August 22nd 2016

Wise Practitioner – Predictive Analytics Interview Series: Scott Zoldi at FICO

By: Eric Siegel, Founder, Predictive Analytics World

In anticipation of his upcoming conference keynote presentation, Fraud Screening for 2/3rds of All Card Transactions: A Consortium and Its Data at Predictive Analytics World Financial in New York City, October 23-27, 2016, we asked Scott Zoldi, Chief Analytics Scot Zoldi IMAGEOfficer at FICO, a few questions about his work in predictive analytics.

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

A:  Falcon models generate transaction scores which determine the likelihood of the transaction being fraudulent, by incorporating multi-faceted fraud risk assessments from the cardholders’ historic spending behavior, characteristics of merchants and transaction instruments. The higher the score is, the more likely the transaction is fraudulent.

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

A: Falcon models are data-driven analytical models built from the FICO Fraud data consortium where banks agree to pool their payment card transaction and fraud data for building the most predictive models. These predictive models mitigate the financial impact of credit and debit card fraud and improve customer experiences.  The pursuit of the highest performing model has driven the creation of a payment card fraud consortium data asset that contains more than two decades of historical data. 

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

A: Since introducing Falcon models based on the data consortium, we monitor 2.5 billion payment cards world-wide, and the payment card fraud losses have been reduced by more than 70% in the US. The data consortium has enabled the creation of 88 granted fraud analytic patents and more than 44 fraud analytic patent applications pending derived from over 3,000 researcher-years working with the asset to maximize fraud detection while minimizing customer impact.

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

A: One is that in 2014 nearly half of the fraudulent cross-border transactions on UK debit cards took place in the US, as US had not yet implemented EMV technology. The card-not-present transactions (for example, online purchase) are seen to grow rapidly from year to year in volume and the fraud rate is much higher than card-present transactions. The data also reveal the occurrence of data compromise events associated with some merchants that lead to exposure to card data in the dark web.

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

A:  One take away that will be discussed is how we utilize auto-encoder technologies to ensure that the data on which our models are built is similar to the data seen in production.  This provides an ability to monitor data through these deep learners in order to determine when we may have models that will perform sub optimally based on transaction pattern shifts in production data environments.

———————

Don't miss Scott’s conference keynote presentation, Fraud Screening for 2/3rds of All Card Transactions: A Consortium and Its Data at Predictive Analytics World Financial NYC, on Wednesday, October 26, 2016 at 8:50 to 9:40 am. Click here to register to attend.

By: Eric Siegel, Founder, Predictive Analytics World

No Comments yet »

August 19th 2016

Wise Practitioner – Predictive Analytics Interview Series: Thomas Klein at Miles & More GMbH

By: Eric Siegel, Founder, Predictive Analytics World

In anticipation of his upcoming conference co-presentation, Using Predictive Models for Demand Simulation – Purchase, Response and Uplift Modeling in Practice, at Predictive Analytics World New York, October 23-27, 2016, we asked Thomas Klein, Associate Thomas_Klein IMAGEDirector Advanced Analytics at Miles & More GmbH, a few questions about his work in predictive analytics.

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

A: Our advanced analytics team at Miles & More is responsible for turning data into insights that help our organization to make the right decisions and choose the right measures to take action. The larger part of business problems we work on are focused on optimizing our multichannel marketing campaigns and creating a better customer experience: We build predictive models in order to better target marketing campaigns and deliver personalized recommendations. We also use analytics to identify, predict and test meaningful events that serve as trigger to provide relevant information and create marketing offers based on the customer's context and current needs. Always with the aim to create and improve customer loyalty for our program and for partner companies.

Precisely, predictive analytics helps us to answer questions like the following:

  • Which customers will respond to a marketing campaign?
  • Which customers will book their flight on an airline of the Lufthansa Group or Star Alliance rather than the competition?
  • Which customers will book a hotel room, rent a car or buy a product from one of our various non-aviation Miles & More partner companies?
  • Which products or services fit with the customers’ preferences and needs?

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

A: With more than 29 million participants, Miles & More is the largest frequent flyer and awards program in Europe. Miles & More has more than 40 airline partners, including the Lufthansa Group and 28 airlines of the Star Alliance. In addition there are already more than 270 non-aviation companies affiliated with Miles & More, these include partners from the hotel, car rental and cruise industries, banks and insurance, telecommunications and electronics industries as well as shopping and lifestyle.

