GE Software Center: The Big Data Land of Oz

Bill Schmarzo By Bill Schmarzo February 24, 2014

As many of you know, I get to spend all of my time working with customers on identifying where and how to start their software analytics and big data journeys.  And I’ve been fortunate enough to work with some outstanding companies across a wide range of industries.  I have had the chance to collaborate with credit card companies, public schools, wind energy producers, mobile phone companies, credit unions, entertainment and hospitality companies, product manufacturers, retailers and many others to uncover best practices about how organizations can leverage big data to transform their business models – to optimize their key business processes and uncover new monetization opportunities.  The result has been the development of the Big Data Business Model Maturity Index (BDBM) to help organizations (see Figure 1):

  • Understand where they stand today with respect to the integration of big data into their business models, and
  • Provide a roadmap or guidelines for identifying what to do next to help the organization to progress up the maturity index and realize the 4 M’s of Big Data: “Make Me More Money.”
Figure 1: Big Data Business Model (BDBM) Maturity Index

Figure 1: Big Data Business Model (BDBM) Maturity Index

This past week I had the chance to meet with my good friend Brad Surak, who is the General Manager for Industrial Internet Programs at the GE Software Center.  Brad is chartered with leading GE’s efforts to advance the Industrial Internet by creating the next-generation software-based applications and services that will drive operational efficiencies across aviation, power generation, oil & gas, healthcare and transportation.  Brad and the outstanding team at GE Software are probably the best example of an organization that has “bet the house” on figuring out how to integrate big data into each aspect of their business operations.  GE isn’t developing a big data strategy, but instead is developing a business strategy that incorporates big data.  And that’s where things really get interesting!

How GE Is Igniting the Next Industrial Revolution: The Predix™ Platform

GE sees a huge business opportunity to leverage the Industrial Internet and the resulting machine-to-machine (M2M) and human-to-machine communications. This is something that I discussed in a previous blog “Understanding Skynet: Internet of Things vs. Industrial Internet.” Advances in networked sensors, computing, software development, analytics, and security provide the foundation to create the next generation of “intelligent” machines that are self-aware, that can connect and interact with other machines and their human operators, that can be provisioned, managed, upgraded and decommissioned remotely, that can function safely and securely, and that can dramatically improve operational effectiveness.  GE is poised to translate these “intelligent” machine and these next generation operational services into savings that range from $320 billion to $640 billion annually[1]

The GE “Big, Hairy, Audacious Goal” (BHAG) has got many supporters as exemplified in Figure 2 from Bank of America Merrill Lynch.  This chart highlights that GE brings a unique combination the Industrial Internet M2M data management and analytics capabilities combined with deep domain expertise to lead the development of the new generation of intelligent machines and operational applications.

Figure 2: Bank of America Positioning of GE as Leader in Industrial Internet Market

Figure 2: Bank of America Positioning of GE as Leader in Industrial Internet Market

The EMC Big Data Business Model Maturity Index

GE provides a live model for how an organization can leverage the BDBM Index to identify your key business processes (in light of your organization’s overall business strategy), leverage big data to uncover new customer, product and actionable operational insights that can be used to optimize key business processes and uncover new monetization opportunities.  But GE has an even grander vision that they are working to put in place – to provide an industrial-strength analytics platform and a supporting ecosystem that allows GE to transform their business (from selling products and services to providing business critical solutions that generate positive outcomes) while allowing customers and partners to make money by extending the value of this platform.  Damn, that’s exciting!

Let’s look at what GE is doing at each phase of the BDBM Index, and what sort of business value they are deriving out of each of these phases.

  • Phase 1 is the Business Monitoring phase, where organizations are leveraging data warehousing (DW) and business intelligence (BI) investments to identify and prioritize their key business processes. The BI/DW foundation is a good starting point for an organization’s big data journey, as organizations have already mapped out their key business processes, identified the key metrics and key performance indicators against which they will measure success, identified the dimensions and hierarchies against which they must monitor performance, captured, aggregated and cleansed the necessary transactional data, and have created the reports, alerts and dashboards that are used by the key business stakeholders to report on business performance.
For GE, their two targeted key business processes are Asset Optimization and Operations Optimization.  For each of these business processes, GE has identified the key personas or business stakeholders that need this information.  They have identified the types of decisions these stakeholders need to make and the questions that the users need to answer to support the decision making.  They have identified the key measures and key performance indicators against which success will be measured.  Finally, they are identifying the operational data and how that operational data will be monitored via reports and dashboards.
  • The goal of Phase 2 – the Business Insights phase – is to leverage the four big data business value drivers to uncover new customer, product and operational insights.  These four big data value drivers are: 1) access to ALL the detailed transactional data, 2) access to new internal and external unstructured data sources, 3) real-time (right-time) data access, analysis and action, and 4) integrated predictive analytics to look across the wealth of data, metrics and dimensions to find areas of unusual operational performance (see Figure 2).
Figure 3: Four Big Data Value Drivers

Figure 3: Four Big Data Value Drivers

GE is doing the following to move to exploit the Business Insights stage:

