Archive for May, 2015

May 19th 2015

Wise Practitioner – Manufacturing Predictive Analytics Interview Series: Jeffrey Thompson of Robert Bosch, LLC

By: Bala Deshpande, Conference Co-Chair, Predictive Analytics World for Manufacturing 2015 

In anticipation of his upcoming Predictive Analytics World for Manufacturing conference Jeffrey_Thompsonpresentation, Manufacturing Analytics at Scale: Data Mining and Machine Learning inside Bosch, we interviewed Jeffrey Thompson, Senior Data Scientist at Robert Bosch, LLC. View the Q-and-A below to see how Jeffrey Thompson has incorporated predictive analytics into manufacturing at Robert Bosch, LLC. Also, glimpse what’s in store at the PAW Manufacturing conference.

Q: In your work with predictive analytics, what behavior do your models predict?

A: At Bosch we work on a wide variety of different use-cases.  The target applications of our predictive models include manufacturing, supply chain and logistics, engineering, and Internet of Things and Services.    

Q: How does predictive analytics deliver value at your organization? What is one specific way in which it actively drives decisions?

A: Predictive models are used across products, processes, and operations at Bosch. An example use of our predictive models is in providing important insight into the root causes of failures on manufacturing lines.  These insights often lead directly to the resolution of a problem and are an important part of our continuous improvement efforts.

Q: Can you describe a successful result, such as the predictive lift (or accuracy) of your model or the ROI of an analytics initiative?

A: In one case, we used a predictive model to narrow down the root cause of a particularly expensive, internal defect in one of our automotive manufacturing lines.  The relative improvement in this case was 85%.   

Q: What surprising discovery have you unearthed in your data?

A: Data often tells you many things that you are not looking for and hence offers many surprises in all our projects. For example, many data sources or measurements never change over the course of a product’s or a process’ lifetime. It might be expensive to measure them, yet they continue to be measured based on an initial and incomplete design specification.

Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World for Manufacturing.

A: In many of our applications, high accuracy, and especially, low false alarm rates are critical. In such scenarios, even with large amounts of data, we find that better features often beat better algorithms, and subject matter experts are the key to getting those features. 

—————————-

Don't miss Jeffrey Thompson’s conference presentation, Manufacturing Analytics at Scale: Data Mining and Machine Learning inside Bosch, at PAW Manufacturing, on Tuesday, June 9, 2015, from 2:40-3:25 pm. Click here to register for attendance. 

By: Bala Deshpande, Founder, Simafore and Conference Co-Chair of Predictive Analytics World for Manufacturing.

No Comments yet »

May 14th 2015

Wise Practitioner – Manufacturing Predictive Analytics Interview Series: Kumar Satyam of PricewaterhouseCoopers, LLP

By: Bala Deshpande, Conference Co-Chair, Predictive Analytics World for Manufacturing 2015

In anticipation of his upcoming Predictive Analytics World for Manufacturing conference co-presentation, Utilizing On Board Technologies To Improve Maintenance Practices in Airlines, we interviewed Kumar Satyam, Manager, Advisory at PricewaterhouseCoopers, LLP. View the Q-and-A below to see how Kumar Satyam has incorporated predictive analytics into manufacturing at PricewaterhouseCoopers, LLP. Also, glimpse what’s in store for the PAW Manufacturing conference.

Q: In your work with predictive analytics, what behavior do your models predict?

A: PwC Analytics group works on a wide range of advanced analytics projects. Some of the examples of types of predictive models developed are customer choice, market response, demand forecasting, lifetime value analysis, sentiment analysis, failure modeling, supply chain optimization and propensity modeling

Q: How does predictive analytics deliver value at your organization? What is one specific way in which it actively drives decisions?

A: Through predictive analytics, PwC enables its clients to rapidly discover, quantify and deliver value from data with intelligent analytics and scalable end-to-end business solutions.  By enabling our clients to quantify the potential benefits from analytics engagements, we help them prove the value of predictive analytics to their respective organizations.

Q: Can you describe a successful result, such as the predictive lift (or accuracy) of your model or the ROI of an analytics initiative?

