KDnuggets Home » News » 2014 » Jul » Opinions, Interviews, Reports » Manufacturing Analytics Summit 2014 Chicago: Day 1 Highlights ( 16:n16 )

Manufacturing Analytics Summit 2014 Chicago: Day 1 Highlights

Highlights from the presentations by Analytics leaders from McCormick, HP, Patheon and Boeing on day 1 of Manufacturing Analytics Summit 2014 in Chicago.

To keep up with the pace of development in modern business, it is essential to invest in more sophisticated and innovative manufacturing processes. Investment in analytics offers this opportunity as it allows organizations to streamline their manufacturing process, improve efficiency and reactivity to customer demand by utilizing real time data. For any company manufacturing its own products, this is an essential area to gain an advantage over competitors.

The Manufacturing Analytics 2014Manufacturing Analytics Summit (May 21 & 22, 2014) was organized by the Innovation Enterprise in Chicago, IL to bring together analytics leaders from manufacturing sector to share their success stories and key learning. The summit addressed the issues mentioned above and brought together business leaders who are implementing and experimenting with new analytical methods for manufacturing - driving remarkable improvements in their organizations. Executives at the forefront of analytics shared their innovative approaches, providing insight into how they have gained valuable information from raw data.

We provide here a summary of selected talks along with the key takeaways.

Here are highlights from Day 1 (Wednesday, May 21, 2014):

Grace WooGrace Woo, Director of Procurement, McCormick spoke on one of the most common challenge visionary employees face, in her talk "Supply Chain Analytics: How To Get The Buy-in That You Need To Get Started". A lot of data science practioners have a vision of how analytics can bring value, but find it difficult to convince thier boss to invest so that they can get started. Many of them are frustrated with end to end data quality, but find it difficult to convince their peers to collaborate in improving data quality. For quite some time, Analytics has persistently appeared in strategic priorities, yet it is important to understand that most companies are just at the beginning of this long journey and share the struggle.

Analytics evangelists should start with knowing themselves, their boss and their organization. In context of the "crossing the chasm" curve, most of such analytics evangelists will fall in the "Visionaries" category, whereas their boss will fall in the "Pragmatists" category; both categories separated by the "chasm" valley. In Supply Chain Analytics, one is more likely to be drowned by inflated expectation than data. In conclusion, she provided the following advice for designing the first steps:
  1. Find and organize like minded peers to locate potential pilot projects
  2. Find internal data talent and empower them with tools that will solve their most immediate problems
  3. Select projects that your sponsor will likely say yes

Milind ShinganeMilind Shingane, Director, Global Supply Chain Systems, Hewlett Packard shared his experience and learning at HP in his talk "Transforming the Largest IT Supply Chain in the World". Describing the magnitude of data processing, he mentioned that every year HP's global supply chain systems process 36M purchase order lines, 87M sales order lines, 17M deliveries, and 19M invoices. Diverse nature of products means HP has to employ multiple strategies in all aspects of its supply chain from manufacturing to distribution. Accelerated technological innovations are continuously reducing time to market and fierce competition demands that products be delivered with high quality and exceptional customer service. The challenge is to transform this complex supply chain while maintaining business continuity and laying the foundation for longer term profitable growth.

He shared HP's vision to build the industry's best supply chain through predictive analytics, in-transit inventory allocation models, collaborative supply chain networks and enhancing customer experience. In the march towards its vision, HP has developed a cloud-based next generation supply chain architecture connecting customers, channel partners, sales staff, HP social networks, suppliers, manufacturing partners, and transportation carriers. HP's Buy/Sell solution gives it significant advantages (over the legacy applications) such as real-time supply chain analytics & decision support and metwork with over 36,000+ tranding partners. The "Logistic Cloud" layer in the systems architecture provides various important capabilities (such as collaboration, visibility, control) to Analytics applications.

Barry GujralBarry Gujral, Director of Operational Excellence, Patheon gave an insightful talk on "Pharmaceutical Analytics - An Innovative Tool to Maximize Returns". Pharmaceutical manufacturers are improving their marketing and product distribution strategies using analytics. Analytics are used to build mathematical models, computer simulations and decision strategies to extract value out of large pharmaceutical data. Thus analytics helps the industry to streamline operations to minimize risks and to get maximum returns. Predictive analytics helps to find the root cause of the failure and to bring process optimization. In pharmaceuticals, data is being generated from different sources such as R&D, Physicians, Retailers, and Patients. The pharmaceutical companies can better identify and develop new potential drug entities using innovative analytics tools such as Quality by Design (QbD). These drugs are potentially getting approval more quickly than from those developed with traditional methods. Thus analytics adds further value to the developers and consumers.

In order to maximize the returns, processes in pharma industry should focus on consistency, quality, efficacy, safety and reducing waste. The tools used to maximize returns include quality analytics, lean six sigma, quality by design, risk assessment and simulations. Next, he defined Capability Indices and the relevance of that metric. A process capability index uses both the process variability and the process specifications to determine whether the process is capable. The capability index is useful for measuring continual improvement using trends over time, or for prioritizing the order in which process will be improved, and for determining whether a process is capable of meeting customer requirements. Quality cannot be tested into products, it should be built in or it should be by design.

Paul OrtmanPaul Ortman, Manager, Manufacturing and Quality Advanced Analytics, Boeing delivered a great talk on "Insight Incorporation: Transformative Value Through Analytics". Manufacturing is generally under constant pressure to enhance quality and productivity to improve margins and fund investments in other parts of the business. Data analysis in conjunction with methodologies such as Six Sigma and Lean Manufacturing have been leveraged for decades and have achieved step-function improvements. Further strategies involving work placement have also been employed to increase competitiveness through cost reduction.

More recently, data, cutting-edge modeling tools, self-service architecture, and data-savvy analysts and managers can apply advanced analytics techniques for insights beyond isolated factory processes. However, investments in IT and talent are often required to bridge data sets to make analysis possible. In an environment that demands constant improvements in quality and productivity, analytics solutions must be quickly available and easily incorporated into business decisions to transform the business. In other words: gone are the days of proof-of-concept analytics; manufacturing leaders must make decisions based on the insights to drive value.

We must be deliberate in our quest to ensure insights are incorporated into the business. This is more likely by having a systematic methodology focused on the business - process, problems, people, and systems - utilizing a model that converts analytics-derived insights that are reliable, explainable, deployable and scalable. In conclusion, he mentioned that the path from discovery to implementation should follow the R.E.D.S. approach (which stands for Reliable, Explainable, Deployable, and Scalable). Analytics should deliver trust and data-driven understanding, not black-box magic.

Highlights from presentations on day 2.