Carnegie Mellon University (CMU) has developed a system that can translate a wide variety of 3D shapes into stitch-by-stitch instructions that enable a computerised knitting machine to automatically produce those shapes.

This ability to generate knitting instructions without the need for human expertise could make on-demand machine knitting more available. It may be possible to use the industrial knitting machines to produce customised pieces one at a time or in small quantities, allowing to choose the desired color, pattern and fit.

Knitting machines could become as easy to use as 3D printers.

As of now a floor of knitting machines also needs a department of engineers. Garment designers rarely have the specialised expertise necessary to program the machines. This makes doing one-off customised pieces very difficult.

V-bed flat knitting machines are highly capable but have certain limitations in terms of flexibility. The CMU algorithm takes these constraints into account, producing instructions for patterns that work within the limits of the machine and reduce the risk of yarn breaks or jams.

A number of brands have experimented with the on-demand production concept. In 2017, Intel teamed up with women’s brand Eileen Fisher to experiment with 3D sweater knitting. Last year, Adidas carried out a project, where shoppers could design a sweater, have a body scan to determine fit and get it knitted by a state-of-the-art Stoll flat knitting machine within a couple of hours.