• Nederlands

Sector

Industry & maritime

Design work still done by hand, and rejects that only become visible once the product is finished. Three of our cases sit here: shipbuilding with MasterShip Software, generating ship hulls, and a design module for custom stairlifts.

A white ship hull section on a workbench, with a nesting drawing of steel plates in front of it

The situation

The bottleneck is in the design, not in production

CAD/CAM has been standard for product design for decades. Automating the tasks within that design has lagged behind: an engineer still draws by hand what is in effect a repeat of something drawn last week, within the same structural rules.

That is a different kind of AI problem from a prediction model. There is no dataset with thousands of labelled examples. What there is, are design rules: hard requirements the outcome must meet, and rules of thumb that live in the heads of experienced engineers. The work starts by getting those rules onto the table, and only then with generating.

Why generating is not the hard part

A model that produces a thousand variants is quickly built. The value is in the constraint: which variants meet the structural requirements, can be produced on the machines that are already there, and fit the existing way of working. That approach, first uncovering the design rules and then generating automatically within hard requirements, was developed on the ship hulls and later applied to custom stairlifts.

Roles

What we do here

Twentynext works in four ways: advice, people, projects and maintaining what runs. In this domain these three occur most often. What advice means here, you will find under AI consultancy for industry.

Uncovering design rules and automating them

Modules that run inside the design environment that is already there, usually AutoCAD, taking over the repeatable part of the drawing work within the requirements the structure imposes.

Seeing rejects before the customer does

Image assessment in the production line, on your own rejection standards and trained on your own products. What that demands in image quality and lighting we say beforehand, not afterwards.

Specialists in your own team

Data engineers and developers who work inside your environment, with our projects and R&D behind them.

Evidence

Where this has already been done

Case

AI for Lightweight Construction

With MasterShip Software: an AutoCAD module that automates technical design tasks in shipbuilding.

Read the case

Case

Designing ship hulls

From hull shape to construction plan: the arrangement of steel plates with which the yard assembles the ship like a puzzle.

Read the case

Solution

CAD/CAM automation

A module that generates a production-ready design automatically from a stairlift measurement. Carried out for a manufacturer of stairlifts, whose name we do not mention.

View the solution

Scope

What we do and do not claim here

Of the three cases above, we do not name the client for two. For the stairlifts that is on substantive grounds: this work touches that manufacturer’s competitive position. That is not a formality but precisely why the work has value.

We do not supply machines or off-the-shelf production software. What we build are models and modules that run inside the design and production environment that is already there. If an existing package solves your question, we will say so; that is the honest answer more often than it sounds in this market.

View all cases and projects

Want to know where your organisation stands? Take the free Data Maturity Scan: sixteen questions, immediate result.

Work with us

Realise your project together?

The people who build it also run it afterwards. Eindhoven, since 2014.

Martijn van Grieken

Martijn van Grieken

Director Data & AI

Get in touch