INFORMS honors: How the FCC won the Edelman Award

Jean Kiddoo (holding award) and the rest of the FCC’s Edelman Award team are all smiles after winning the “Super Bowl of O.R.” Former INFORMS presidents Michael Trick (sixth from left) and Karla Hoffman (third from right) were key members of the team.

Jean Kiddoo (holding award) and the rest of the FCC’s Edelman Award team are all smiles after winning the “Super Bowl of O.R.” Former INFORMS presidents Michael Trick (sixth from left) and Karla Hoffman (third from right) were key members of the team.

The U.S. Federal Communications Commission turns to operations research and incentive auction to relieve the “spectrum crunch.”

By Peter Horner

Spectrum, the range of radio frequencies used for communications in a particular service area, is finite. Local broadcast television stations control a lot of spectrum through paid licenses. Meanwhile, the demand for spectrum in the United States and Canada has exploded over the past decade due to the proliferation of smart phones, tablets, computers and other wireless devices.

So, given a finite capacity, how do you accommodate the rising demand for wireless spectrum for communication devices while respecting the legal rights of local television stations across much of North America? That was the problem facing the U.S. Federal Communications Commission, the agency that authorizes and manages the use of spectrum throughout the United States. The solution, several years in the making, not only satisfied the various stakeholders and raised billions of dollars, it also resulted in the FCC winning the 2018 Franz Edelman Award for Achievement in Advanced Analytics, Operations Research and Management Science.

How did the FCC do it? With the vital help of operations research (O.R.), the FCC conducted the world’s first two-sided “incentive auction” to meet the demand for wireless services by reclaiming valuable low-band electromagnetic spectrum from TV broadcasters. By reselling spectrum to wireless providers, the auction repurposed 84 MHz of TV spectrum for mobile broadband, next-generation “5G” and other wireless uses, raised $20 billion in revenue. In addition, thanks to O.R. and optimization, many TV stations remained on their original channels, saving an estimated $200 million in relocation costs.

The prestigious Edelman was presented by INFORMS at the Edelman Gala held in April in Baltimore in conjunction with the 2018 INFORMS Conference on Business Analytics and Operations Research.

“To be presented with the Franz Edelman Award, which recognizes the greatest achievements in operations research and analytics, is an incredible testament to the ground-breaking work of our incentive auction team,” says FCC Chairman Ajit Pai. “The auction would not have been possible without the use of operations research tools to solve complex design and implementation challenges. We are deeply honored that our effort has been recognized with this prestigious award.”

Two-Sided Incentive Auction

The auction would not have been possible without the use of operations research tools to solve complex design and implementation challenges. – FCC Chairman, Ajit Pai

The auction would not have been possible without the use of operations research tools to solve complex design and implementation challenges. – FCC Chairman, Ajit Pai

In 2012, based in part on the FCC’s preliminary O.R. studies demonstrating the feasibility of the concept, Congress passed a law directing the FCC to relieve the “spectrum crunch” by conducting the world’s first two-sided “incentive auction.” The objective: use market incentives to repurpose some over-the-air broadcast television spectrum to wireless communications services.

Politics and data-driven decision-making in government do not always mesh, so how did the FCC get past that hurdle?

“It took many years of hard work,” says Jean Kiddoo, chair of the Incentive Auction Task Force at the Federal Communications Commission, who accepted the Edelman on behalf of the FCC. “We used operations research to test our theories and prove that it could be done to convince legislators, regulators and the people we needed for support. They thought this was a concept that would work. The broadcast and wireless industries, on the other hand, were very skeptical. They didn’t think it could be done, and they certainly didn’t think the FCC could accomplish it.”

As FCC officials explained in its award-winning presentation, the auction would offer television stations “incentive” payments to voluntarily relinquish their spectrum (i.e., TV channel assignments) and reassign the remaining television stations into the smaller segment of the spectrum allotted to television broadcasting after the auction. The vacated spectrum would then be converted from television broadcasting into expanded capacity for wireless communications services.

Conducting the auction required far more than the creation of a standard exchange for a largely homogeneous commodity. The television channels to be purchased in the auction differ in terms of the population they covered based on their location, frequency and other factors. Adding significantly to the auction design challenges, the wireless spectrum “products” to be sold in the auction would not be usable by wireless serve providers unless reconfigured into contiguous blocks whose dimensions differ in both frequency and geographic scope from the broadcast licenses. This required the FCC to use innovative O.R. techniques and analytics to create a successful auction.

In addition, the FCC was required to determine how much spectrum was available to be repurposed, how to determine channel availability for nearly 3,000 TV stations across the United States and Canada on as few as 28 channels, and what new channel assignments would be feasible substitutes for television broadcasters who would remain on the air after the auction. This required analyzing massive quantities of data, creating optimization and feasibility models, and determining appropriate policy objectives to serve American consumers, the wireless industry and television broadcasters.

To solve these challenges the FCC created two tools: 1) a distributed optimization solver to address all optimization problems, using Gurobi, combined with a set of custom heuristics and decomposition approaches to determine optimal or near- optimal solutions, for each problem; and 2) a highly customized portfolio of satisfiability solvers that determine usually within a fraction of a second whether a given station repacking was feasible or infeasible.

So how critical was operations research to the project?

“It was essential, absolutely essential,” says Kiddoo. “During the auction we needed feasibility checkers to be able to determine what stations we needed to buy and what stations we didn’t need to buy, which stations could be repacked into new channels, and which ones couldn’t. It had to be done on a real-time basis between every round of the auction, and we needed it to happen really fast. That was the challenge. We ended up being able to run six rounds of the auction per day rather than one round of the auction every three weeks. The auction was still very long – it went on for a year – but it would have gone on for the next 10 years had we not used O.R.”

