Wednesday, February 6, 2013

The Future of Engineering Dataism

In his Tuesday New York Times column (The Philosophy of Data), David Brooks had an important question that engineering needs to be thinking about.  The column covered several of the issues and opportunities created by the data revolution.

Brooks closed the column with the following:

"In sum, the data revolution is giving us wonderful ways to understand the present and the past.  Will it transform our ability to predict and make decisions about the future?  We'll see."

Both the medical and engineering professions exist in three worlds - - the past, present, and future.  Future to both professions is measured in decades - from the new baby just delivered to the new bridge that needs to be functional next century.  The is different than law and accounting.  Their focus is much more about the past and the moment (hang around an accountants office on April 14th).

The data revolution has the potential to change our ideas regarding the future.  In the context of engineering, three ideas or issues regarding the future are important.  The first is pattern recognition.  The data revolution will allow engineers greater access to a wealth of historical data as well as real-time data.  Finding patterns in the data will be highly critical.  Seeing both the past and present in greater clarity allows for a better understanding of potential future events.

The second issue critical to engineering is the idea that data and the data revolution will illuminate patterns of behavior we haven't yet noticed.  Using data and statistical tools to look at previous and real-time behavior for human-to-human and human-to-system interactions will open up new fields and opportunities for engineers.  Using the past and present to model future behaviors will be critical - - from highway design to airport terminal layouts to drive-thru bank system designs.

The third issue will be a question increasingly posed to the engineering community:

"What kinds of events are predictable using statistical analysis and what sorts of events are not?"

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.