Question 01
Can we quote without chasing drawings and quantities?
AnswerMissing information stands out as soon as the request comes in, with your comparable quotes alongside. The estimator starts from a complete base.
Industry: Manufacturing
Your plant produces. HUB AI finds, tracks and prepares.
For companies of 15 to 500 employees.
Revisions, delays, shift handovers and non-conformances: linked in a single memory, with their sources.
The planner, the buyer and quality decide. HUB AI prepares.
Three questions · Manufacturing
Quote, plan and see what slipped in production.
-
-
Question 02
Which orders are affected by a supplier delay?
AnswerThe confirmation received is linked to the orders and due dates concerned, with the reminders ready to send.
-
Question 03
Were there any production delays?
AnswerMachine stoppages and unavailability are linked to the orders and due dates concerned; the planner sees what slipped.
LimitYour managers keep every decision.
Let’s choose a first case in your plant.
Book my DiagnosticYour private foundation
Dedicated capacity for your plant, and a model trained on the way you manufacture.
Drawings, processes and cost prices stay in your environment. They are processed on supercomputers reserved for your company, where a model can learn your business.
-
Private infrastructure
An environment dedicated to your company, hosted at Hub AI, in Quebec. It is not shared with other clients; your teams access it over the web.
Drawings, routings, cost prices and processes stay in this environment.
-
Computing power
Your processing runs on supercomputers reserved for your company. That same power is also used to train your model. HUB AI and its agents are available 24 hours a day, 365 days a year. This power is yours: you use it as you need it, with no bill for each job.
Manufacturing files, drawings and histories are processed on capacity reserved for your plant.
-
Model training
HUB AI can train a model on your ways of working, from examples validated by your teams. It is trained and hosted on these same machines, in Quebec, then tested before it goes into service. Your examples train only your model, and stay walled off, in Canada.
The model can learn your manufacturing vocabulary and your ways of working, from examples validated by your teams.
The Diagnostic will tell whether a tailored model is the right lever, or whether memory and agents are enough.
For which manufacturers
Six types of manufacturers where SMBs of 15 to 500 employees are numerous in Canada and Quebec. The pain points are working hypotheses, to be validated with you during the Diagnostic.
NAICS 332
Fabricated metal product shops
Machining, welding, sheet metal, custom job work.
Pain points to validate: Quotes, drawing revisions, material certificates, non-conformances.
NAICS 333
Custom machinery and equipment
Industrial and agricultural machinery, special equipment.
Pain points to validate: Changing bills of materials, missing parts, scattered technical documentation.
NAICS 326
Plastics and rubber
Injection, extrusion, thermoforming.
Pain points to validate: Press settings, mould downtime, lot traceability.
NAICS 321 · 337
Wood products and furniture
Doors and windows, cabinets, commercial furniture.
Pain points to validate: Product variants, order configuration, lead times.
NAICS 336
Transportation equipment, excluding aerospace
Trailers, specialized vehicles, parts.
Pain points to validate: Configurations by customer, welding and assembly, warranties.
NAICS 311
Agri-food (variant)
Food processing.
Pain points to validate: Food safety, lot traceability, allergens: specific stations to document.
Sources: Statistics Canada, tables 33-10-1014-01 (June 2025) and 16-10-0117-01 (2024); ISED, Canadian Industry Statistics, manufacturing 31-33 (2025); Institut de la statistique du Québec (2024). Aerospace has its own page.
HUB AI at work in your plant
Department shown: 1. Sales, estimating and orders
The information missing from a customer request (drawing, quantity, material, requested date) stands out as soon as the request comes in, with your comparable quotes alongside. The estimator starts from a complete base and sets the price.
The same chain in every plant: the order, methods, planning, purchasing, production, quality, shipping. One agent per step, under one person’s responsibility: it prepares, you decide.
1. Sales, estimating and orders
“Can we quote without chasing drawings and quantities?”
The complete request before the price. The information missing from a customer request (drawing, quantity, material, requested date) stands out as soon as the request comes in, with your comparable quotes alongside. The estimator starts from a complete base and sets the price.
2. Engineering, methods and industrialization
“Which routings does this revision affect?”
The revision, linked to what it affects. Link a new drawing or bill of materials revision to the routings, instructions, moulds or jigs and orders to review. The update follows the decision of the methods manager.
3. Planning and scheduling
“If we slot in the rush order, what slips?”
Options, not surprises. Gather priority orders, confirmed materials and machine unavailability to prepare the options for the planner, who decides.
4. Purchasing, receiving and inventory
“Which orders are affected by a supplier delay?”
The delay, linked to the orders affected. The confirmation received from the supplier is linked to the orders and due dates concerned, with the reminders ready to send. The buyer sends the reminders.
5. Production and supervision
“What’s still open for the next shift?”
A documented handover. Prepare the handover (work completed, stoppages, settings noted, next operation) for a machining cell, a press, a welding line or an assembly line. Checking the program or the setting is still done at the station.
6. Quality and non-conformances
“Have we had this defect before?”
The defect, with its context. Gather measurements, material lot, certificates, drawing revision and the history of the defect. Quality acceptance remains with the quality manager.
7. Shipping, delivery and after-sales
“Which lot does this customer return come from?”
From order to customer, without losing the thread. Prepare the shipping file, then link a request, a return or a warranty claim to the lot, the delivery and the original order.
In support. Maintenance and health and safety
“Is this the third time this machine has stopped?”