Put in a nutshell, our large partner network offers a wide variety of different products and services and our goal is to provide each customer with the most relevant offers, services and information based on his or her individual preferences, interests and current needs.

Predictive models enable us to choose the best offers or the most relevant information to dynamically tailor content and use the right channels to interact with our customers based on the data available. More personalized marketing communicaton instead of a one-size-fits-all approach generates additional revenue and results in higher customer satisfation. At the same time we can automate the decision process as well as our marketing campaign workflows to increase operational efficiency and save costs.

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

A: The answer to this question is not as straightforward as it seems, since the predictive lift always depends on the baseline you use to measure your results and we often try to improve models and processes which are already pretty optimized and mature. When it comes to improving marketing campaign it is also very important to set the right expectations and keep in mind that predictive analytics is a critical success factor, but still only one piece of the puzzle and no panacea for all challenges.

From my point of view, the recipe for success is doing a lot of experimentation and accept that you will fail more often than you succeed. During this process most improvements will result from the aggregation of marginal gains and sometimes you will hit a long shot (https://www.youtube.com/watch?v=ohCavVVxX0M).

And to finally give you one specific example; which we also presented in our case study on uplift modelling at Predictive Analytics World Berlin in 2014. Using an uplift modelling approach we were able to increase the lift of direct marketing campaigns for our Lufthansa Miles & More Cedit Card by a factor of 3.

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

A: Our analytics team was often asked about the typical customer of a given partner, product or service. The expectation was almost always to get a demographic segmentation or description of the perfect target group. When we started to educate colleagues from various departments on how a multivariate predictive model works they were often surprised to learn, that very different customers are actually equally valuable and that a predictive model always beats an old-school demographic or heuristic segmentation approach. 

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

A: Acting as awards program and expert in customer loyalty for more than 300 partners across all industries and providing a marketing platform to interact with 29 million customers certainly has its own and unique challenges.

We want to show some use cases about how we successfully apply predictive analytics at Miles & More and share our practical experience about what works best in such a setting. We hope to give our audience some new ideas they can take-away and apply to their specific challenges within their own business setting. We are looking forward to a fruitful discussion, in our Q&A session as well as during the whole conference.

———————

Don't miss Thomas’ conference presentation, Using Predictive Models for Demand Simulation – Purchase, Response and Uplift Modeling in Practice, at Predictive Analytics World New York on Tuesday, October 25, 2016 from 11:20 am to 12:05 pm.  Click here to register for attendance.  

By: Eric Siegel, Founder, Predictive Analytics World

No Comments yet »

August 15th 2016

Wise Practitioner – Predictive Analytics Interview Series: Meina Zhou at Bitly

By: Eric Siegel, Founder, Predictive Analytics World

In anticipation of her upcoming conference presentation, Predictive Analytics for Different Business Types:  Optimize All the Funnels, at Predictive Analytics World New York, Meina Zhou IMAGEOctober 23-27, 2016, we asked Meina Zhou, Data Scientist at Bitly, a few questions about her work in predictive analytics.

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

A: My marketing qualified lead scoring model helps to identify high-quality leads. My churn prediction model helps to identify the current customers that have the highest probability to churn. My upsell prediction model helps to identify the best customers to upsell.

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

A: My predictive analytics helps the customer success team to better understand our customers’ behaviors and to identify key factors churning. It also helps the customer success team to design proactive retention actions accordingly.

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

A: The marketing team compared my model with third-party software and determined that my model was on par. Building the software in-house would allow us to save the money we would have spent on a third-party tool.

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

A: Some of the customers churned because they never understood how our product worked. They decided to pay for the product before they knew how to use it.

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

A: Utilize the usage data of your product to understand the customers’ engagement with your product. The change in the customers’ usage behavior can be key factors for churning and upselling.

