December 8, 2020 in Intel wins Edelman
O.R. Inside Intel: Project Produces $25 billion in Benefits, Earns Edelman Award
Tech giant receives coveted award from INFORMS for integrating product architecture design with supply chain planning.
SHARE: PRINT ARTICLE:
https://doi.org/10.1287/orms.2020.06.06
Beginning in 1991, “Intel inside” [1] became a wildly successful branding slogan and logo for the famed U.S. tech corporation best known for its semiconductors driving countless computers around the world. Now, after receiving the 2020 Franz Edelman Award for Achievement in Advanced Analytics, Operations Research and Management Science, it’s an appropriate time to celebrate “O.R. inside Intel.”
A panel of Edelman Award Committee members and judges named Intel Corporation the winner of the coveted honor for its use of operations research (O.R.) to maximize revenue and minimize cost thanks to a multiphase project that resulted in a whopping total benefit of $25.4 billion since 2009.
How do you achieve such impressive results? In the case of Intel’s Edelman-winning O.R. team, you start with a massive, complex problem, which the team outlined as follows:
Due to its scale, the complexity of its products and manufacturing processes, and the capital-intensive nature of the semiconductor business, efficient product architecture design integrated with supply chain planning is critical to Intel’s success. In response to an exponential increase in complexities, Intel has used advanced analytics to develop an innovative capability that spans product architecture design through supply chain planning with the dual goals of maximizing revenue and minimizing costs. (For technical details of the product feature design problem and the team’s approach, see the accompanying sidebar story.)
“The system has been vital to improving the process of corporate decision-making, and the results have been in the billions since the system began,” says George Davis, CFO of Intel. “It is an innovative, cost-effective and sustainable strategy that uses operations research and data analytics to create efficiency across the corporation.”
Karl Kempf, senior fellow and director of decision engineering at Intel, as well as a longtime member and Fellow of INFORMS, led Intel’s Edelman presentation. In addition to the monetary benefits, Kempf noted that Intel has realized a number of organizational benefits. “One is the establishment of personnel on the front lines of product design and supply chain planning to make faster, better, more integrated decisions,” he said. “Another gives senior management new insights and abilities to influence decisions impacting go-to-market strategies that they have never had before. Finally, our success has provided an existence proof of the power of applying advanced analytics to Intel business problems for other teams inside Intel, as well as teams across the advanced analytics community.” (Shortly after the Edelman Award presentation, the co-authors interviewed Kempf via Zoom in preparation for this article. To listen to the interview, click here).
The Edelman Team
The Edelman-winning team was comprised of members from two Intel groups, both based in Chandler, Ariz. Kempf led the Decision Engineering, Data Platforms Group that included Ryan Lovrien, Nicholas Mason, Evan Rash and Harry Travis. Shamin Shirodkar, a former master’s student of Kempf’s at Arizona State University, led the Supply Chain Decision Solutions, Global Supply Chain Group that included John Heiney, Irfan Ovacik, Nandini Sarkar, Zhenying Zhao and Kalani Ching. Their presentation title says it all: “Intel realizes $25 billion by applying advanced analytics from product architecture design through supply chain planning.”
When asked what stood out about Intel from her perspective as chairperson of the Edelman Award Committee, as well as a competition judge, Pooja Dewan said, “All finalists’ submissions were impactful and creative. As always it was very hard to decide the winner. Intel’s project stood out because of the wide impact the project had on Intel’s entire organization and the period of time over which the impact had changed their design and manufacturing process. Their leadership desire to endorse and participate in the competition despite the adversity we all are facing is also commendable.”
The product feature design problem
A large part of the product feature design problem is defining a finite number of stock keeping units (SKUs) with associated values for key features to best cover market needs and balance against manufacturability and costs. Given the broad range of Intel’s application areas and customers, the combinatorics faced by product feature designers is daunting and involves search spaces of plausible plans numbering from 1010 to 1050. The output of the product feature design process feeds the supply chain planning and manufacturing execution process, which routes products through 6+ manufacturing stages, at 200+ internal and outsourced factories across 13 countries.
Intel’s manufacturing flow begins with wafer fabrication. Each wafer is a thin circular disk of silicon 200-300mm in diameter and containing 15 to 15,000 complete integrated circuits, or die, depending on die size. Each wafer is separated into a number of die that are placed into a protective container with external connectors, which will enable attachment to printed circuit boards. At times multiple die from different wafers are used for multi-chip products. Intel is especially focused on the efficiency of using every die on every wafer to supply a saleable product across the architectures of the entire Intel product portfolio. The assembled items are then tested to measure actual performance. Tests help separate items into categories defined by performance ranges for a combination of the key features. Lastly, the items go through a Finish process that results in individual SKUs. Each wafer could have hundreds of thousands of possible manufacturing routes before it ends up as one of the SKUs shipped to customers.
Faster, better and more integrated decision-making is increasingly important to Intel’s decision process, as its products have become more complex. Each year, Intel introduces several new products and plans for roughly 400 unique wafer types totaling 2.5 million of wafer volume, which results in 4,000 unique SKUs accounting for 600 million in SKU volume. Products can flow through six or more manufacturing stages at more than 200 internal and outsourced factories across 13 countries.
Source: Intel
Gala Goes Virtual
Originally scheduled as the Edelman Gala during the 2020 INFORMS Business Analytics Conference in the spring, the finalists’ presentations were showcased – and the judges’ decision announced – at a virtual Edelman Award presentation on Sept. 29 due to the coronavirus. The Edelman Award, widely considered the “Super Bowl of O.R. Applications,” is a yearlong competition that begins with a call for nominations, followed by a vetting and verification process of the nominated work. Once the nominations are culled to a handful of finalists, the competition culminates with presentations – advised by INFORMS-appointed team coaches – before a panel of award committee members and judges. After listening to the presentations and questioning the presenters, the judges confer behind closed doors to select a winner.