A: Recently PwC Analytics team and Travel and Transportation team worked together to develop a delay/cancellation prediction model for predictive maintenance for a large US carrier. Through a couple of months of  pilot test run, we were able to demonstrate that the developed model can help the carrier save up to 25-30% of the currently occurring maintenance related delays/cancellations for the modeled fleets and components. This effectively translates to millions of dollars’ worth of savings per year if expanded across fleets.

Q: What surprising discovery have you unearthed in your data?

A: One of the surprising discoveries from our analysis was that the maintenance logs of airlines contain lot of relevant information about a potential delay/cancellation that can occur in the future once the unstructured test data is analyzed.

Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World for Manufacturing.

A: Key Takeaway – Current computing power and analytics capabilities enable us to combine sensor data from aircrafts with unstructured maintenance log data to predict a significant percentage of delay/cancellations before they occur. This can enable airlines to perform predictive maintenance resulting in improvement of on-time performance and customer satisfaction with reduction in delay associated costs.

———————-

Don't miss Kumar Satyam’s conference co-presentation, Utilizing On Board Technologies To Improve Maintenance Practices in Airlines, at PAW Manufacturing, on Wednesday, June 10, 2015, from 3:30-4:15 pm. Click here to register for attendance. 

By: Bala Deshpande, Founder, Simafore and Conference Co-Chair of Predictive Analytics World for Manufacturing.

No Comments yet »

May 12th 2015

Wise Practitioner – Predictive Analytics Interview Series: Thomas Schleicher of National Consumer Panel

By: Eric Siegel, Founder, Predictive Analytics World

In anticipation of his upcoming conference presentation, The Importance of an Early Win in Promoting a Greater Predictive Analytics Capability, at Predictive Analytics World Chicago, June 8-11, 2015, we asked Thomas Schleicher, Sr. Director, Measurement Science at National Consumer Panel (by Nielsen), a few questions about his work in predictive analytics.

Q: In your work with predictive analytics, what behavior do your models predict?

A: The simple answer is attrition. Our prospective panelists who scored well according to our predictive model were shown to have a statistically significant, lower 12-month attrition rate than those who the model scored as poor prospective panelists. Panelists, for us, are analogous to subscribing customers in other industries, such as telecommunications.

Q: How does predictive analytics deliver value at your organization? What is one specific way in which it actively drives decisions?

A: Because we are obligated to maintain a 100,000 member consumer panel as a client KPI, reducing attrition is critical. Deploying our predictive model lowers attrition, reducing our need to recruit new panelists, which of course also has the effect of reducing recruitment costs.

Q: Can you describe a successful result, such as the predictive lift of your model or the ROI of an analytics initiative?

A: The predictive model lowered our rate of attrition by an average of 3.5% over two independent evaluations of its impact. We’re still working on establishing the best gauge of ROI, given that there are direct, short-term benefits achieved through reduced recruitment costs, as well associated savings through a lower need to purchase scanners for panelists. Senior management is interested in ongoing improvement of the model as well as clearer measures of its value.

Q: What surprising discovery have you unearthed in your data?

A: Some of the variables that were part of the predictive mix were not immediately intuitive. This has helped to reinforce the Measurement Science team’s quest to develop an integrated database that will enable expanded predictive capability.

Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World.

A: I plan to share how the success of this model has helped to promote not just the benefit, but the need for a predictive analytics capability. I believe our success is applicable to other organizations also interested in developing and/or expanding a data driven culture that adds true value. Knowing where to start is critical.

———————

Don't miss Thomas Schleicher’s conference presentation, The Importance of an Early Win in Promoting a Greater Predictive Analytics Capability, at Predictive Analytics World Chicago on Wednesday, June 10, 2015 from 4:15-5:00 pm.  Click here to register for attendance.

By: Eric Siegel, Founder, Predictive Analytics World

No Comments yet »

May 7th 2015

Wise Practitioner – Predictive Analytics Interview Series: Delena D. Spann of US Government

By: Eric Siegel, Founder, Predictive Analytics World

In anticipation of her upcoming conference presentation, 21st Century Fraud Analytics (Detection) in a Predictive Analytics World, at Predictive Analytics World Chicago, June 8-11, Delena_Spann2015, we asked Delena D. Spann, Fraud Analytics Expert at a United States Government Agency, a few questions about her work in predictive analytics.

