Depuis le 1er octobre 2025, AS Kalve coffee participe au projet de recherche "Latvian Food Competence Centre" n° 2.2.1.3.i.0/1/24/A/CFLA/002 de la Recovery and Resilience Facility (RRF ) : "Développement d'une solution numérique unifiée basée sur l'intelligence artificielle pour l'approvisionnement, la production et la comptabilité afin d'améliorer l'efficacité du travail et l'automatisation d'une entreprise de fabrication de l'industrie alimentaire."

L 'objectif du projet est de développer une solution numérique unifiée pour améliorer l'efficacité du travail et l'automatisation d'une entreprise de fabrication de produits alimentaires en intégrant des solutions d'intelligence artificielle.

Dans la première phase du projet (1er octobre 2025 - 31 décembre 2025), une étude du processus de production et des sources de données sera mise en œuvre.

Date limite de mise en œuvre du projet : 31 décembre 2026.

During the first quarter, research on data and data sources was conducted. This included
fieldwork, process and data mapping, and the integration of available data within a laboratory
environment.
The following activities were performed, yielding these results:

1. Business Process Analysis: Identified key processes and mapped interactions across
the production, warehouse, and logistics dimensions.
2. Warehouse Data Research: Analysed inventory turnover and warehouse data,
including data structures and schematic representations. Focus was placed on
historical data and data granularity. Warehouse data sources are suitable for further
structuring and processing.
3. Production Resource and Process Data Analysis: Identified specific resources used
in the coffee production process (human resources, equipment, raw materials, etc.).
The production stages and steps for the coffee product (coffee cans) were mapped
schematically, including a time dimension.
4. Recipe and Roasting Data Research: Examined coffee roasting recipes and
validated the data structure for use in production planning and forecasting algorithms.
Evaluated the potential for using recipe data to predict production time, resources, and
costs.
5. Identified necessary supplemental data (seasonality, holidays, external demand
factors, logistics delays) and assessed its availability for further use and integration in
laboratory conditions.

Outcome

Acquired a precise technical dataset required for the development of automation and
optimization algorithms, including:

1. A structured and validated technical dataset suitable for algorithm development.
Existing and potential data sources and their applications have been explored.

2. Defined requirements for data processing, enrichment, and integration for future
project activities.

3. Prepared information for algorithm architecture design and laboratory testing in the
next research phase.