Ford Scours Big Data to Improve Manufacturing

With the recent popularity of "Big Data", it comes as no surprise that some of America's biggest auto manufacturers including Ford are looking to gain insights from the vast amounts of data within their processes.  Michael Cavaretta, a technical leader in predictive analytics at Ford's research and advanced engineering group has used large data sets to try and predict what features and specs customers respond to the most.  The following are highlights from an interview with Mr. Cavaretta at the Open Group Conference in Newport Beach, California.

Sea change in attitude

The biggest change in the past 5 years supporting big data analysis according to Mr. Cavaretta is the availability of storage and the increasing processing power of computers.  The "sea change in attitude" started with a view that all data is good, treating it as a fixed asset, and not worrying about the immediate associated costs.  An interesting point Cavaretta makes also is that he has yet to find a case where somebody is giving him too much data.  Given all the data you never have to worry about answering a question because some information was thrown out.. It is all there.

Internal consultants

 Within the research and advanced engineering dept at Ford, each member treats their job as an internal consultant would.  This means going out and trying to find opportunities for growth internally from a business perspective.  In this context new technologies have helped, especially with big data technologies where data can be stored and analyzed on the spot. On the notion of obtaining a total feedback loop across the lifecycle of an automobile, Caveretta explains that the potential is there but there's still work in getting the data from multiple sources. 

Huge number of sensors

There is a large amount of data coming from a vehicle.  This data can be captured by sensors and processors within the vehicle.  One group, the University of Michigan Transportation Research Institute is investigating the communication between a person's vehicle and their home system.  They are gaining insight into things such as letting your home know when you're on the way so that the system can start running the a/c or heat. 

According to the energi, cars generate about 25 gbs/hr of information.  With the amount of data being generated from vehicles, an analysts challenge becomes filtering which information is important or not.  Finding mashups between information from external sensors, temperature, humidity, etc is something very cool to think about.

 The link to the article can be found at:

http://www.zdnet.com/ford-scours-for-more-big-data-to-bolster-quality-improve-manufacturing-streamline-processes-7000010451/