Ayesa Digital has developed a platform based on Generative AI to optimize bidding processes at Iberdrola. The Generative AI solution uses advanced language models and machine learning techniques to speed up the categorization of documentation, evaluate risks according to predefined templates and carry out technical-economic pre-assessments in an automated manner. Thanks to the use of Generative AI, the efficiency of document analysis is significantly improved.
The Generative AI platform also allows the generation of assessment criteria, comparisons between offers, risk analysis and scenario simulations. Likewise, Generative AI facilitates intelligent searches in more than 100,000 documents per year and offers comparative analysis in real time, improving operational efficiency, decision-making agility and process reliability.
The Ayesa Digital project aimed to streamline the bidding process and make the evaluation of offers more efficient through Generative AI, facilitating quick access to key information, centralizing evaluations and promoting data-based decisions thanks to advanced Generative AI capabilities.
4 essential areas powered by Generative AI
- Smart classification: Generative AI allows searches in natural language, generation of automatic summaries, alerts for missing documents and early identification of risks (cybersecurity, outsourcing, among others).
- Technical criteria: Through Generative AI, the system generates evaluation criteria based on technical specifications, proposes improvements and detects ambiguities in the specifications.
- Technical evaluation: Generative AI compares offers based on automatically generated dynamic templates, highlights discrepancies and evaluates their technical impact.
- Economic evaluation: Supported by Generative AI, the platform facilitates cost comparisons, financial scenario simulations, anomaly detection and advanced commercial analysis.
The application of Generative AI reduces time, costs and manual errors, improves the traceability and quality of the purchasing process, increases objectivity and allows 100% of the available documentation to be analyzed.
Applied innovation and good practices
The transformation project in the Purchasing area stands out for applying Generative AI in a tangible way in a critical business process. Thanks to Generative AI, the bidding process becomes a digital flow assisted by algorithms, from the definition of criteria to the final comparison of offers. Semantic search engines and dynamic generation of criteria based on Generative AI are integrated, which suggest improvements, detect ambiguities and adjust the specifications to real needs.
Through RAG architectures, Generative AI uses updated and contextualized information to compare offers, analyze inconsistencies and evaluate technical-economic impacts that previously required many hours of manual work. Machine learning allows Generative AI to iteratively refine the tool based on historical data, progressively increasing its reliability. The adoption of SAFe for global coordination and Scrumban for daily execution has facilitated continuous iteration, the incorporation of feedback and a constant focus on business value, reinforcing technological innovation based on Generative AI.
Use of technologies (ICTs)
The Generative AI-based solution is built on a modern, secure and scalable ecosystem. Deployed on AWS, it uses SageMaker to train Generative AI models, Bedrock for access to generative models and RDS PostgreSQL for data management, combining computing, embedding storage and advanced analysis in a single environment. The Generative AI platform integrates with internal systems such as SAP, document repositories and RPA tools to cover the entire process cycle, from receiving offers to generating final reports.
Ayesa Digital’s Generative AI facilitates quick access to key information, centralizes evaluations and promotes data-based decisions
The design is responsive and accessible from multiple devices, allowing natural language queries powered by Generative AI, which simplifies access to complex information. It includes real-time monitoring to identify incidents, bottlenecks and updated metrics, reinforcing transparency and decision making. Security is a fundamental pillar: the Generative AI solution uses advanced encryption, access control and policies aligned with standards such as OWASP, guaranteeing the integrity and confidentiality of supplier and financial data.
