In-Person event
Gurobi Industry Days: Telecom
April 5-6, 2022
3:00 PM
Event Recap
Gurobi Industry Days: Telecommunications on April 5 – 6, 2022 was a success. You heard from global telecom experts how mathematical optimization is transforming critical roles in the telecommunications industry, including:
Network planning, including fiber network, FTTH and 5G
Retail location planning and supply chain challenges
Maximizing the use of your spectrum
Optimizing your backhaul network for the new demands of 5G and FTTH
Our speakers presented how they’re using mathematical optimization to overcome complex, mission-critical challenges amid global change, supply chain disruptions, regulatory restrictions, and complex interdependencies.
For those of you that missed it or just want a refresher, here is the agenda, speaker info and presentations:
Agenda
Tuesday, April 5
Gurobi Optimization – “Introduction”
Sascha Haake, Sales Director – EMEAI
Welcome of all participants, quick introduction to Gurobi technology.
Slides | Recording
University of Missouri – Opening Keynote – “”Innovations via Advanced Analytics in the Telecom Industry”
Haitao Li, Professor
The rapid growth of use of sensors, 5G networks and Internet-of-Things (IoT), plus volatile market conditions and potential supply disruptions, have created significant challenges for the telecom industry. On the other hand, the abundant availability of data empowered by Advanced Analytics enables data-driven decision-support tools to turn challenges to opportunities. In this talk, I will first elaborate the suite of techniques under Advanced Analytics and delineate their connections, differences, and more importantly, synergies with the traditional AI methods such as machine learning and neural networks. We shall then provide a taxonomy of the diverse range of applications in the telecom industry, identify their combinatorial characteristics, and provide a road map of optimization modeling approaches to address decision-making in the deterministic vs. stochastic and static vs. dynamic environment. In the third part of my talk, I will share some best practices for developing industry-strength optimization applications and decision-support tools.
Slides | Recording
Vodafone Germany – “Combining Telecommunications, Optimization, and Machine Learning
Rolf Bardeli, Lead Data Scientist
The telecommunications industry offers a wide variety of settings with optimization problems at their core. These appear both in a pure form and in the combination of discrete optimization and machine learning.
In this presentation, Rolf Bardeli will give an overview of business use cases in which optimization and its combination with machine learning can provide major benefits for network planning, campaign planning, and footprint optimization. He will also discuss how machine learning is used both to define the parameters of the optimization problems and provide warm-start solutions.
Slides | Recording
Federal Communications Commission (FCC) – “Addressing the Spectrum Crunch at the Federal Communications Commission”
Brian Smith, FCC Optimization Team Lead
From mobile devices to improved Wi-Fi, from connected cars to satellites, the demand for more wireless connectivity grows every day. These uses require more dedicated spectrum but much of the available spectrum has already been allocated. Resolving the “spectrum crunch” will require creative solutions and new approaches. In this talk, we will discuss three different ways the Federal Communications Commission has attempted to use the available spectrum more efficiently: repacking television stations, reconfiguring 37 and 39 GHz licensees, and creating dynamic licenses in 3.5 GHz. We will discuss how mathematical optimization could be used in each of these scenarios.
Slides | Recording
Wednesday, April 6
KPN – “More efficient roll-out by route optimization in the backhaul network”
Dick van Huizen, Senior Data Scientist
Flattening the curve has been a hot topic over the last two years in the global news. Similarly, KPN has been trying to flatten the demand of fibers in the backhaul network. The introduction of new technologies, such as 5G and FttH, puts pressure on the capacity of the existing backhaul network. Most of this capacity is currently used by network elements that will be obsolete when all innovations are live. As a result, almost double the backhaul capacity is needed during roll-out. By creating a mathematical optimization model for routing and rerouting of existing connections, KPN has been able to make more efficient use of the existing assets, limiting the need for new cable roll-out. Our optimization model has helped to save CAPEX and increase roll-out speed.
Slides | Recording
baobab soluciones – “The road to 5G requires adding high-level mathematical optimization technology to the telecom’s toolkit”
David Sánchez, Head of Product & Business Development
All optimization solutions require a key decision with business impact, a complex context that constrains that decision making, and a huge variability of options, unmanageable without mathematical tools. Based on baobab’s analysis methodology, we will review some of the common decisions in telecom environments. Furthermore, we will discuss how the main directions in the roadmap to 5G (speed, latency, availability, virtualization, and customization of services, etc.) impose very strict demands on infrastructure control. This path certainly calls for solvent optimization specialists and a first-class mathematical solver.
