![\"Siemens](\"http:\/\/community.plm.automation.siemens.com\/legacyfs\/online\/wordpress\/images\/2014\/10\/Siemens-PLM-Tecnomatix-Big-Data-Solution-for-Production-Quality.png\")
<\/A><\/P>
Take more control of production quality by gaining real-time mashup-like insight into design and production data that includes 3D part models, design requirements, as-measured results and a summary of overall quality statistics.<\/EM><\/P> This report compares the measured dimensional variation of a sub-assembly consisting of four different parts produced in four different plants.<\/EM><\/P> This is an interactive 3D view of the profile deviation between the as-designed and actual product.<\/EM><\/P> Tecnomatix 12 is coming soon, November 2014, so stay tuned as more information will become available. Follow this link for additional information about Tecnomatix DPV:<\/EM><\/P>
If the unrelenting media interest in big data has made you curious about the topic, and about whether your company might be able to use big data to improve production quality, read on. We at Siemens are on top of this, with products that can already turn the huge volumes of data collected by shop floor measurement devices into insights that haven\u2019t previously been available \u2013 insights that will have a significant effect on product quality and operational output.
First, let\u2019s look at basic concepts behind big data. One involves the data itself, which as the term implies, is voluminous. But actually people use three other words that being with \u201cv\u201d to describe big data: high v<\/SPAN><\/STRONG>olume, high v<\/STRONG><\/SPAN>elocity, and high v<\/STRONG><\/SPAN>ariety. I think we\u2019re finally hearing so much about big data in the popular press because this kind of data about people, and the insights drawn from it, are becoming available. (Current example: the book Dataclysm by the founder of OkCupid<\/A>, who uses the big data generated by his site\u2019s users to make some interesting conclusions about human behavior.)
Manufacturers have long had access to huge volumes of data in the form of the information captured by shop floor measurement equipment (handheld devices, CMMs, vision systems). This data meets all three of the criteria (volume, velocity and variety). Is it big data? Possibly not. Even though your company may have several terabytes of measurement data, unless it\u2019s aggregated in some way that it can be, it can\u2019t really be considered big data, in the sense that it can provide new and valuable insights.
The second aspect of big data involves analyzing it. As you may recall from college statistics courses, this involves asking the right questions. Doing that indicates the appropriate analyses to use. I\u2019ll say more about this in a minute.
Probably the most important aspect of big data involves displaying analyses results in ways that makes sense. This is a challenge, as you can see if you look at the illustration on the top right of the Wikipedia page for big data<\/A>.<\/A>
“Viegas-UserActivityonWikipedia<\/A>” by Fernanda B. Vi\u00e9gas<\/A> – User activity on Wikipedia<\/A>.
Licensed under CC BY 2.0<\/A> via Wikimedia Commons<\/A>.<\/P>
The chart shows \u201ca visualization created by IBM of Wikipedia edits,\u201d and it does make some sense, I suppose, especially given all of the information it\u2019s trying to convey. But you can see the difficulty. When you have so much information to work with, how can any information gleaned from it be presented in a useful way?
A Big Data Solution for Measured Quality<\/STRONG><\/SPAN>
Let\u2019s look at the aspects of big data in terms of Siemens\u2019 big data solution for production quality: Tecnomatix Dimensional Planning and Validation (DPV)<\/A>. As you might have guessed, this solution is supported on Siemens\u2019 Teamcenter <\/A>software, which is used by manufacturers worldwide to store and manage the vast amounts of data related to the entire product lifecycle.
Collecting and managing the data.<\/STRONG> Tecnomatix DPV solution, with its Teamcenter foundation, easily handles the first aspect of big data \u2013 collecting and aggregating disparate sources of data into a managed data set. It\u2019s hard to say how much measurement data companies have because it\u2019s normally stored it in so many different places and formats. But the fact that a leading automaker is using Tecnomatix DPV to collect and store over a million measurement data files per month from many different facilities makes it clear that this solution is up to the challenges of big data.
Analyzing the data<\/STRONG>. Tecnomatix DPV helps a company query its measurement data to uncover ways of improving quality, mainly by making it very easy to set up analyses. For example, the automaker mentioned above dynamically sets up analyses conditions (i.e. not using preset templates) that combine different facilities, different production process, different vehicles and different measurement devices. They simply choose the elements they want to investigate from drop down menus.<\/A>
The software retrieves the appropriate data and performs the appropriate analysis. Tecnomatix DPV incorporates extensive statistical analysis functionality, including the ability to perform root cause analysis, for instance, which is very important to improving quality.
Reporting the results.<\/STRONG> This is where Tecnomatix DPV really shines. Compare the color-choked image from the Wikipedia big data page to this Tecnomatix DPV screen shot:<\/A>
The ability to combine multiple charts, graphs and product geometry comes from Teamcenter Visualization software. In addition to displaying information in multiple ways, you can display analysis results on the actual product geometry, as shown below. This way to see and use measurement data is more likely to have a direct impact on product quality than numbers in a spreadsheet.<\/A>
We\u2019ll be writing more in the future about how Siemens PLM Software supports the use of big data. For now, you should know that in Tecnomatix DPV, we offer a closed-loop system that allows you to gain new insights from measurement data, feed those insights back into design process, and ultimately deliver higher-quality products.
Additional notes:<\/SPAN><\/STRONG>
![Viegas-UserActivityonWikipedia<\/A>” by](\"http:\/\/commons.wikimedia.org\/wiki\/File:Viegas-UserActivityonWikipedia.gif#mediaviewer\/File:Viegas-UserActivityonWikipedia.gif\"){kind=link}