———————

Don't miss Meina’s conference presentation, Predictive Analytics for Different Business Types:  Optimize All the Funnels, at Predictive Analytics World New York on Wednesday, October 26, 2016 from 11:15 to 11:35 am.  Click here to register for attendance

By: Eric Siegel, Founder, Predictive Analytics World

No Comments yet »

August 12th 2016

Wise Practitioner – Predictive Analytics Interview Series: Dr. Shantanu Agrawal at Centers for Medicare & Medicaid Services

By: Sean Robinson, Program Chair, Predictive Analytics World for Government

In anticipation of his upcoming conference keynote presentation, Implementing Predictive Analytics at CMS: Lessons Learned and Future Directions at Predictive Analytics World for Government, October 17-20, 2016, we asked Dr. Shantanu Agrawal, Deputy Shantanu Agrawal Pic 08-2016Administrator for Program Integrity and Director of the Center for Program Integrity at the Centers for Medicare & Medicaid Services (CMS), a few questions about his work in predictive analytics.

Q: How would you characterize your agency’s current and/or planned use of predictive analytics? What is one specific way in which predictive analytics drives decisions in your agency?

A: Predictive Analytics is the backbone of the Centers for Medicare & Medicaid Services (CMS) Fraud Prevention System (FPS), which applies proven and effective predictive modeling tools into the Medicare claims processing system. The FPS was a result of a CMS mandate by the Small Business Jobs Act of 2010 (SBJA), Section 4241, to implement a predictive analytics system to analyze Medicare claims to detect patterns that present a high risk of fraudulent activity. Signed by President Obama in fall 2010, the SBJA enabled CMS to stop payment on high-risk claims, and perform post-pay analysis to identify emerging trends of potentially fraudulent activity.  The FPS analyzes information from multiple Medicare data sources (Part A, Part B, and DME claims, Composed Numbers Checklist, FID, and 1-800-Medicare Tips) to predict whether observed billing patterns or trends are likely to be fraudulent, similar to the way the credit card industry evaluates the consistency of a cardholder’s new charges against past transactions as a way of identifying potential fraud. 1 CMS designed the FPS to accommodate 4 model types including rule-based, anomaly, and social network analysis models, in addition to predictive models. These model types address multiple kinds of fraud schemes. In addition, they work together to increase the effectiveness of one or more models to identify potential and actual fraudsters.

Since June 30, 2011, the FPS has run predictive algorithms and other analytics nationwide on a daily basis against 4.5 million Part A and Part B claims to generate alerts for Program Integrity (PI) purposes.  The FPS alerts when egregious, suspect, or aberrant activity is identified, and automatically generates and prioritizes leads for review and investigation by the Agency.

1 Department of Health and Human Services Centers for Medicare & Medicaid Services, MLN Matters, Number SE1133, 2011.

 

Q:  Can you describe the challenges you face or have already overcome in establishing a data-driven environment in your agency?

A: The FPS “big data” effort has had a profound impact by allowing CMS to quickly identify issues and take action. CMS has saved over $1.5 billion in inappropriate payments since implementing this system, producing a 11:1 return-on-investment in the latest year of operations.

There are a number of things an organization has to do well to really drive value from data intensive approaches. First, often disparate data has to be compiled into a single usable structure and technology solution that promoted ease of use and allows business-focused analytics. This can only be done by engaging end users early and promoting the use of data. Second, analytics have to be incorporated into existing business processes. In our case, program staff, policy analysts, and investigators had to be convinced that “big data” would make their work more efficient and effective so that it became part of their usual work-flow, instead of an add-on. Third, the business has to constantly feedback on the analytics to make sure that useful, high yield models are constantly being designed and rolled-out – and where necessary, models and analytical approaches are being improved to provide more actionable results to the business. Finally, you have to establish performance metrics and measure results, to know how “big data” is impacting work and hopefully making it better. Data without a measurable ROI is not necessarily an improvement.

Done right, big data leads to more confident decision making, and better decisions can result in greater operational efficiency, cost reduction and reduced risk. For example, the FPS identified a home health agency in Florida that billed for services that were never rendered. As a result, CMS placed the home health agency on prepayment review and payment suspension, referred the agency to law enforcement, and ultimately revoked the agency’s Medicare enrollment. In Texas, the FPS identified an ambulance company submitting claims for non-covered services and services that were not rendered. Medicare revoked the ambulance company’s enrollment. Likewise, FPS identified that an Arizona, medical clinic had questionable billing practices, such as billing excessive units of services per beneficiary per visit. Upon review of medical records, it was discovered that physicians had been delivering repeated and unnecessary neuropathy treatments to beneficiaries. That medical clinic was revoked from Medicare enrollment. 