Slides | Recording
“Demand Forecasting and Optimizing Transportation Decisions Based on Inventory Targets and Store Locations”
Chandu Bhujang, Senior Data Scientist
Businesses are going through most uncertain time in the history. Growing customer expectations, increasing operational complexity and lack of visibility. In this situation, how machine learning and operations research capabilities can be leveraged to successfully solve supply chain related problems?
Slides | Recording
Round Table ““How mathematical optimization, as an AI tool, is solving key industry challenges in the telecom industry”
All of our speakers will gather to discuss this topic. This round table will be moderated by Manuel Rassi, Telecom Account Director at Gurobi.
Recording
Agenda
Tuesday, April 5
Gurobi Optimization – “Introduction”
Sascha Haake, Sales Director – EMEAI
Welcome of all participants, quick introduction to Gurobi technology.
Slides | Recording
University of Missouri – Opening Keynote – “”Innovations via Advanced Analytics in the Telecom Industry”
Haitao Li, Professor
The rapid growth of use of sensors, 5G networks and Internet-of-Things (IoT), plus volatile market conditions and potential supply disruptions, have created significant challenges for the telecom industry. On the other hand, the abundant availability of data empowered by Advanced Analytics enables data-driven decision-support tools to turn challenges to opportunities. In this talk, I will first elaborate the suite of techniques under Advanced Analytics and delineate their connections, differences, and more importantly, synergies with the traditional AI methods such as machine learning and neural networks. We shall then provide a taxonomy of the diverse range of applications in the telecom industry, identify their combinatorial characteristics, and provide a road map of optimization modeling approaches to address decision-making in the deterministic vs. stochastic and static vs. dynamic environment. In the third part of my talk, I will share some best practices for developing industry-strength optimization applications and decision-support tools.
Slides | Recording
Vodafone Germany – “Combining Telecommunications, Optimization, and Machine Learning
Rolf Bardeli, Lead Data Scientist
The telecommunications industry offers a wide variety of settings with optimization problems at their core. These appear both in a pure form and in the combination of discrete optimization and machine learning.
In this presentation, Rolf Bardeli will give an overview of business use cases in which optimization and its combination with machine learning can provide major benefits for network planning, campaign planning, and footprint optimization. He will also discuss how machine learning is used both to define the parameters of the optimization problems and provide warm-start solutions.
Slides | Recording
Federal Communications Commission (FCC) – “Addressing the Spectrum Crunch at the Federal Communications Commission”
Brian Smith, FCC Optimization Team Lead
From mobile devices to improved Wi-Fi, from connected cars to satellites, the demand for more wireless connectivity grows every day. These uses require more dedicated spectrum but much of the available spectrum has already been allocated. Resolving the “spectrum crunch” will require creative solutions and new approaches. In this talk, we will discuss three different ways the Federal Communications Commission has attempted to use the available spectrum more efficiently: repacking television stations, reconfiguring 37 and 39 GHz licensees, and creating dynamic licenses in 3.5 GHz. We will discuss how mathematical optimization could be used in each of these scenarios.
Slides | Recording
Wednesday, April 6
KPN – “More efficient roll-out by route optimization in the backhaul network”
Dick van Huizen, Senior Data Scientist
Flattening the curve has been a hot topic over the last two years in the global news. Similarly, KPN has been trying to flatten the demand of fibers in the backhaul network. The introduction of new technologies, such as 5G and FttH, puts pressure on the capacity of the existing backhaul network. Most of this capacity is currently used by network elements that will be obsolete when all innovations are live. As a result, almost double the backhaul capacity is needed during roll-out. By creating a mathematical optimization model for routing and rerouting of existing connections, KPN has been able to make more efficient use of the existing assets, limiting the need for new cable roll-out. Our optimization model has helped to save CAPEX and increase roll-out speed.
Slides | Recording
baobab soluciones – “The road to 5G requires adding high-level mathematical optimization technology to the telecom’s toolkit”
David Sánchez, Head of Product & Business Development
All optimization solutions require a key decision with business impact, a complex context that constrains that decision making, and a huge variability of options, unmanageable without mathematical tools. Based on baobab’s analysis methodology, we will review some of the common decisions in telecom environments. Furthermore, we will discuss how the main directions in the roadmap to 5G (speed, latency, availability, virtualization, and customization of services, etc.) impose very strict demands on infrastructure control. This path certainly calls for solvent optimization specialists and a first-class mathematical solver.
