9 years ago
Predictive Analytics Can Help with the Challenges Facing Manufacturing in the 21st Century

Historically, data and analytics have been key to the success of manufacturing. The biggest contributor to the success of the 100+ year old assembly line technology was the development of interchangeable parts. Clearly, data was central to this concept. It was really the emphasis on making interchangeable parts that led to the invention of statistical quality control (SQC) and statistical process control. The recognition of variability as a manufacturing issue and the recognition that this variation is different from variation seen in natural or social phenomenon was essential to the development of methods to track, diagnose and control variability. Methods such as SQC were the first fully fledged analytics efforts, that came out of any industry and manufacturing was the one which led this.

However, in a study that was reported in the well-known “Competing on Analytics” book by Davenport and Harris (2006), manufacturing companies represented less than 10% of those considered as Analytics Competitors. Somewhere along the road, manufacturing businesses dropped the leadership on analytics, which was taken over by financial and technology companies.

Has this situation changed in the last 10 years? What are the areas of growth and opportunity for manufacturers? Can manufacturing resurrect itself as a leader in the development of new technologies, during these challenging times, when it is fighting for its own survival in many developed countries, including the US?

Manufacturers must now focus on using data beyond the conventional application areas such as efficiency and quality. Traditional lean production and cost cutting are no longer enough to remain competitive. Manufacturers generate and store data from many sources across the supply chain. The goal today is to integrate and gain insights from data across their complex global and fragmented supply chains.

But the big story that will unfold in the near future is the data coming from systems that monitor the performance of products after they’ve been sold. Consider the following:

• A fully instrumented car can generate 25 GB/hr
• A Self driving car can generate nearly 10 times as much, about 250 GB/hr
• A fully instrumented jet engine easily generates 50 TB/hr

This is the emerging area of application that will dwarf all varieties of big data that other industries are rushing to squeeze value from. These insights can lead to improvements in design and production, product quality, forecasting, more targeted products and distribution, reduce warranty and recall, and identify hidden bottlenecks in the production process.

A basic use case for the data from these connected devices is Preventive Maintenance. Today, the world wide annual maintenance spend on production related assets is about $0.5 Trillion. By systematically using data from connected products and using advanced methods such as predictive analytics, the value that the industry can generate is enormous. For example, a

• 1% fuel savings, for an airline company can result in savings of nearly $30B
• 1% system efficiency (in rail transport) can generate $27B
• 1% reduction in capex on assets (in Oil and Gas) can result in $90B

To quote W. Edwards Deming, a pioneer in applying statistical techniques and predictive analytics to manufacturing, “The big problems are where people don’t realize they have one in the first place.”

Author Bio:

Bala Deshpande, Ph.D. (Carnegie Mellon), MBA (University of Michigan), is the founding partner at SimaFore, a boutique consulting company that focuses on providing custom analytics solutions for manufacturing, marketing and non-profits.

Dr. Deshpande’s has two decades of experience in using analytical techniques. His first exposure to predictive models and analytics was in academia, in the field of biomechanics – for identifying correlations and building regression models to predict muscle forces based on electrical activity in muscles. He began his career in the industry as an engineering consultant at EASi Engineering, following which he spent several years analyzing data from automobile crash tests and helping to build safer cars at Ford Motor Company. He has been actively involved in promoting information theory based analytical techniques for a range of applications from performance measurement in organizations to healthcare. He is an active blogger and is currently wrapping up a book on Predictive Analytics and Data Mining to be published by Morgan Kaufman.

He is currently the Co-Chair for Predictive Analytics World – Manufacturing, Chicago.