One of my businesses is a design and manufacturing operation. Cactus manufactures backpacks and outdoors clothing. For years we’ve been using PAD which is a CAD/CAM software package to make patterns for the apparel industry. PAD works well but has always been a little backwards in terms of development, online availability and the like.

A new offering that one of my colleagues mentioned to me today is pretty interesting. Pattern pieces get created in the installed application as they always have but from here things have changed. In the past, when one wanted to make a marker (the template to cut out all the pattern pieces) one spent ages playing around with the pieces to get the highest efficiency. Obviously given labour costs, this is less than efficient.

The new system combined the PAD offering, with some technology created by New Zealand business ShapeShifter. ShapeShifter is apparel optimisation software – it uses some cool algorithms to reduce fabric consumption and optimise production planing.

In a nice little SaaS/Traditional combination, we can now email our pattern pieces directly to PAD where, as an added service to owners of their software, an automatic process occurs which creates the most efficient lay plans possible using ShapeShifter’s offering.

In our case – it formerly took a highly skilled (and paid) worker 30 minutes or so to create a marker with 80% efficiency. The new offering achieves higher efficiencies, at no cost and in a similar amount of time.

Nice work guys!

Ben Kepes

Ben Kepes is a technology evangelist, an investor, a commentator and a business adviser. Ben covers the convergence of technology, mobile, ubiquity and agility, all enabled by the Cloud. His areas of interest extend to enterprise software, software integration, financial/accounting software, platforms and infrastructure as well as articulating technology simply for everyday users.

3 Comments
  • Falafulu Fisi |

    optimisation software

    Yeah, Ben I’ve read about ShapeFitter via business articles on the NZ Herald, it is a good software since it uses Operations Research (OR) techniques in which Linear Programming (LP) is dominant in OR.

    I do use OR algorithms in my development and it is a hugely an untapped domain (largely because it is a mathematical/engineering branch, which is not taught at IT/Computer science courses, since they’re quite complex to comprehend or very difficult to implement algorithmic-wise), but over time, we’ll see OR being adopted in web 2.0 applications as ShapeShifter has shown. OR is also used in certain search engine algorithms. There are commercial APIs available in OR (the leader here is ILOG Corporation), some mentioned int the Optimization newsletter (which I subscribed to) that for a single developer license is about ~ US $20,000 . Optimizing operations in a factory (machine job schedules , etc…) is very simple using the ILOG optimization API, since they have taken the complex task of OR and made simple for developers who don’t have expertise on the subject. Some major airlines flight attendants shift (rotation) schedules had been developed using the ILOG commercial APIs. There are also open source projects available on the internet (C++, Matlab, Java, etc…) in OR. The advantage of using OR is that it minimizes the loss (wasted resources etc…) while maximizing the profit (gain).

    I do use OR in investment optimizations (using the modern economic Markowtiz Portfolio Theory), such how to divide up an amount X dollars into a portfolio that contains Y assets that will minimize the risk or loss (regardless of how the market move) and maximize profits (returns). This technique is widely adopted in the financial industry. Eg, suppose that an investor has $200,000 to invest in say 5 assets, Xero, Telecom, Burger Fuel, Pumpkin Patch, Rakon. An investor could divide the $200,000 into equal amount, such as $40,000 to each asset or perhaps just divide it according to some personal criteria such as $100,000 to Xero, and $20,000 to the remaining 4. There are infinite ways to dish up the $200,000 into the 5 assets portfolio. The problem here, is that these random allocations of funds to the assets, the investor takes a huge loss if the market moves in an unfavorable direction to his portfolio and if the market moves in a favorable direction then his/her gain is minimal. Buy using optimization algorithms, it gives exactly the weights (allocation amounts) of how the $200,000 should be divided into the 5 assets. In this way, if the market moves in an unfavorable condition to the portfolio, the loss is going to be minimal (out of a vast possibilities in terms of amounts). On the other hand, when the market moves in a favorable condition to the portfolio, then the gain is maximum (and again out of a vast possibilities in terms of amounts).

    I have seen OR being used in Job Costings where the aim is to minimize the spending (time-wise/material-wise) while maximizing the profits. It is certainly being used in budget allocations and I have pointed out to Rod last year (2007) about a document that shows how it is used in budget allocations. Now I see that Xero is doing Job Costings, I reckon that they should explore OR algorithms for competitive advantage compared to Xero’s competitors including SaaSu, MYOB where they haven’t implemented OR yet.

    The beauty of using OR, is that users can see immediately its benefits, ie, it saves them money that might have been wasted by doing things or running a business by pure gut-feeling/intuition or based on sales figures which are again not the optimal way of how to run it.

    There is an Auckland company that has been growing at a very high rate over the past few years (Optima Corporation) that uses OR as their core technology.

  • Falafulu Fisi |

    Hehe, my math wasn’t quite accurate, in there was an obvious error in the following line:

    divide it according to some personal criteria such as $100,000 to Xero, and $20,000 to the remaining 4

    It should be correctly stated as:

    divide it according to some personal criteria such as $120,000 to Xero, and $20,000 to the remaining 4

    Now the correction sums up the total to $200,000 rather than the incorrect sums of $180,000

  • I hadn’t heard of shapeshifter….but doesn’t PAD come with marker making? I recently purchased a CAD system (optitex) but decided against PAD for strategic reasons. I’m wondering how SS would work; I mean, I got a marker making module added on for $1,000. I’m puzzled about the third party add on.

Leave a Reply