      • Storing and analyzing ALL the operational or machine-to-machine (M2M) data – at the lowest levels of granularity – that is coming off of their bevy of sensors.  This is the sensor data that is being kicked off by the tens of thousands of sensors in their jet engines, wind turbines, locomotives, energy plants, medical devices, etc.
      • Integrating new unstructured data sources including internal data sources (e.g., text comments and photos captured via tablets and smart phones from technicians, machine operators, mechanics, train engineers, service crews, pilots, nurses, etc.) as well as external data sources (e.g., weather, temperatures, humidity, elevation, traffic, track conditions, etc.).
      • Applying real-time access, analysis and action to key data events within their targeted business processes.  There are a select number of data events occurring within the devices and machines on the Industrial Internet for which real-time or right-time access and analysis is critical.  For example, products, components or sub-assemblies of jet engines, energy plans, wind turbines or locomotives that are operating outside of normal performance boundaries for which an immediate corrective action might be required.  But not all data events require real-time analysis, so GE is going through the process of identifying those specific data events for which real-time analysis is operationally critical.
      • Integrating predictive analytics to look for unusual product or device patterns, trends, propensities and operational behaviors.  The data volumes are so massive and the number of metrics and dimensions so large that no human can possible monitor all the important combinations of metrics and dimensions to identify areas of “unusualness.”  This is a perfect example of leveraging predictive analytics and even basic six sigma methodologies (like control charts) to flag these operational performance areas of “unusualness” that requires human intervention to asses the scope, severity, causes and repeatability of the problem.  By the way, GE’s expertise with Six Sigma[2] (using techniques like control charts and Pareto charts to monitor and optimize manufacturing processes) is certainly making the integration of big data into their business models much easier.
  • Stage 3 is the Business Optimization stage where organizations build upon the insights uncovered in Phase 2 to identify, quantify and analytically model repeatable event occurrences.  Organizations leverage the S.A.M. Principle to ensure that they are focused on modeling data events that are strategic, actionable and material to the business.  The end goal is to move towards prescriptive analytics where the analytics are delivering recommendations to the front-line decision makers.

GE is aggressively applying the SAM Principle to identify and model the most critical product usage events.  They have implemented a data analytics and science discipline (see Figure 4) that aggressively looks to quantify these insights with the goal of ultimately delivering recommendations to the key front-line decision makers such as the train engineers, mechanics, product developers, pilots, maintenance crews, nurses and the like.

Figure 4: Data Science Discipline

Figure 4: Data Science Discipline

  • In the Data Monetization phase, organizations seek to monetize the customer, product and operational insights that are the byproduct of the BDBM phases 1 through 3.  We see organizations monetizing insights in one of three ways:

1) Packaging, selling and/or trading customer, product or operational insights

2) Embedding analytic insights into the physical product to create “intelligent” (self-monitoring, self-healing, self-optimization) products (intelligent thermostats, refrigerators, cars, jet engines, wind turbines)

3) Creating an entirely new customer experience by leveraging unique insights about your customers and partners behaviors, tendencies, propensities and trends to make product purchase and usage recommendations that provide a more rewarding, more compelling, more sticky customer experience

GE is leveraging all three of these monetization opportunities, by creating a data platform (built on open data principles) to sharing operational data and insights with key partners and customers, integrating analytics (integrating machine learning with physics-based models) to create intelligent products and ultimately leverage best-in-class user experience design skills and techniques to deliver prescriptive insights to the front-line decision makers as well as internal business stakeholders.

  • Finally, the Business Metamorphosis phase.  This is the Holy Grail – leveraging your data, analytics and insights to create an ecosystem or platform upon which others can provide value and “make money.”  This empowers the legions of independent developers to create value by leveraging an easy-to-use, easy-to-scale, easy-to-deploy/provision environment where third party developers can focus their efforts on creating value added products and services, and less time trying to stand up the environment.

To realize this business value bonanza and squarely take the leadership role in the Business Metamorphosis phase of the BDBM, GE has developed the Predix platform. Predix will help GE and its partners bring new industrial solutions to market more quickly, and in doing so drive important advances in functionality as well as efficiency and cost-savings. GE is also building an ecosystem of partners to extend the value of Predix. These partners are engaging with GE in establishing standards and driving the knowledge and products needed to make Predix an industry standard.

Creating An Environment For Innovation And Ideation

The GE Software has taken this entire approach one step further, by creating an innovation center where GE, customers and partners can convene to brainstorm how and where big data can power their business and operations.  This helps translate the potential of big data into something meaningful to all parties.  And as big advocates of the Vision Workshop process, we think that this may be the most powerful innovation – creating a failsafe, creative ideation environment where the realm of what’s possible can be freely explored.  Heck, it even has a Star Trek-like Holodeck (see Figure 5) that they use to fuel innovative thinking with their customers! Most excellent!!

Figure 5: Nina Grooms Lee (left), Me (center) and Brad Surak (right) in the Holodeck

Figure 5: Nina Grooms Lee (left), Me (center) and Brad Surak (right) in the Holodeck

Truly a most excellent adventure, but more importantly, a living example for other companies to follow – companies who believe that there are opportunities to leverage big data to optimize key business processes and uncover new monetization opportunities.  It’s simple; just follow the Big Data Business Model Maturity Index, or maybe it’s better to label it the Yellow Brick Road.


[1] Source: “Industrial Internet, a Key Growth Initiative for GE.” Bank of America Merrill Lynch Global Research. Sept., 2013

[2] Six Sigma is a set of management techniques and statistical methods that seek to improve the quality of process outputs by identifying and removing the causes of defects (errors) and minimizing variability in manufacturing and business processes. Each Six Sigma project carried out within an organization follows a defined sequence of steps and has quantified value targets, for example: reduce process cycle time, reduce pollution, reduce costs, increase customer satisfaction, and increase profits.

The term Six Sigma is a term associated with statistical modeling of manufacturing processes. The maturity of a manufacturing process can be described by a sigma rating indicating the percentage of defect-free products it creates. A six sigma process is one in which 99.99966% of the products manufactured are statistically expected to be free of defects.

Bill Schmarzo

About Bill Schmarzo

Read More

Share this Story
Join the Conversation

Our Team becomes stronger with every person who adds to the conversation. So please join the conversation. Comment on our posts and share!

Leave a Reply

Your email address will not be published. Required fields are marked *