Slides | Recording
“Demand Forecasting and Optimizing Transportation Decisions Based on Inventory Targets and Store Locations”
Chandu Bhujang, Senior Data Scientist
Businesses are going through most uncertain time in the history. Growing customer expectations, increasing operational complexity and lack of visibility. In this situation, how machine learning and operations research capabilities can be leveraged to successfully solve supply chain related problems?
Slides | Recording
Round Table ““How mathematical optimization, as an AI tool, is solving key industry challenges in the telecom industry”
All of our speakers will gather to discuss this topic. This round table will be moderated by Manuel Rassi, Telecom Account Director at Gurobi.
Recording
Agenda
Tuesday, April 5
Gurobi Optimization – “Introduction”
Sascha Haake, Sales Director – EMEAI
Welcome of all participants, quick introduction to Gurobi technology.
Slides | Recording
University of Missouri – Opening Keynote – “”Innovations via Advanced Analytics in the Telecom Industry”
Haitao Li, Professor
The rapid growth of use of sensors, 5G networks and Internet-of-Things (IoT), plus volatile market conditions and potential supply disruptions, have created significant challenges for the telecom industry. On the other hand, the abundant availability of data empowered by Advanced Analytics enables data-driven decision-support tools to turn challenges to opportunities. In this talk, I will first elaborate the suite of techniques under Advanced Analytics and delineate their connections, differences, and more importantly, synergies with the traditional AI methods such as machine learning and neural networks. We shall then provide a taxonomy of the diverse range of applications in the telecom industry, identify their combinatorial characteristics, and provide a road map of optimization modeling approaches to address decision-making in the deterministic vs. stochastic and static vs. dynamic environment. In the third part of my talk, I will share some best practices for developing industry-strength optimization applications and decision-support tools.
Slides | Recording
Vodafone Germany – “Combining Telecommunications, Optimization, and Machine Learning
Rolf Bardeli, Lead Data Scientist
The telecommunications industry offers a wide variety of settings with optimization problems at their core. These appear both in a pure form and in the combination of discrete optimization and machine learning.
In this presentation, Rolf Bardeli will give an overview of business use cases in which optimization and its combination with machine learning can provide major benefits for network planning, campaign planning, and footprint optimization. He will also discuss how machine learning is used both to define the parameters of the optimization problems and provide warm-start solutions.
Slides | Recording
Federal Communications Commission (FCC) – “Addressing the Spectrum Crunch at the Federal Communications Commission”
Brian Smith, FCC Optimization Team Lead
From mobile devices to improved Wi-Fi, from connected cars to satellites, the demand for more wireless connectivity grows every day. These uses require more dedicated spectrum but much of the available spectrum has already been allocated. Resolving the “spectrum crunch” will require creative solutions and new approaches. In this talk, we will discuss three different ways the Federal Communications Commission has attempted to use the available spectrum more efficiently: repacking television stations, reconfiguring 37 and 39 GHz licensees, and creating dynamic licenses in 3.5 GHz. We will discuss how mathematical optimization could be used in each of these scenarios.
Slides | Recording
Wednesday, April 6
KPN – “More efficient roll-out by route optimization in the backhaul network”
Dick van Huizen, Senior Data Scientist
Flattening the curve has been a hot topic over the last two years in the global news. Similarly, KPN has been trying to flatten the demand of fibers in the backhaul network. The introduction of new technologies, such as 5G and FttH, puts pressure on the capacity of the existing backhaul network. Most of this capacity is currently used by network elements that will be obsolete when all innovations are live. As a result, almost double the backhaul capacity is needed during roll-out. By creating a mathematical optimization model for routing and rerouting of existing connections, KPN has been able to make more efficient use of the existing assets, limiting the need for new cable roll-out. Our optimization model has helped to save CAPEX and increase roll-out speed.
Slides | Recording
baobab soluciones – “The road to 5G requires adding high-level mathematical optimization technology to the telecom’s toolkit”
David Sánchez, Head of Product & Business Development
All optimization solutions require a key decision with business impact, a complex context that constrains that decision making, and a huge variability of options, unmanageable without mathematical tools. Based on baobab’s analysis methodology, we will review some of the common decisions in telecom environments. Furthermore, we will discuss how the main directions in the roadmap to 5G (speed, latency, availability, virtualization, and customization of services, etc.) impose very strict demands on infrastructure control. This path certainly calls for solvent optimization specialists and a first-class mathematical solver.