This year the virtual Edelman Award Ceremony was truly like no other. The traditional hotel/convention center ballroom packed with INFORMS Business Analytics Conference attendees gave way to an online stream of names and faces. Instead of an Oscar-like moment with the Edelman Award Committee chairperson handing an envelope containing the name of the winner to the INFORMS president for the big reveal, Committee chair Dewan live-texted INFORMS President Pinar Keskinocak a logo of the winning company. Pinar held up her cell phone for all those online to see. Instead of a standing ovation, the virtual announcement triggered a silent outpouring of two-word messages in the chat room: “Congratulations Intel!”
Even in the coronavirus era, the show must go on, and thanks to Mary Leszczynski, senior manager and creative director, and other INFORMS staff members who helped put the virtual event together – and thanks to Grace Lin and Peter Bell who served as co-hosts – the surreal soiree was an inspiring success. It did what it was supposed to do in half the time of the traditional ballroom gala: showcase the finalists (see accompanying sidebar story) – Carnival Corporation, Deutsche Bahn, IBM, Intel and Walmart – with prerecorded presentations and videos of the most outstanding achievement in applied O.R. and analytics … and crown a winner.
“I would like to thank the competitors, judges, coaches and most of all INFORMS staff, who helped us orchestrate the award in a virtual environment,” Dewan said. “It was a challenge and new way of hosting the competition that required all of us to adjust the way we operate in other years. All the folks rose up to the challenge, so we can take the time to recognize the teams creating excellence in applying O.R. and analytics this year despite the adversity.”
Tremendous Honor
Accepting the Edelman Award on behalf of Intel, Kempf said, “What a tremendous honor for Intel Corporation. We have traversed a long and winding path to get here. Craig Barrett started the advanced analytics effort at Intel in 1987 when he was in charge of manufacturing. He later became CEO, and I can assure you that being the favored project of the CEO was very beneficial. Even more beneficial has been delivering the benefits of applying advanced analytics to a variety of important problems across the corporation.
“After 20 years we thought we had a rather impressive portfolio of success stories, Kempf continued. “That was ratified in 2009 when we were awarded the INFORMS Prize. That was about the time that we started on the project that is the focus of our 2020 Edelman Award work, our integrated Product Architect Design and Supply Chain Planning system. We started with an attack on what we saw as one of the most important mathematical problems for realizing the system. Success on that problem not only propelled us forward but was awarded the Wagner Prize in 2011.”
Kempf closed by thanking the Edelman judges for “giving us the opportunity to showcase our work,” as well as the team’s coaches, Jeff Alden and Aaron Burciaga, CAP, whose “help contributed tremendously to our success.” He also congratulated the other finalists, adding that, “The very best feature of the Edelman process is there are no losers, everybody wins. The committee picks first place, but it is very clearly first among equals.”
Reference
- Wikipedia, “By launching its Intel Inside marketing campaign in 1991, Intel was able to associate brand loyalty with consumer selection, so that by the end of the 1990s, its line of Pentium processors had become a household name.”
2020 Edelman Award Finalists
In addition to Intel, the other finalists and their presentations for the 2020 Edelman Award include:
Carnival Corporation & plc (“Joint revenue and inventory optimization for multiple heterogeneous brands in the world’s largest cruise company”): Carnival & plc built Yield Optimization and Demand Analytics (YODA), a cutting-edge program using advanced analytics and algorithms to provide dynamic price recommendations and inventory management. The program is a global collaboration involving leaders from six brands and teams across three continents who manage prices in six currencies. Since the system deployed in late 2017, it has been used to set prices on thousands of cruise voyages. The system was built in collaboration with Revenue Analytics.
Deutsche Bahn (“On the right track: Deutsche Bahn schedules train rotations using hypergraph optimization”): The largest European railway company is moving more efficiently thanks in part to a new decision support system. The system optimizes train rotations in cargo, regional and long-distance passenger transport divisions. The changes have resulted in direct savings in addition to better planning and productivity. The productivity of the locomotives and railcars has increased up to 10%, while the time savings were up to 80%.
IBM (“Predictive analytics for server incident reduction in the wild”): IBM Services (Global Technology Services Unit) has developed a cutting-edge system that utilizes machine learning and advanced data analytics to identify devices, such as servers, that are at risk for an outage. The system determines contributing risk factors, e.g., outdated hardware, and prompts a fix before the problem occurs. Since 2013, executing recommended actions has reduced outages by 23%. This translates into savings of more than $1 billion every year for IBM’s clients. The IBM team is also using this system to drastically reduce the number of problematic systems for clients by as much as 85%.
Walmart (“A multiobjective price optimization framework in stores using reinforcement learning”): Walmart’s model of offering consistently low prices depends on constantly innovating to find new ways to reduce costs. The company developed a new system using machine learning to help perfectly time markdowns to optimize sales and clear excess inventory in its stores. The technique relies on operations research and provides feedback to help avoid excess inventory ordering in the future. The technology has already helped the company save millions of dollars.
For more detailed information about this year’s finalist projects, check out the Resoundingly Human podcast where host Ashley Kilgore talks to each 2020 finalist team (https://pubsonline.informs.org/magazine/orms-today/podcasts).
Peter Horner is the editor of Analytics magazine.
([email protected])
