How Machine Vision Can Be Used In Manufacturing

How Machine Vision Can Be Used In Manufacturing

By Mikaela Berman, July 5, 2018      Categories: AI & Machine Learning      Tags: , ,

Artificial intelligence is permeating through nearly every sector and disrupting industries around the globe.  Whilst robotics isn’t a new thing, in the manufacturing sector when AI and machine vision blend together the potential benefits to business can’t be ignored.

In this blog post, we look at how machine vision and AI are being used in manufacturing. Specifically, we dive into:

  • Predictive maintenance
  • How machine vision can be used to read barcodes
  • How machine vision can be applied to goods inspection
  • How AI and machine vision can help improve health and safety
  • How defects in the production line can be reduced

 

Want the extra insights? Download our free “Top 10 Machine Vision and Imaging Technology Solution Providers of 2018” info sheet!

 

Predictive Maintenance

Machines on production lines often must be maintained at regular intervals or they run the risk of breaking down mid-production.  This can be catastrophic for a business that manufactures components or even full-blown automobiles.

robotic arms in manufacturing production line

Not to mention, that undergoing routine checks, whilst important, can also result in scheduled downtime which still brings production lines to a standstill thereby resulting in potential reductions in productivity or profit. For example, just 60 seconds of downtime in an automotive factory can result in the loss of as much as $20,000 (on high-profit vehicles).

 

While robotics isn’t a new thing in the manufacturing sector, when AI and machine vision blend together, the potential benefits to business can’t be ignored. Click To Tweet

 

This is where predictive maintenance, powered by machine vision and artificial intelligence really shine.  FANUC, based out of the US, developed a software program called ZDT (Zero Down Time) that collects images from cameras attached to robots. The images are augmented with metadata which is then pushed to the cloud for processing which subsequently can identify problems before they happen.

During an 18-month trial, ZDT was deployed to 7,000 robots in nearly 40 production lines around the world and could detect and prevent over 70 failures.

 

Reading Barcodes

The inner workings of cell phones and mobile devices often feature printed circuit boards (or PCBs) which help the device function.  As consumer demand for such goods shows no signs of slowing down, manufacturers are always looking for a way to maximize the number PCBs that can be produced in the manufacturing process.  One way to do this is called “panelization”.

With this approach, a number of identical PCBs are printed onto a large panel. Each circuit is then separated by the robotic machine and is identifiable by a unique barcode.

barcodes being used as panelization to maximize prodcuction process

Historically, these barcodes were manually read by a human but this was time-consuming and open to error.  Enter “PanelScan”, this high-tech machine vision solution can automatically read the barcodes thereby improving the productivity of the production line and consequently help improve profits.

 

Want the extra insights? Download our free “Top 10 Machine Vision and Imaging Technology Solution Providers of 2018” info sheet!

 

Goods Inspection

During the production of medical tablets or capsules, pharmaceutical companies need to count the number of tablets being placed into containers.  To solve this business problem, a firm based in the UK called Pharma Packaging Systems built a solution that can augment an existing production line’s capability.

One such feature of the solution is the use of machine vision to check for broken or malformed tablets.  As a tablet makes its way along the production line, images are taken which are then automatically transferred to a dedicated computer for further analysis to check the tablet or capsule’s various attributes.  The software will check for attributes that include, but are not limited to:

  • Color
  • Length
  • Width
  • If the tablet is whole or not

The Vision Inspection System features a counting mechanism that validates if each container contains the correct number of items in it.  If the total number of tablets isn’t correct, or if a single tablet or capsule has been flagged as defective, when the container reaches the end of the production line, any containers that contain “defective” tablets are rejected which removes the risk of packing and shipping defective medical tablets.  You can read more about Pharma Packaging Systems solution here if you’re interested.

 

Improving Health and Safety

The application of artificial intelligence and machine vision isn’t just restricted to components or products in manufacturing production lines.  Machine vision and artificial intelligence can also be used to help improve the health and safety of humans.

For example, the tech firm NVIDIA, often associated with high performant computer graphics cards, formed a partnership with Komatsu Ltd.  Komatsu is a firm based in the UK that are pioneers in the manufacturing of mining and construction equipment.

The partnership between both firms will see the integration of NVIDIAs Jetson AI platform into Komatsu’s drilling, excavation and mining equipment.  Using a combination of real-time cameras and video analytics, the businesses plan to track human movement, and powered by deep learning, predict the movement of equipment to help avoid dangerous situations on a construction site.

 

Machine vision and artificial intelligence can also be used to help improve the health and safety of humans. Click To Tweet

 

There are roughly 150,000 construction site accident injuries each year according to the Bureau of Labor Statistics, a solution like this can only help to reduce this number.

 

Defect Reduction

Manufacturers understandably want components that roll off the production line to be free of defects and issues, but being able to do this at scale can pose problems for more manual efforts.  Machine vision, however, is the ideal technology that can help businesses automate a problem like this.

A company called Shelton have a surface inspection system called WebSPECTOR which can identify defects, store images and accompanying metadata related to said image.  As items are processed along the production line, any defects that are identified get classified according to their type, they are then assigned an accompanying grade to help further identify the severity of the defect.

surface inspection system called WebSPECTOR

With this sort of information, manufacturers are able to differentiate between the types of defect that are occurring, and it can also help them implement procedures and policies.  For example, manufacturers can introduce a process that halts the production line when X number of Y types of defect have occurred.

This software technology, coupled with state of art cameras has been able to improve the productivity of a fabric producer by as much as 50%!  You can read more about that story here.

 

Want the extra insights? Download our free “Top 10 Machine Vision and Imaging Technology Solution Providers of 2018” info sheet!

 

Summary

In this blog post, we’ve looked at how machine vision and artificial intelligence can be used in manufacturing.  We’ve looked at some diverse use cases for the technology and how it can be applied from predictive maintenance, to even improving health and safety.  We closed out by looking at how machine vision and software can be used to help reduce defects in manufacturing lines and help improve productivity.

Here at Growth Acceleration Partners, we have extensive expertise in many verticals.  Our nearshore business model can keep costs down whilst maintaining the same level of quality and professionalism you’d experience from a domestic team.

Our Centers of Engineering Excellence in Latin America focus on combining business acumen with development expertise to help your business.  We can provide your organization with resources in the following areas:

  • Software development for cloud and mobile applications
  • Data analytics and data science
  • Information systems
  • Machine learning and artificial intelligence
  • Predictive modeling
  • QA and QA Automation

If you’d like to find out more, then visit our website here.  Or if you’d prefer, why not arrange a call with us?

The following two tabs change content below.
Mikaela Berman is a Senior Marketing Manager at GAP. Throughout the past 8 years, Mikaela Berman has worked as a consultant for companies from startups to multinational enterprises, helping define their business models and marketing strategies. She also sits on the board of two Austin startups. She has a BA from the University of Maryland and a MS in Technology Commercialization from the University of Texas at Austin.

Latest posts by Mikaela Berman (see all)

Comments


Leave a Comment

Your email address will not be published Required fields are marked *

Subscribe to our Newsletter