Q: Can you discuss any near term goals you have for improving your agency’s use of predictive analytics?

A: Now in its fifth year, CMS has begun work to build a new and more robust FPS system (FPS2) that is scheduled to go into production the first quarter of CY 2017. FPS2 will be enhanced to improve predictive analytics by (1) decreasing the time it takes to market and recode models to custom language, (2) expanding the toolset to include Social Network Analysis visualization, (3) improving the user interface based on previous and past user studies, (4) implementing the ability to suspend payments within the tool, and (5) providing an easy to use computer interface, enabling users to access and view claims level data in maps and charts.  Ultimately, FPS2 will continue to increase the CMS CPI return on investment noted in the original FPS.

———————

Don't miss Dr. Agrawal’s conference keynote presentation, Implementing Predictive Analytics at CMS: Lessons Learned and Future Directions on Monday, October 17, 2016 from 1:00 to 1:45 pm at Predictive Analytics World for Government. Click here to register to attend. 

By: Sean Robinson, Program Chair, Predictive Analytics World for Government

No Comments yet »

August 10th 2016

Wise Practitioner – Predictive Analytics Interview Series: Madhusudan Raman at Verizon

By: Eric Siegel, Founder, Predictive Analytics World

In anticipation of his upcoming conference presentation, Best Practices Enhancing Contextual Experience with Predictive Analytics, at Predictive Analytics World New York, October 23-27, 2016, we asked Madhusudan Raman, Innovation Incubator at Verizon, a Madhusudan_Raman IMAGEfew questions about his work in predictive analytics.

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

A: As in the past, Contextual Intent, Contextual Mood, Internet Advertising Bureau (IAB) Context, Contextual Interestingness, and Propensity to Buy/Act continue to be strong prediction targets.

Q: What emerging needs are you seeing related to the behaviors your models predict?

A: Recently the prediction of Thing behaviors to identify Contextual Interestingness has gained in importance due to the recent push to extend the embedded device boundaries to include cloud activity; in other areas, there seems to be increased interest in Propensity for Attrition where a body of best practices are emerging.

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

A: Knowing which market incubations to pursue and which ones to put on hold is a valuable insight. Predictive Analytics – specifically scoring ‘Propensity to Succeed’ has been a helpful operational support measure for prioritization of resource constrained pursuit of ideations.

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

A: Applying Context (Information) Gain to constrain datasets increases the resulting lift from the model deployment from 2X-9X when applied to a range of Consumer Experience enhancement scenarios.

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

A: One interesting discovery was that archetype (‘think extreme’) analysis seems to always be more effective than standard statistical (‘think average’) measures and analysis for predictive segmentation incorporating near-real-time sensor streams from Things.

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

A: Contextual understanding enhances Consumer Experience. We will share a systematic Predictive Analytic technique for dealing with context, its quantification within a dataset, and how channels and touchpoints can use this derived “context” in near-real-time across Industry verticals especially with Thing Sensor data.

———————

Don't miss Madhusudan’s conference presentation, Best Practices Enhancing Contextual Experience with Predictive Analytics, at Predictive Analytics World New York on Wednesday, October 26, 2016 from 4:20 to 5:05 pm.  Click here to register for attendance

By: Eric Siegel, Founder, Predictive Analytics World

No Comments yet »

August 8th 2016

Wise Practitioner – Predictive Analytics Interview Series: Sanjay Gupta at PNC Bank

By: Eric Siegel, Founder, Predictive Analytics World

In anticipation of his upcoming conference co-presentation, Predictive Analytics for Stress Testing – Industry Challenges at Predictive Analytics World Financial in New York City, October 23-27, 2016, we asked Sanjay Gupta, Executive Vice President and Head of Sanjay Gupta IMAGEModel Development at PNC Bank, a few questions about his work in predictive analytics.

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

A: We predict defaults of our clients and losses in our portfolio. This is used for the calculation of the Capital, Reserves and Profits of the Bank.

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

A: The predictive analytics helps us acquire customers, evaluate losses and defaults. By looking at the drivers of default or losses, we can design account management strategies that can mitigate loss or default. It can also provide early warnings that can help to better inform our stakeholders.