The intervention prepared, the procedure up to date. Prepare the intervention with previous breakdowns, available parts and the machine’s approved procedure (energy control, CNESST). The return to service is still confirmed by the person in charge.
Use cases in your plant
Each use case relies on HUB AI products. The products used are shown in parentheses.
Examples of use cases
- Drawing revision propagated. A new revision updates the routing, notifies scheduling and the shop floor, and archives the old version. (Workflows, Documents)
- Rush order simulated. The effect of an addition on the production load is shown before you say yes to the customer. (Watch Centre, Connect)
- Shift handover with nothing lost. Stoppages, adjustments and instructions from the shift summarized and linked to the right production orders. (Feed, Conversations)
- Non-conformances documented. Measurement report, photos and similar cases brought together; the corrective action is prepared for approval. (Business agents, Search)
- Equipment history found. Past breakdowns, parts and interventions found with one question before lifting the hood. (Company memory, Role-based intelligence)
- Prevention tracking. Inspections and corrective measures tracked through to closure, by person responsible. (Workflows, Feed)
- Lighter repetitive data entry. Receipts and slips copied from one software tool to another by an agent, on screen, subject to approval. (Application control, Connect)
What HUB AI can prepare for the manufacturing industry
The industry’s other functions, in addition to those shown in the scene. Each example is a proposed path: it depends on the connected sources, the access rights and the appropriate tools, and the person responsible reviews or decides.
Organization represented: Manufacturing SMB with 15 to 500 employees that produces by order, by lot or by assembly. The perspective follows the chain common to all manufacturers: from order to after-sales. Variants to document: Food processing (food safety, traceability, allergens); pharmaceutical; continuous production; design and R&D.
Finance, costing and billing
Match order, receipt and invoice; only the discrepancies reach the person in charge, who decides on payment.
Automation and industrial IT
Gather documented versions, affected equipment and planned tests for a requested change. Documentation support, separate from machine controls.
Continuous improvement and performance
Group the documented causes of recurring scrap and stoppages for the problem-solving meeting.
Food safety and traceability (food processing)
Variant to document: hygiene logs, lots, recipes and allergens. Applicable stations and requirements.
Illustrative example Proposed path across several teams
- Trigger A supplier confirms that a component is delayed.
- Documents gathered HUB AI links the confirmation received to the affected orders and deadlines.
- Preparation The follow-ups to the supplier and the planner’s options are prepared.
- Decision The buyer sends the follow-ups; the planner decides on the adjustments.
- Result A customer reply is prepared with the new confirmed commitment; the file keeps the decision and the documents that explain it.
Indicators to measure: Time to identify the affected orders; incomplete quality files; supplier follow-ups with no next action. No quantified gain has been established.
Responsibilities: Prices, material substitutions, technical decisions, quality acceptances and returns to service are still confirmed by the people in charge.
Functions common to all industries (12)
Management and steering
Bring together the blocked files, their owner, the source documents and the decision expected, to prepare the management meeting.
Human resources and recruitment
Prepare the onboarding of a new employee with the tasks, requested access and documents that match their position.
Training and knowledge transfer
Find an approved procedure, its version and the person to consult; prepare a training path by role.
Finance, accounting and payroll
Match order, receipt and invoice; present the discrepancies to the person responsible before payment.
Business development and communications
Prepare a proposal from a request and authorized references; keep the commitments made to the client.
Client relations and handling of dissatisfaction
Group the requests tied to a file and prepare a reply with the real status of the work.
Purchasing and supplier management
Compare offers against defined criteria and track the supplier’s missing documents.
IT, cybersecurity and support
Link an incident to the equipment, the symptoms already observed and an authorized procedure to help the technician.
Document management and protection of information
Find documents by file and by access rights; track access requests and the applicable retention dates.
Legal, contracts and insurance
Flag contract deadlines and gather the documents for renewal or for review by the person responsible.
Quality, risk, health and safety
Link an event to a corrective action, an owner and evidence of completion.
Buildings, equipment and general services
Prepare maintenance work and track service contracts, equipment and scheduled inspections.
Private infrastructure
Your drawings, your prices and your processes: handled in an infrastructure that serves only you
A plant handles information that its customers and competitors should never see circulating. Here is why the place where it is processed matters, and what HUB AI’s private infrastructure changes.
Which data
- Drawings, routings and manufacturing processes
- Cost prices, quotes and margins
- Drawings and requirements entrusted to you by your customers
- Employee files and health and safety records
Why it’s sensitive
- This information is often covered by confidentiality agreements with your clients and your partners.
- Law 25 (Quebec’s privacy law) governs the collection, communication and retention of personal information, including its communication outside Quebec.
- A public cloud AI tool may process data outside the country, reuse it under its own terms or make it accessible to third parties; these risks vary by provider and contract.
How HUB AI keeps them in your hands
- Hosting in QuebecYour processing runs on infrastructure hosted at Hub AI, in Quebec, in Canada.
- Machines that serve only youAn environment dedicated to your company, hosted at Hub AI: it is not shared with other clients.
- Your teams access it over the webNo hardware to install on your premises: your teams work in HUB AI from their browser.
- Controls to define with youAccess rights per person, access logging, encryption and retention periods: capabilities to confirm for your deployment.
These references describe general obligations; they are not legal advice. Private infrastructure does not guarantee compliance on its own: compliance depends on your practices, your contracts and the assessment made for your organization.
Let’s talk about your plant.
The Diagnostic tells you where HUB AI starts in your company.