Q: In your work with predictive analytics, what behavior do your models predict?

A: The behaviors that my models predict are within the elements of fraud.

Q: How does predictive analytics deliver value at your organization? What is one specific way in which it actively drives decisions?

A: I can't speak as to how it delivers value within my organization due to it being comprised of a wide range of specific divisions. However, I will say that within the realms of academia in which I am an Adjunct Professor as well; its value has provided me with an in-depth understanding of my findings in a credit card fraud case study. It drives decisions based on the mere fact that it allows one to gain a better understanding of the variables that are derived directly or indirectly to the findings of concern. 

Q: Can you describe a successful result, such as the predictive lift of your model or the ROI of an analytics initiative?

A: Within the ROI analytics initiative the predictive analytics methodology used ensured the most thorough approach to deterring and detecting fraud in the credit card industry. 

Q: What surprising discovery have you unearthed in your data?

A: Although, using predictive analytics can be an exhaustive and reiterative process with built-in flexibilities; it can enhance the looks for stability and repeatability of its findings. 

Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World.

A: The one take-away is that Predictive Analytics is the emerging tool of the 21st Century.

———————

Don't miss Delena D. Spann’s conference presentation, 21st Century Fraud Analytics (Detection) in a Predictive Analytics World, at Predictive Analytics World Chicago on Tuesday, June 9, 2015 from 3:05-3:25 pm.  Click here to register for attendance.

By: Eric Siegel, Founder, Predictive Analytics World

No Comments yet »

May 5th 2015

Wise Practitioner – Predictive Analytics Interview Series: Dr. Patrick Surry of Hopper

By: Eric Siegel, Founder, Predictive Analytics World

In anticipation of his upcoming conference presentation, Buy or Wait? How the Bunny Predicts When to Buy Your Plane Ticket, at Patrick_SurryPredictive Analytics World Chicago, June 8-11, 2015, and keynote at Predictive Analytics World Boston, Sept 27-Oct 1, 2015, we asked Dr. Patrick Surry, Chief Data Scientist at Hopper, a few questions about his work in predictive analytics.

Q:  In your work with predictive analytics, what behavior do your models predict?

A: At Hopper we predict how airfares are likely to move so that we can advise consumers whether they should buy now or wait for a better price.

Q:  How does predictive analytics deliver value at your organization? What is one specific way in which it actively drives decisions?

A: Our goal at Hopper is to help consumers save money on flights with data-driven advice about when to fly and buy, and even where to go.   Airfare predictions are a core part of our value proposition: On average the “buy or watch” advice in our app saves consumers 10% over the first price they see for a trip.

Q:  Can you describe a successful result, such as the predictive lift of your model or the ROI of an analytics initiative?

A:  In extensive back-testing across tens of thousands of trips, we’ve shown that 95% of the time our “buy or wait” recommendations will either save the consumer money or do no worse than the first price they saw, saving 10% on average.

Q:  What surprising discovery have you unearthed in your data?

A:  We’ve been surprised at how much airfare fluctuates.  Although on average airfare rises as departure approaches, watching continuously is very effective at capturing lower prices.  More than half the time we’re watching a trip, we’ll see a price that’s 5-10% lower than our initial price within just 24 hours.

Q:  Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World.

A:  Airfare trends are surprisingly predictable, enabling consumers to save up to 40% on their flights using the Hopper app, while avoiding the frustration of manual comparison shopping.  Top tips for saving: Start watching prices early, be flexible with dates and destinations if possible, and do your homework for the route(s) you’re interested in so you know what prices other people are paying.  One example: Fares rise higher for routes with lots of business travelers, so you won’t find a last minute deal from New York to San Francisco, but you might find one from New York to Hawaii.

———————

Don't miss Patrick Surry’s conference presentation, Buy or Wait? How the Bunny Predicts When to Buy Your Plane Ticket at 1) Predictive Analytics World Chicago on Wednesday, June 10, 2015 from 11:15 am-12:00. Click here to register to attend.  2) Keynote presentation at Predictive Analytics World Boston on Monday, September 28, 2015 at 1:30-2:15 pm. Click here to register to attend.

By: Eric Siegel, Founder, Predictive Analytics World

No Comments yet »