Gurobi Day Korea
April 17, 2025

Iain Jardin #
Regional Director, APJ, Gurobi Optimization

Dr. Lee Kangju
Account Director, Korea, Gurobi Optimization

This presentation will give an overview of Gurobi Optimization, an update on the global industry trends, and customer use cases.

Dr. Kostja Siefen
Senior Director of Technical Account Management, Gurobi Optimization

Optimization technology plays a crucial role in supporting rational decision-making across various industries. With advancements in artificial intelligence and the increasing availability of data, AI-driven optimization techniques are emerging, enabling the resolution of more complex problems. However, practical challenges such as data scarcity, high computational costs, and integration issues with existing systems often hinder the effective application of optimization methods in industrial settings. This presentation explores the latest trends in optimization technology and analyzes the real-world challenges faced during its implementation. Additionally, we propose practical approaches and solutions to enhance the applicability of optimization techniques, providing insights into strategies for their successful adoption in industrial environments.

Prof. KangBok Lee
Industrial & Management Engineering, POSTECH

This presentation examines the role of Generative AI, which is rapidly revolutionizing approaches to complex optimization problems, from three key perspectives. Firstly, we will examine AI-driven optimization algorithms, ranging from pure end-to-end AI-based algorithms to hybrid approaches that combine AI with metaheuristics and mathematical optimization algorithms. Secondly, let’s discuss automated algorithm design methods where AI systems construct optimization methods and search for high-quality solutions. Thirdly, we will investigate the role of large language models (LLMs) in optimization, analyzing their potential as prompters for problem definition, modelers for optimization problem formulation, and solvers for solution generation. As the design of systems where interconnected AI agents collaborate across industrial optimization processes becomes feasible, this will fundamentally transform our approach to complex optimization challenges.

Prof. Changhyun Kown
Industrial and Systems Engineering, KAIST

In this session, we will be chatting with Dr. Zonghao Gu, Co-Founder of Gurobi on the history of Gurobi Optimization, the challenges and the successes faced along the way, the future of optimization, and sharing his memorable experiences as an entrepreneur. He will also be answering selected pre-submitted questions received from the attendees. This is a rare opportunity to hear directly from Dr. Gu about his experiences and vision while gaining insights into the past, present, and future of optimization.

Dr. Zonghao Gu
Chief Technology Officer and Co-founder, Gurobi Optimization

• Track 1 – Business Session
• Track 2 – Technical Session

• Track 1 – Business Session
• Track 2 – Technical Session

In the era of digital innovation, businesses are increasingly relying on data-driven decision-making to strengthen their competitiveness. LG CNS has been supporting clients across various industries by leveraging mathematical optimization consulting to maximize business performance.In this presentation, we will introduce LG CNS’s approach to mathematical optimization, along with consulting case studies from industries such as manufacturing and logistics. We will delve into the tangible outcomes achieved by improving the accuracy and agility of decision-making through optimization models, with real-world examples like production scheduling and logistics network optimization. Additionally, we will demonstrate how our cloud-based SaaS solution, SCMable, helps businesses apply optimization models more easily and enhance their decision-making capabilities.

This presentation will provide participants with insights into how mathematical optimization can turn data into actionable insights and create measurable business value.

Moo-Sung Sohn
Consultant Leader, LG CNS

Satellite constellation design is becoming increasingly complex. Beyond just providing global coverage, there is a growing demand for focused regional observation, precise mission execution, and real-time data delivery. As user expectations rise, these evolving requirements make optimization challenges even more difficult, while also maintaining cost-effectiveness.

Particularly, autonomous distributed satellite systems require inter-satellite cooperation and real-time decision-making. Unlike traditional centralized operations, where a centralized control center manages all satellites, distributed environments demand new optimization approaches that allow satellites to make autonomous decisions while coordinating with each other. This presentation will explore the latest research methodologies addressing these challenges and their real-world applications.

Additionally, active research is being conducted on accelerating optimization solvers in edge computing environments, further expanding the scope of aerospace optimization. As the space industry continues to evolve, this presentation will provide an opportunity to explore how optimization is shaping the future and driving advancements in space technology.

Dongshik Won, Ph.D. Candidate, KAIST
Director, TelePIX

Hospitals and healthcare institutions have significant potential to improve the efficiency of their workforce and resource management. Interest in this area is steadily growing. The Nurse Rostering Problem (NRP), despite being extensively studied over the years, remains a challenge in many healthcare settings. Even today, duty rosters are often created manually, taking more than six hours to complete.

In this presentation, we will introduce the nurse scheduling problem and present a nurse scheduling solution developed using Gurobi. We will also share case studies of its practical implementation in real healthcare environments.

Jun Kim
CEO, WorkFlip

Supply chain management encompasses a broad set of processes, from production to delivery, aimed at meeting customer needs. Practitioners often rely on diverse heuristics to optimize these processes, thanks to their ease of implementation and rapid solution times. However, a commonly overlooked aspect is the lack of procedures to gauge solution quality relative to the true optimum. In certain operations, it is crucial not only to enhance solution optimality but also to measure the gap from the best possible outcome in a reasonable amount of time, as this can save substantial resources and highlight additional cost-saving opportunities. To address these problems, Kurly, a leading Korean online grocer, adopted Gurobi Optimizer, a commercial solver. In this presentation, we will showcase Kurly’s use of Gurobi Optimizer in tackling challenges across its warehouse and last-mile delivery operations and illustrate how Gurobi surpasses other solver in improving Kurly’s critical logistics processes.

