OneBoxVision Blog

Vision system networks will reduce CapEx and improve payback

Why industry is beginning to integrate machine vision as part of their industrial network.  The new way forward is to combine the use of embedded smart technology, imaging workstations and post processing solutions to optimize the payback of a vision system investment. Lets look at what's involved in building a vision network.

 Vision purchases are not always planned

Manufacturing has completely changed in the past 20 years. Operations no longer use paper records to manage orders. Companies have adopted enterprise resource planning solutions to take care of all of their operational business needs, from supplies, asset allocation to order execution.

Productivity gains rose in the planning end of business, but most businesses then had to go through long manual operating procedures to adjust and set up their lines for each order. In the past ten years advanced operations have introduced manufacturing execution systems to support and integrate this automation. As part of these MES solutions, a crucial element sits behind called the historian. Life Science, Oil & Gas, and other leading industrial sectors in automation have deployed historians for many years. The historian stores all relevant process data for all sensors and machines.

vision.jpg

However the integration of machine vision systems into the factory network has very much been piecemeal. The trend in manufacturing is the use of smart cameras, and distributed processing. Vision networks are carefully planned infrastructures that are an essential element to support the smart factory and the goals of greater productivity and flexibility, all goals of Industry 4.0 and the connected enterprise. What are the key attributes of a vision network?

  • A vision network requires that each element of an operation where a vision system can be used is planned, it is known as the Vision map.
  • Each manufacturing cell or operation is identified where a vision system can add value.
  • The key requirements and benefits are identified. This includes the role of automated quality alerts for the operator (traditional role of a vision solution) , the collection of process data for engineering, integration with the manufacturing historian, targets for increased productivity and goals.

Think about how machine vision systems have been used and purchased to date. The system has been bought for one purpose only, to inspect a product on a line and to alert the operator of any defects. This is usually in response to a customer complaint. This is very much analogous to security systems in the home. Often we do not take action until we have had an incident.

The inspection system as used is dedicated to alert major defects to an operator. If an inspection system is on a line where the product can be rejected in real time then corrective action may not even be taken until almost all product is rejected. Very often the purchase of an inspection system is done as part of a larger project and is deemed an add on. There is often no thought how it is going to integrate with companies wider operations. The equipment acts as an island used by the operator for that moment on that line.

The return on investment is based on reduced customer complaints and if used at the point of production (this is not always the case) a reduction in inline waste. If an inspection system is used to inspect material on a final process for defects introduced upstream it will not reduce waste. It may reduce customer returns. The net result is that the vision system is seen as a cost with no perceived value add and often not used by production properly as it is also perceived as a bottleneck.

So what is the difference in planning a vision network compared to buying an inspection system?

Identify the appropriate location for the point of inspection. If it is to stop a customer complaint, do not assume that it must be installed on the final process where little to no value can be derived. Build the vision map as it should be one element of a planned vision network. Each manufacturing cell should be identified. The operation and function that a vision system can perform should be identified for all operations, and the feed forward and backward value identified. This almost always includes some element of video storage. Based on the function and added value from the data, it becomes very clear which location provides the best net return on that asset. Make sure that all other elements of the purchase are considered so that the result is that you have a vendor independent infrastructure that supports independent software applications and that the data is stored in a documented open format.

Calculation of ROI and NROA (Net Return On Asset) on a business infrastructure is different to the purchase of a standalone vision system. The network is a necessary element of any modern manufacturing operation.  Vision networks differ from normal business networks only in that the data bandwidths are much larger. 

Why are the benefits of a vision network?

  • Customer facing business benefits are probably the hardest to quantify and the most valuable. Reduction if not elimination of a variety of customer returns. Not all returns are due to visual items, this is where the vision network data combined with other quality sensors as part of a MES provide value. For example, the vision system sees a process change that on its own is not defective, but that data is linked to another sensing technology to identify product that when both data points are combined the product may be out of specification. It is part of any business strategy to include an element of waste with deliveries, as the risk of return has been identified and reduced by the vision network. This is not a viable business strategy if the product quality has not been determined and
  • Insurance claims due to product issues. If a proper track and trace system is implemented, product liability claims may be fought with hard evidence of product quality through use of
  • Root cause analysis and process improvements reduce inline waste and increase asset OEE - Vision networks integrated with sensor historians enable this. This requires the use of
  • Smart converting - This is the use of data to optimize the follow on processes. This includes identifying optimum converting strategies for products that will have value added. The product may need to be reworked prior to the value added process to avoid additional material waste. The use of inspection results, video and sensor data will have an effect of optimizing output and reducing waste, but it should also reduce rework. Smart converting analysis can be automated between process steps using a variety of
  • Lower maintenance costs can be expected through root cause analysis and smart conversion as there will be less unplanned stops.
  • Increased Output - Operators only deal with issues due to their process, there will be an increase in throughput.

Let's summarize

This change in approach to the purchase of quality and vision equipment can completely change the economics of a manufacturing operation. There is no benefit in being vendor specific when building an infrastructure. The key to success is to maintain these key elements:

  • Vision map
  • Vendor independence
  • Video storage
  • Independent software applications
  • Open formats

OneBoxVision is an expert in Industrial IT and machine vision. Feel free to contact us to learn more.

Download the whitepaper on vision networks

Topics: Vision network Machine Vision Building vision networks