The INFORMS Conference on Business Analytics and Operations Research kicks off April 15 in Huntington Beach, CA. I had a chance to preview a presentation by Glenn Bailey, Sr. Director of Operations Research at Manheim (the $3B wholesaler auto auctioneer). Glenn's talk is on "The Need for Speed: Responsive Predictive Analytics," and he makes an important statement:
There's no correlation between Analytic Complexity and Business Value -- so conduct your analytics accordingly
We are in an age of amazing possibilities when it comes to the big data at our fingertips, and the high-performance analytics available for extracting knowledge and improving decisions. The technology lets us attack more types of problems, approaching them from different angles with more methods, and to do all this in much less time than it ever took before.
While the availability of this technology is a blessing, and our love of the technology is much deserved, we mustn't lose sight of the business problems we use this technology to solve. And, as Glenn asserts, we mustn't lose sight of the business value we are attempting to deliver.
For example, it is well-established in forecasting that simple methods can, and frequently do, perform better than more complicated methods. A moving average or simple exponential smoothing will be the most appropriate forecasting model in some situations, and there is no shame in that. While more complex or sophisticated methods can always give our model a better fit to history, this does does guarantee better forecasts. (See Makridakis, et al, Forecasting Methods and Applications (pp. 526-527) for a brief but excellent discussion of simple versus complex methods.)
Supply Chain & Forecasting Tracks
While at INFORMS, some other noteworthy presentations include:
- Sam Iosevich of Prognos on "How can you integrate demand planner knowledget into the statistical forecast"
- Kean Chew of HAVI Global Solutions on "The numbers behind burgers and fries"
- Robert Fildes of Lancaster University Centre for Forecasting on "Improving supply chain forecasting"
- Anne Robinson of Verizon on "Influencers, Skeptics, and Data Geeks"
- Manoj Chari of SAS on "Tools and Approaches for Big Analytics"