Admit it. After buying something online—whether directly from a company or person, or from a non-Prime Amazon order—you’ll often find yourself tracking the online purchase, wondering when it will arrive. I know I do. I recently ordered a hooded sweatshirt of a band I want to support from a music and entertainment online merchandising company. For a few days, the tracking system kept telling me that the order had not yet been filled, so there was obviously nothing to track just yet.
Eventually I began to wonder if the item was even in stock, so I emailed customer service.Within a day or two, a representative responded, informing me that it takes two-to-three business days for an order to be processed, then another two-to-four business days for the items to be produced and shipped, plus whatever it took for the actual shipment. Given that my order was made on January 16 and the company does not ship on weekends or holidays, the earliest I could expect my delivery—estimating that shipping would take another two-to-four days—was January 31. (It did arrive on January31.)
Now, this isn’t necessarily a long time to wait, especially for something like an article of clothing, but Amazon Prime has spoiled a lot of us. Of course, this order was made from a smaller company with significantly less demand than Amazon, so maybe these longer times will always be the norm. But—in addition to Amazon—how do larger companies selling products online keep up with demand and customer requirements? One definite answer isautomation.
Take for example, DHL’s new logistics robot in the Netherlands, which is described on page 21 of this issue. Guided by 3D machine vision and designed by robotic machine vision specialist Robomotive, the robot is the first fully-automated system of its kind at DHL, where it locates and moves boxes of printer cartridges onto a conveyor belt for shipping. DHL’s system achieves a speed of 400 picks per hour and replaces a tedious task previously performed by human operators. This is just one example of companies using robotic and machine vision technologies to automate a process and save time, money, and labor. There are, of course, many other companies doing similar things.
There are obvious ways machine vision can make things more convenient on a day-to-day basis. Then there are scenarios like this where the technology is used behind the scenes to improve processes, increase efficiency, and so on. One other example in this issue is on page 25, where Andy Wilson describes a smart vision system used to inspect cylinderbores.
An important part of combustion engines, cast cylinder blocks consist of holes called bores in which components are mounted. These bores are sprayed with a wear-resistant coating and must be inspected to ensure the coating is even and that there are no deposits of coating on the inside walls. In short, these are important automotive parts that must be fully and accurately inspected, and here, machine vision and deep learning technologies team up to automatically doso.
What about you? Have you been involved in the design or integration of a machine vision system lately? We’d love to hear aboutit.