Juyoung Wang
Data Scientist, Kurly

Power Systems are among the most complex infrastructures ever built by humanity, requiring stable and efficient operation. One of the key optimization challenges in this domain is Unit Commitment (UC), which focuses on minimizing power system operation costs by determining the on/off status and output levels of power generators to meet hourly electricity demand.

This problem is formulated as a Mixed-Integer Programming (MIP) model, incorporating various constraints such as generator operational status, output limits, and ramping restrictions. As power system size increases, the problem complexity grows exponentially, making the role of optimization solvers crucial.

In this presentation, we provide a brief introduction to power system operations and the unit commitment problem. We then showcase a real-world case study where Gurobi was used to efficiently solve large-scale unit commitment planning problems. Through this case study, we will explore Gurobi’s performance and discuss its potential applications in the domestic power system industry.

Dr. Yong-Kyu Kim
Senior Researcher, Korea Electrotechnology Research Institute

This presentation introduces a software solution that enables easy modeling of energy network systems (including steam, electricity, fuel, and CO₂) for operation optimization and energy production planning based on demand. The software features visual modeling, allowing users to intuitively create energy network models by simply dragging and dropping components. It also collects real-time data from the energy grid, taking into account equipment variables and operational constraints to determine the optimal operation and production plan.
The modeled energy network system is converted into an MILP (Mixed-Integer Linear Programming) problem, which is efficiently solved using the Gurobi solver, ensuring fast computation speeds and highly accurate solutions.
Currently, the solution is deployed and actively used at over 20 sites worldwide. As the need for adaptive energy demand management and dynamic pricing optimization continues to grow, this solution has become an essential tool for efficient energy infrastructure operation.

Wonchoul Kim
Team Leader, Infotrol Technology

The presentation will provide insights into how Gurobi’s custom GPTs—Gurobi AI Modelling Prompt Engineer & Gurobi AI Modelling Assistant—can guide users through the modeling process, from ideation to solution implementation. We’ll walk through practical classic optimization problems, to showcase how these tools can streamline problem definition, model creation, and solution implementation. You’ll gain an understanding of best practices, common pitfalls, and how these tools can accelerate the learning curve for those new to optimization.

Dr. Chung-Kyun Han
Technical Account Manager, Gurobi Optimization

This session explores the latest enhancements in Gurobi 12.0, including significant performance improvements and the new NLExpr object for intuitive nonlinear modeling. Attendees will gain insights into the challenges of nonlinear optimization and its key applications across finance, power systems, and engineering sectors. The session then introduces best practices for modeling and solving nonlinear problems, covering essential techniques such as scaling and variable bounding. Finally, we conclude with Gurobi’s evolving role in the MINLP solver ecosystem and a look at future advancements in nonlinear optimization.

Dr. Zonghao Gu
Chief Technology Officer and Co-founder, Gurobi Optimization

Gurobi offers more than just solving LPs, MIPs, and MIQCPs—it also provides many useful features that can enhance your optimization work. In this talk, we will review modeling features such as multiple objectives, scenarios, solution pools, and more. Additionally, we will introduce tools that help users utilize servers more effectively.

Dr. Chung-Kyun Han
Technical Account Manager, Gurobi Optimization

This presentation provides a detailed overview of Gurobi’s architecture options and licensing mechanisms, helping users make informed decisions on model building and solving strategies. It covers three key architectural approaches: local solving, centralized solving with Compute Server, and cloud-based solving using Gurobi Instant Cloud. The session also compares licensing mechanisms, including Machine-bound, Floating, and Web License Service (WLS), and explains which mechanisms best fit various environments—from on-premise deployments to containers and serverless infrastructures. Additionally, participants will gain practical insights into choosing the optimal license type based on key factors such as flexibility, security, performance, usage patterns, and scalability.

Dr. Chung-Kyun Han
Technical Account Manager, Gurobi Optimization

• gurobi-machinelearning: Integrating machine learning models with optimization workflows
• gurobi-optimods: Providing pre-built modules for common optimization problems
• gurobi-modelanalyzer: Diagnosing numerical issues and comparing heuristic results

Participants will gain practical insights and hands-on techniques to bridge theory with implementation and enhance their optimization workflows.

Dr. Kostja Siefen
Senior Director of Technical Account Management, Gurobi Optimization

This presentation offers a practical guide to tackling complex mathematical models through effective diagnosis and advanced formulation strategies. Participants will explore proven techniques to troubleshoot model failures and enhance solver performance, especially for long-running models with performance bottlenecks. The session covers best practices for data scaling to ensure numerical stability, techniques for managing nonlinearities to simplify model complexity, and methods for building stronger formulations to accelerate solver convergence.

At the end of the presentation, participants will walk away with a clear understanding of how to diagnose performance issues, apply advanced formulation strategies, and leverage best practices to create efficient mathematical models for their most challenging problems.

Dr. Chung-Kyun Han
Technical Account Manager, Gurobi Optimization