technical Manuals
On this page
    Year of implementation
    2024
    Industry
    Automotive

    Global company specializing in the supply of components and advanced high-technological solutions for the modern automotive industry.

    Artea developed an automated labeling system for OEM technical manuals for a global leading company in the mobility, industrial, and energy sectors. The solution combines advanced artificial intelligence models, such as GPT-4 mini and Nomic BERT Base, to accurately classify and organize technical documents. Thanks to a multi-modal pipeline with specific prompts, the system ensures efficient management of manuals, enhancing their usability and integration into the company’s diagnostic and operational processes.

    Approach and methodology

    The project involved the development of an automated labeling system for OEM technical manuals, based on a multi-modal approach that integrates GPT-4 mini and Nomic BERT Base. This combination of AI models enables documents to be analyzed from multiple perspectives, ensuring precise and context-aware classification of technical content. The solution includes an evaluation pipeline with specific prompts, capable of processing large volumes of documents and generating labels consistent with operational needs. This system significantly enhances the usability of manuals within diagnostic workflows, reduces information search times, and optimizes the integration of documents into maintenance and technical support processes, increasing the company’s overall efficiency.

    We implemented a multi-modal approach, integrating GPT-4 Mini and Nomic BERT Base to automate the labeling of OEM documents. Using a structured evaluation pipeline with tailored prompts, the solution ensures precise classification, significantly improving the accessibility and usability of technical manuals within diagnostic workflows.

    Staff involved in the project
    3 Specialists
    Execution time
    1 Months
    Technologies used
    • BERT
    • Databricks
    • LLM
    • OpenAI
    92
    %
    Classification accuracy
    1
    sec
    Average labeling time per document
    1
    Number of Documents

    Get in touch with
    Abacus Group

    Contact us

    Related Case Studies

    case-studies
    XBRL financial statement
    case-studies
    Virtual showroom for personalized and immersive experiences
    View all

    Contact Us

    Thank you for providing the required information below, marked with an asterisk (*). This will help us provide you with the best possible response.

      Abacus Group is committed to protecting your information. Your information will be used in accordance with the applicable data privacy law, our internal policies and our privacy policy. As Abacus is a global organisation, your information may be stored and processed by Abacus and its affiliates in countries outside your country of residence, but wherever your information is processed, we will handle it with the same care and respect for your privacy.