FHS-JULAUG24-PG38+39_Layout 1 07/08/2024 10:14 Page 38
PACKAGING
SOLVING THE SIZEABLE PROBLEM OF NON-MACHINABLE ITEMS
N
on-machinable items, often called ‘uglies’ because of the difficulties they pose, can take many different forms – from big TVs, brooms and car tyres to large pots of paint.
What they all have in common is that, due to their characteristics, they cannot be accommodated in a standard, automated sorting system. As postal services and e-commerce companies invest heavily in fixed infrastructure that is optimised for smaller items, non- machinables present a sizeable problem. They must be identified, separated, and accommodated in a different process. Traditionally, this has been accomplished manually. Warehouse staff identify non- machinables by eye, then move them by hand to a separate chute or destination. This is time consuming and costly, placing unnecessary strain on workers. Furthermore, a manual approach isn’t fact based, raising the possibility that a non-machinable item might slip through. Improving the identification and transport of these goods has long been an uphill battle, but computer vision and robotics can provide a solution.
EYES ON THE OVERSIZED PRIZE Computer vision technology offers the capability to quickly and accurately identify non- machinable items at the beginning of the sorting process. This eliminates any manual intervention, greatly improving efficiency.
Non-machinable or non-conveyable items are the bane of any postal or e-commerce sorting operation. Whether large, heavy, fragile, unstable or oddly shaped, these goods can’t be automatically sorted and instead require inefficient, manual processing. However, with new automation technologies like computer vision and robots, businesses have an opportunity to change this. In the near future, every item could be machinable. Lars Pruijn, innovation director, and Mart Ruijs, product manager at Prime Vision explore how computer vision and robotics could make non-machinable items a thing of the past.
Correct identification of a wide range of items is
possible thanks to machine learning. By training artificial intelligence (AI) models using examples from a business’s real-world operations, a computer vision system can not only recognise objects based on dimensions and weight, but other characteristics such as shape, stability, packaging type and more. The more items the system is exposed to, the better it becomes at recognition. As well as relying on hard data to correctly
determine non-machinable items, the system can assist in automatically pre-sorting them to an appropriate chute, conveyor, area or robot – so no unsuitable packages slip through.
38 JULY/AUGUST 2024 | FACTORY&HANDLINGSOLUTIONS
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70