The FCC’s unique and unprecedented auction generated nearly $20 billion in revenue from wireless providers, of which over $10 billion was paid to television broadcasters, $1.85 billion went to support television stations required to switch to a new channel, and $7.3 billion was sent to the U.S. Treasury to reduce the federal deficit. The auction repurposed 14 channels (84 MHz) of valuable low-band spectrum, 70 MHz for licensed use to support technologies such as next-generation 5G mobile broadband and the Internet of Things, and 14 MHz for unlicensed use to support technologies such as Wi-Fi and wireless microphones.

The FCC’s use of optimization enabled the market to determine the maximum amount of spectrum to repurpose while giving all stations that remain on air a channel equivalent to their pre-auction channel. Importantly, these tools ultimately enabled the FCC to assign most of the remaining TV stations to their original channel, thereby significantly reducing transition costs and TV viewer inconvenience.

“I cannot be more proud of this team, and the fact that the government let us do it is just amazing to me. They believed we could do it when it was iffy, but we worked at it because they believed in us and we believed in them,” says Karla Hoffman of George Mason University and a past president of INFORMS, as was fellow team member Michael Trick of Carnegie Mellon University.

Asked to sum up the monumental project and its financial impact, Hoffman says, “Forget the $20 billion dollars, forget the $7 billion for deficit reduction. This relates to the future ability of the United States to have the wireless spectrum it needs for 5G and the Internet of Things. This is the wireless economy. It’s the app economy, and if the FCC didn’t do this, we couldn’t move forward.”

When asked what he thought most impressed the Edelman judges, Irv Lustig of Princeton Consultants, a longtime member of INFORMS and a coach of the FCC Edelman-winning team along with Luz Adriana Caudillo Fuentes of Universidad Anáhuac, backs Hoffman’s assessment. “It was not just about the benefits,” Lustig says. “I think the coolest thing was that O.R. helped create the policy and the congressional bill that authorized the auction. They did amazing work on the O.R. side, but they also did a great job of explaining a very complex problem and convincing the industry – both television broadcasters as well as wireless companies – that this new auction would work, and that the operations research behind it would work as well.”

Going Forward

According to the team’s Edelman presentation, the FCC has continued to embrace operations research solutions, including using optimization to develop the post-auction transition scheduled for broadcasters. This schedule helps broadcasters coordinate limited transition resources to enable all broadcasters to transition by the final deadline so that wireless providers can promptly use their newly acquitted spectrum to prove expanded services to American consumers. The FCC also sees operations research as a pillar of its new Office of Economics and Analytics that will help to inform and improve future policy decisions. The FCC is fundamentally changing the way that it approaches problems. Under its current chairman, the FCC has set up a new office of economics and data analytics – an umbrella office that has an impact on every decision that other bureaus make. Instead of just the silos of wireless and media and wireline, the FCC will now have economics and data analytics to help inform its policy decisions. It’s an important breakthrough and part of the impetus for the success of the effort with O.R. research in the incentive auction.

Says FCC Chairman Pai: “The success of [operations research] to solve complicated design and implementation challenges of the incentive auction speaks for itself, and the team’s work is exemplary of the data-driven approach to policymaking that I believe should be this agency’s hallmark.”
Here Come the Judges

The Edelman Award is a nearly yearlong competition that begins with a call for nominations, followed by a vetting and verification process. Once the nominations are culled to six finalists, the competition culminates each spring with presentations – advised by INFORMS-appointed team coaches – before a panel of judges at the INFORMS Conference on Business Analytics and Operations Research. After listening to the presentations and questioning the presenters during a series of sessions, the judges confer behind closed doors to select a winner. Their decision is announced in dramatic fashion at the Edelman Gala that same evening.

So, what put the FCC over the top in a highly competitive and diverse field of Edelman finalists (see sidebar)? We asked several of the Edelman competition judges, all of them longtime members of INFORMS, their thoughts:

  • “It was complete in the sense that it was a hard technical problem, an original methodology, and it had immediate and lasting policy impact.” - Harrison Schramm, CAP, CANA Advisors
  • “The most amazing thing is the FCC had to satisfy so many different constituencies – thousands of TV stations and governmental officials – on getting approval. It was a totally transparent process, and operations research made it possible. It never would have happened without the mathematical rigor and the objectivity.” - Michael Gorman, University of Dayton
  • “All the Edelman finalists were wonderful. The FCC had a unique application with a lot of creative and technical O.R. work underneath it. They built sophisticated algorithms to calculate in real time and figure out whether the bids are acceptable or not, which allowed the whole auction mechanism to run through and create a solution. Considering the amount of bipartisan support it took, and to have the Canadian government agree to get involved in the process, as well as the broadcasting and wireless industries, it had a huge impact.” - Pooja Dewan, BNSF Railway
  • “I think the judges were impressed with the level of complexity and the amount of change management that was required for a government organization to deliver such an enormous level of analytical work and drive such tremendous value.” - Anne Robinson, Verizon and competition chair

First awarded in 1972, the Franz Edelman Award, considered the “Super Bowl of O.R.,” recognizes and rewards outstanding contributions of advanced analytics and O.R. in the for- and non-profit sectors around the globe. Each year, INFORMS honors finalist teams that have improved organizational efficiency, increased profits, brought better products to consumers, helped foster peace negotiations and saved lives. Since its inception, the cumulative dollar benefits from Edelman finalist projects have surpassed $250 billion (not including this year). ORMS

Peter Horner (peter.horner@informs.org) is the editor of OR/MS Today and Analytics magazines.

Sources: FCC, INFORMS, Gurobi

Visit these Advertiser Sites