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

A: Data does not lie; sometimes our ability to see the truth is hampered by our biases.

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

A: Looking at fundamental credit analysis of a client and knowing your client can truly add value to your predictive work.

———————

Don't miss Sanjay’s conference co-presentation, Predictive Analytics for Stress Testing – Industry Challenges at Predictive Analytics World Financial NYC, on Tuesday, October 25, 2016 at 11:20 am to 12:05 pm. Click here to register to attend.

By: Eric Siegel, Founder, Predictive Analytics World

No Comments yet »

August 1st 2016

Wise Practitioner – Predictive Analytics Interview Series: Brian Reich, Former Director at The Hive

By: Eric Siegel, Founder, Predictive Analytics World

In anticipation of his upcoming conference presentation, The Data Set You Can No Longer Ignore: Consumer Engagement with Social Issues, at Predictive Analytics World New York, October 23-27, 2016, we asked Brian Reich, former Director at The Hive, a Brian Reich IMAGEfew questions about his work in predictive analytics.

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

A: We started by analyzing current donors. From our current donors, we created lookalike models that gave us an appreciation for who was already responding and taking action to support refugees – and, of course, the much larger universe of people who were not yet engaged. We also conducted a national survey to measure people’s knowledge and support of refugee aid, along with a host of related issues, and we used that information to build models that explored the population beyond the lookalikes – what they know about refugees, what they might need to understand better in order to be motivated to take action, and similar.

We started all this work when awareness and interest in the global refugee crisis was pretty limited – and certainly confined to only people who were already deeply knowledgeable on the subject. Then, at the end of last summer, that all changed. The number of refugees flooding out of Syria into Europe exploded. Heart-breaking pictures of a young refugee who drowned trying to make it to safety suddenly made headline news and took over social media channels. All of a sudden, the global refugee crisis was the top story all over the world. President Obama challenged Americans to step up and help. The Pope said it was our moral responsibility as a society to aid refugees. The attention and the interest in the refugee crisis was higher than ever before. And we were able to organize our response, both short-term and long-term using the data as a guide.

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

A: Everything is informed by the data, both directly and indirectly. We have an unprecedented level of sophistication that we can apply – who we target, how we position issues related to the refugee crisis, and the ways we work with partners. Beyond our specific efforts:

  • We have demonstrated that we aren’t a traditional non-profit organization, but rather a sophisticated, engagement-oriented organization that approaches engagement with a combination of political, consumer, media, and tech expertise that hasn’t been attempted before.
  • We have clear insights to our audience and can prioritize accordingly – not just look at who might engage based on their past behaviors, but look to engage people who are likely to engage if given the right opportunity.
  • We build our messages based on the rapid-message testing, so we don’t have to rely on what our gut tells us.
  • We know the best ways to reach individuals, whether digital ads or snail mail, through partners, or another method, so we don’t waste time or resources on audiences that aren’t likely to take action.

And with all of this, we’ve been able to improve our organizational capacity and shift the way we think about our challenges. We’re looking at individuals – human beings – not donors, or advocates. We can do a much better job tapping into what we know people are already doing, or comfortable doing, instead of trying to compel people to do what we think benefits our organization best. Our cause is important, but we need to think more about the people we are talking to – understanding who they are and what motivates them to ultimately engage them and support our cause.

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

A: Prior to doing the modeling, we had been making assumptions based on our gut – and they were pretty good. While the data validated our assumptions, it also revealed a whole new set of opportunities. It uncovered some surprising untapped geographic markets, hotspots in the parts of the country that we had never thought of, groups of people who don’t fit the stereotype of those who would have been our targets. Over the past several months we have used the data to reach and engage hundreds of thousands of new people who would not have been targetable without the data.

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

A: Beyond target audiences, the data has expanded the ways we think about how to engage people. Instead of relying on the existing messages – the ways you commonly hear about the refugee crisis from a nonprofit or through media coverage – we now had messages that have been tested and verified by data science. Instead of speaking about the refugee crisis as an emergency situation, or telling stories of the horror refugees face or hope they retain against all odds, we can present issues in ways that Americans understand, connect with on a more personal level, or make sense of through an experience they can appreciate. All of these insights continue to show how data enables us to target more efficiently and effectively.