Slides | Recording
“Demand Forecasting and Optimizing Transportation Decisions Based on Inventory Targets and Store Locations”
Chandu Bhujang, Senior Data Scientist
Businesses are going through most uncertain time in the history. Growing customer expectations, increasing operational complexity and lack of visibility. In this situation, how machine learning and operations research capabilities can be leveraged to successfully solve supply chain related problems?
Slides | Recording
Round Table ““How mathematical optimization, as an AI tool, is solving key industry challenges in the telecom industry”
All of our speakers will gather to discuss this topic. This round table will be moderated by Manuel Rassi, Telecom Account Director at Gurobi.
Recording
Speakers
Meet Your Expert Speaker
Learn from the best in the industry, bringing years of experience and groundbreaking insights to the forefront of AI personalization.
Haitao Li
Professor of Supply Chain & Analytics at University of Missouri
Dr. Haitao Li is Professor and Chair of the Supply Chain & Analytics Department, College of Business Administration at University of Missouri – St Louis (UMSL). Dr. Li has years’ of research experience in optimization modeling and algorithm design, and has been actively working with industry in the application domains of supply chain optimization, resource planning, project scheduling and vehicle routing, among others. He was a recipient of the Young Investigator Award from the US Army Research Office (ARO) in 2010, the Douglas Durand Award for Research Excellence of UMSL in 2015. With two U.S. Patent applications and a number of invention disclosures, he was named 2015 UMSL Inventor of The Year. Dr. Li currently serves as Associate Editor of the Journal of the Operational Research Society and Transportation Journal.
Rolf Bardeli, PhD
lead data scientist in the consumer business unit at Vodafone Germany
Rolf Bardeli, PhD, is currently the lead data scientist in the consumer business unit at Vodafone Germany. He received his MSc (2003) and PhD (2008) in Computer Science from the University of Bonn, Germany. Between trying to teach the computer to recognize bird species by sound during his PhD and trying to teach the computer to understand telco customers at Vodafone, he was a researcher in multimedia pattern recognition at the Fraunhofer Society. Rolf is a senior member of the IEEE and a low-frequency maths blogger at http://mathcination.wordpress.com.
Brian Smith
Operations Research Analyst with NCI Information Systems, Inc
Brian Smith is an Operations Research Analyst with NCI Information Systems, Inc. and project manager for the Optimization Team at the FCC. He has an MS in Operations Research from George Mason University and a BS in Mathematics from the Catholic University of America.
Dick van Huizen
Senior Data Scientist, KPN
Dick van Huizen is a senior Data Scientist at KPN Advanced Analytics TDO. He graduated in Stochastic Operations Research from The University of Twente in 2015, joining Dutch telecom incumbent KPN soon after. Within KPN, Dick mainly focusses on developing optimization algorithms in the fixed network domain. This contains the optimization of cable roll-out plans, rerouting of existing connections and the efficient introduction of new technologies. He does this together with 8 colleague data scientists, all equally passionate about mathematical optimization and its applications.
David Torres Sanchez
Mathematical Optimization QA Engineer
David received his PhD in Operations Research from Lancaster University (UK) in 2019. The topic was aircraft maintenance scheduling and recovery. Since then, David has held research positions at SINTEF Digital (Norway) and Lancaster University, where he has worked on a varied range of combinatorial optimization problems from vehicle routing to multicommodity flow problems.
In his spare time he enjoys bouldering, riding his mountain bike, and maintaining and contributing to several open-source projects.
Speakers
Meet Your Expert Speaker
Learn from the best in the industry, bringing years of experience and groundbreaking insights to the forefront of AI personalization.
Professor of Supply Chain & Analytics at University of Missouri
Haitao Li
Dr. Haitao Li is Professor and Chair of the Supply Chain & Analytics Department, College of Business Administration at University of Missouri – St Louis (UMSL). Dr. Li has years’ of research experience in optimization modeling and algorithm design, and has been actively working with industry in the application domains of supply chain optimization, resource planning, project scheduling and vehicle routing, among others. He was a recipient of the Young Investigator Award from the US Army Research Office (ARO) in 2010, the Douglas Durand Award for Research Excellence of UMSL in 2015. With two U.S. Patent applications and a number of invention disclosures, he was named 2015 UMSL Inventor of The Year. Dr. Li currently serves as Associate Editor of the Journal of the Operational Research Society and Transportation Journal.
lead data scientist in the consumer business unit at Vodafone Germany
Rolf Bardeli, PhD
Rolf Bardeli, PhD, is currently the lead data scientist in the consumer business unit at Vodafone Germany. He received his MSc (2003) and PhD (2008) in Computer Science from the University of Bonn, Germany. Between trying to teach the computer to recognize bird species by sound during his PhD and trying to teach the computer to understand telco customers at Vodafone, he was a researcher in multimedia pattern recognition at the Fraunhofer Society. Rolf is a senior member of the IEEE and a low-frequency maths blogger at http://mathcination.wordpress.com.
