Quebec Forestry Associations

The customer

GFQ (Groupements Forestiers du Québec) is a provincial network comprising 33 independent forestry groups located throughout Quebec. Each group manages its own operations and database, which provides the organization with a wealth of information, though this information is scattered across as many separate systems.

Challenges

The existence of 33 independent databases makes it difficult to get an overall view of the network.

Lack of a centralized tool that allows for reliable and secure querying of all network data.

Risk of conflicts between identifiers and codes from different groups, which could compromise the integrity of the consolidated data.

High volume and frequency of data transfers (backups of up to 500 MB) requiring a secure communication channel.

‍

Initial findings

Some groups use their own database structure, without standardization across groups.

Some identifiers (grouping codes) contain inconsistencies or duplicates in the reference files, posing a risk of data contamination if they are not validated before integration.

Temporary and working tables (e.g., print jobs) had to be excluded from the integration process so as not to overload the central database.

Data integrity constraints were not managed consistently across the various groups.

‍

Objectives

Design and deploy a single central database that consolidates the 33 forestry groups.

Ensure the traceability and auditability of every piece of integrated data, from its source through to its upload.

Secure data exchanges between the groups and the central platform.

Make network data searchable by artificial intelligence agents and a chatbot, for the benefit of all teams.

Provide GFQ users and each forestry group with access to AI agents tailored to their respective needs, while respecting each group’s specific data scope.

‍

Interventions

Diagnosis

Comprehensive mapping of the data schemas for the 33 groups and identification of discrepancies between their respective structures.

Identifying the risk of conflicts between codes and identifiers from different groups.

Analysis of the official reference file of grouping codes to confirm the exact codes and identify ambiguous cases, particularly those related to recent mergers among groups.

Identification of tables and data to be excluded from the consolidation process, such as temporary tables and application caches.

‍

Key actions

Design of a comprehensive central architecture: data model, extract-transform-load (ETL) pipeline, view schemas specific to each group, semantic layer, and audit log.

Implementation of a single prefixing mechanism by group to eliminate any risk of identifier collisions.

Added traceability columns to each table, enabling secure data restores and full data auditability.

Deployment of validation gates that automatically block the integration of any ambiguous data until it is confirmed by the software vendor.

Securing backup transfers and implementing single sign-on with least-privilege role management.

Deployment of a standalone MCP (Model Context Protocol) server connecting the central database to OpenWebUI, enabling artificial intelligence agents to query network data in a secure and structured manner.

Connecting the central database to a chatbot deployed on OpenWebUI, to enable teams to query all network data using natural language.

Deployment of AI agents that address both GFQ’s province-wide needs and the specific needs of users in each group, with each user having access only to their own group’s data through application role management based on the principle of least privilege.

‍

Results

Delivery to GFQ of a complete, operational solution: central architecture (database, ETL pipeline, quality controls, audit log), MCP server, and OpenWebUI interface connected to the central database.

Provision of an OpenWebUI chatbot connected directly to the network's consolidated data, allowing users to query it using natural language.

Deployment of AI agents that make it easier for both GFQ administrators and users within each group to interact with and visualize data, without requiring technical skills.

Elimination of the risk of collisions among the 33 groups thanks to the unique prefixing mechanism.

Implementation of automatic validation checks to prevent the inclusion of data of questionable quality until the issues are resolved.

Comprehensive security for data transfer and access through single sign-on and least-privilege roles.

The entire solution is hosted in a data center located in Canada, with data protection that complies with Bill 25 on the Protection of Personal Information.

‍

Benefits

This results in significant time savings for GFQ administrators and employees of the groupings, who no longer have to manually consult 33 separate data sources; this benefit is consistent with data centralization initiatives underway elsewhere in the forestry sector.

Better-quality and faster decision-making through unified, reliable, and up-to-date access to all network data—a key driver consistently identified in North American forestry digital transformation initiatives, including those supported by Natural Resources Canada.

A reduction in the risk of errors and inconsistencies associated with manual data entry or consulting disparate sources, thanks to the quality controls and validation checks built into the platform.

Greater autonomy for users in each group when querying and viewing their own data using AI agents, without having to rely on technical support for every request.

GFQ's positioning is in line with the global trend in the forestry sector to invest in digital infrastructure and artificial intelligence to support productivity, competitiveness, and decision-making, while ensuring rigorous data governance in compliance with Act 25.
These benefits, which are consistent with trends observed in forest digitization projects in Canada and elsewhere in the world, can be further defined and quantified for GFQ once the agents are fully operational within the teams.

‍

View all real cases

Become a real case!