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

A: We knew that the traditional methods and messages wouldn’t be enough to get new potential supporters engaged. It was clear that Americans cared about helping refugees, but we needed to know more so we could determine how to get people more involved, to take more meaningful actions. We used our data models, plus survey research and rapid-response message testing, to experiment with different efforts we thought could be most effective, and learn the most in the shortest period of time.  And it paid off in a big way.

———————

Don't miss Brian’s conference presentation, The Data Set You Can No Longer Ignore: Consumer Engagement with Social Issues, at Predictive Analytics World New York on Wednesday, October 26, 2016 from 3:30 to 3:50 pm.  Click here to register for attendance.  

By: Eric Siegel, Founder, Predictive Analytics World

No Comments yet »

July 29th 2016

Wise Practitioner – Predictive Analytics Interview Series: Gary Neights at Elemica

By: Eric Siegel, Founder, Predictive Analytics World

In anticipation of his upcoming conference presentation, Predicting Behavior in Chemical Industry Supply Chains, at Predictive Analytics World New York, October 23-27, 2016, we asked Gary Neights, Senior Director at Elemica, a few questions about his work in Gary Neights IMAGEpredictive analytics.

Q: What are the challenges in translating the lessons of predictive analytics from other verticals into manufacturing?

A: Predictive models for pharma and retail are often used to influence consumer behavior or give direction to research efforts. These models yield results that are reviewed by experts before being acted upon.   The predictive system for manufacturing supply chains that I will discuss drive real-time manufacturing execution decisions by front-line employees.  Commitment of resources such as labor, manufacturing capacity, raw materials, and logistics capacity can occur in near real-time.   Accurate, real-time data flows from customers, distributors, suppliers, and carriers are required to develop a full picture of the situation and help provide the best level of decision support.

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

A: One example is under supply or over supply conditions.  Over supplying finished goods may lead to price discounting while under supplying material to a downstream manufacturing process may shut down operations.  Another example is predicting which perishable materials in a complex supply chain network are nearing expiration so they can be expedited to an appropriate manufacturing facility.

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

A: Supply chain decisions have financial impact and need to be taken in near real-time.   Product-by-product and plant-by-plant predictions can lead to information overload and indecision.   For example, if rail cars to a manufacturing site are predicted to be late do I dispatch trucks as a rush shipments… or dip into safety stock?   If trucks, how many?  Over the long-term data may be analyzed systematically and accounted for during periodic planning cycles or contract renegotiations.

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: A predictive Railcar KanBan system drove a working capital savings of greater than $400K / year for one product and inventory replenishment accuracy was increased from less than 55% accuracy to greater than 80%. This allowed a 20% reduction in safety stock levels and 40% reduction of leased railcars. 

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

A: In one case a graphical review of time-series data indicated that a manual supply chain management process was systematically driving costly inventory swings.  The planner was not correctly accounting for transit times between locations, nor the operating hours for shipping and receiving operations. This was corrected by a correctly tuned predictive system.

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

A: A common theme I hear is that the farther you are from the consumer the harder it is to get accurate demand data.  We will share one approach that supports manufacturers systematically aggregating demand to improve predictive accuracy.

———————

Don't miss Gary’s conference presentation, Predicting Behavior in Chemical Industry Supply Chains, and workshops at Predictive Analytics World New York on Tuesday, October 25, 2016 from 3:05 to 3:25 pm.  Click here to register for attendance.

By: Eric Siegel, Founder, Predictive Analytics World

No Comments yet »

July 27th 2016

Wise Practitioner – Predictive Analytics Interview Series: Dr. Sarmila Basu at Microsoft Corporation

In anticipation of her upcoming conference presentation, Predictive & Prescriptive Analytics Helps Keep Kids in School at Predictive Analytics World London, October 12-13, 2016, we asked Dr. Sarmila Basu, Senior Director, Data & Decision Sciences Sarmila Basu IMAGEGroup at Microsoft Corporation, a few questions about her work in predictive analytics.

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

A: We do predictive analytics for business groups within the company as well as for customers outside. For internal groups our predictive analytics involves things like:-

  1. Predictive models for customer attrition for a specific subscription based product.
  2. Models for antipiracy efforts to identify compromised license keys
  3. For some of our external customers we have built models for predictive maintenance to plan for potential failure of their machineries.