Operations Research Analyst with NCI Information Systems, Inc
Brian Smith
Brian Smith is an Operations Research Analyst with NCI Information Systems, Inc. and project manager for the Optimization Team at the FCC. He has an MS in Operations Research from George Mason University and a BS in Mathematics from the Catholic University of America.
Senior Data Scientist, KPN
Dick van Huizen
Dick van Huizen is a senior Data Scientist at KPN Advanced Analytics TDO. He graduated in Stochastic Operations Research from The University of Twente in 2015, joining Dutch telecom incumbent KPN soon after. Within KPN, Dick mainly focusses on developing optimization algorithms in the fixed network domain. This contains the optimization of cable roll-out plans, rerouting of existing connections and the efficient introduction of new technologies. He does this together with 8 colleague data scientists, all equally passionate about mathematical optimization and its applications.
Mathematical Optimization QA Engineer
David Torres Sanchez
David received his PhD in Operations Research from Lancaster University (UK) in 2019. The topic was aircraft maintenance scheduling and recovery. Since then, David has held research positions at SINTEF Digital (Norway) and Lancaster University, where he has worked on a varied range of combinatorial optimization problems from vehicle routing to multicommodity flow problems.
In his spare time he enjoys bouldering, riding his mountain bike, and maintaining and contributing to several open-source projects.
Speakers
Meet Your Expert Speaker
Learn from the best in the industry, bringing years of experience and groundbreaking insights to the forefront of AI personalization.
Haitao Li
Professor of Supply Chain & Analytics at University of Missouri
Dr. Haitao Li is Professor and Chair of the Supply Chain & Analytics Department, College of Business Administration at University of Missouri – St Louis (UMSL). Dr. Li has years’ of research experience in optimization modeling and algorithm design, and has been actively working with industry in the application domains of supply chain optimization, resource planning, project scheduling and vehicle routing, among others. He was a recipient of the Young Investigator Award from the US Army Research Office (ARO) in 2010, the Douglas Durand Award for Research Excellence of UMSL in 2015. With two U.S. Patent applications and a number of invention disclosures, he was named 2015 UMSL Inventor of The Year. Dr. Li currently serves as Associate Editor of the Journal of the Operational Research Society and Transportation Journal.
Rolf Bardeli, PhD
lead data scientist in the consumer business unit at Vodafone Germany
Rolf Bardeli, PhD, is currently the lead data scientist in the consumer business unit at Vodafone Germany. He received his MSc (2003) and PhD (2008) in Computer Science from the University of Bonn, Germany. Between trying to teach the computer to recognize bird species by sound during his PhD and trying to teach the computer to understand telco customers at Vodafone, he was a researcher in multimedia pattern recognition at the Fraunhofer Society. Rolf is a senior member of the IEEE and a low-frequency maths blogger at http://mathcination.wordpress.com.
Brian Smith
Operations Research Analyst with NCI Information Systems, Inc
Brian Smith is an Operations Research Analyst with NCI Information Systems, Inc. and project manager for the Optimization Team at the FCC. He has an MS in Operations Research from George Mason University and a BS in Mathematics from the Catholic University of America.
Dick van Huizen
Senior Data Scientist, KPN
Dick van Huizen is a senior Data Scientist at KPN Advanced Analytics TDO. He graduated in Stochastic Operations Research from The University of Twente in 2015, joining Dutch telecom incumbent KPN soon after. Within KPN, Dick mainly focusses on developing optimization algorithms in the fixed network domain. This contains the optimization of cable roll-out plans, rerouting of existing connections and the efficient introduction of new technologies. He does this together with 8 colleague data scientists, all equally passionate about mathematical optimization and its applications.
David Torres Sanchez
Mathematical Optimization QA Engineer
David received his PhD in Operations Research from Lancaster University (UK) in 2019. The topic was aircraft maintenance scheduling and recovery. Since then, David has held research positions at SINTEF Digital (Norway) and Lancaster University, where he has worked on a varied range of combinatorial optimization problems from vehicle routing to multicommodity flow problems.
In his spare time he enjoys bouldering, riding his mountain bike, and maintaining and contributing to several open-source projects.