 

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

A: Predictive analytics has delivered impact through revenue enablement and cost savings. When we are able to identify pirated/compromised license keys, that helps with revenue recovery. In cases of customer attrition modeling, we were able to start proactive retention efforts.  

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

A: Our antipiracy analytic effort has led to revenue recovery worth $130million last year.

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

A: Surprising discovery varies from project to project. We sometimes find unexpected information or insight by combining multiple data sources. People often complain about their data being inadequate for any meaningful conclusion, but it is more an issue of broken data than bad data. Data Scientists have to be able to stich data from multiple sources together and build a cohesive story.

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

A: I will be talking about some of the use case scenario that has led not only to revenue impact for us but more importantly in terms of social good, it has delivered great impact.

———————

Don't miss Dr. Basu’s conference presentation, Predictive & Prescriptive Analytics Helps Keep Kids in School on Thursday, October 13, 2016 at 11:45 am at Predictive Analytics World London. Click here to register to attend.

No Comments yet »

July 22nd 2016

Wise Practitioner – Predictive Analytics Interview Series: Dae Park and Vijay D’Souza at Government Accountability Office (GAO)

By: Sean Robinson, Program Chair, Predictive Analytics World for Government

In anticipation of their upcoming conference co-presentation, Characteristics for Those Dae Park IMAGEClaiming Social Security Benefits Early, at Predictive Analytics World for Government, October 17-20, 2016, we asked Dae Park, Assistant Director at Government Accountability Office (GAO), and Vijay D’Souza, Director at Government Accountability Office (GAO), a few questions about their work in predictive analytics.

Q:  How would you characterize your agency's current and/or planned use of predictive analytics?  What is one specific way in which predictive Vijay D'Souza IMAGEanalytics actively drives decisions in your agency?

A: A core function of the Government Accountability Office (GAO) is to use data to objectively evaluate government programs and spending. We both develop our own models and evaluate agencies’ own models to evaluate the effectiveness of government programs. For example, we have completed numerous evaluations of Medicare payment policies to evaluate possible effects on quality and access to care.

Q:  Can you describe the challenges you face or have already overcome in establishing a data-driven environment in your agency?

A: The biggest problem we face is acquiring and validating data from the agencies we audit. Data are provided in multiple formats and often have quality issues. These include duplicate, incomplete, or inaccurate entries, and inconsistencies among records. This challenge is compounded when we merge data from multiple agencies or multiple sources to conduct more sophisticated analyses.

Q:  Can you discuss any near term goals you have for improving your agency's use of predictive analytics?

A: There is an increasing awareness across our analyst community of the value of predictive analytics to measure the effectiveness of federal programs. This includes training and outreach efforts from our analytics experts to other agency staff. We are also taking greater advantage of open source tools for analytics and developing strategies to increase the amount of staff that can help with the initial stages of data acquisition, evaluation, and preparation.

Q:  Can you describe a successful result from the employment of predictive analytics in your agency, i.e., cost avoidance, funds recovered, improved efficiency, etc.

A: We analyzed TSA’s program to detect more risky air travel passengers and determined it was not based on sound evidence. In another case, we evaluated a Department of Transportation program that used a large amount of data to calculate safety scores for commercial trucking carriers and to ostensibly determine the likelihood of crashes. We found that violations used in the agency model either occurred too infrequently to be useful, or were not otherwise predictive of crashes.

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

A: To better understand the circumstances faced by those who claim early Social Security benefits, GAO examined demographic and occupational characteristics associated with early claiming. More specifically, we examined the characteristics and income of early claimers using data from the Health and Retirement Study. I look forward to discussing our findings in detail at the session.

———————

Don't miss Dae and Vijay’s conference co-presentation, Characteristics for Those  Claiming Social Security Benefits Early, on Monday, October 17, 2016 from 11:25 am to 12:10 pm at Predictive Analytics World for Government. Click here to register to attend.

By: Sean Robinson, Program Chair, Predictive Analytics World for Government

No Comments yet »

Next »

  • Subscribe Via Email

    Get a daily digest of new posts delivered to your inbox:

  • Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die
    BOOK AWARDS:
  • Recent Posts