Workshop

Session: From Optimization to Intelligence: What's New in Gurobi 14 and the Future of AI

Presenters: Oliver Bastert, Chief Technology Officer, Gurobi | Robert Luce,  Senior Director of Optimization R&D, Gurobi | Lindsay Montanari, Senior Director of Academic Programs, Gurobi

Location: TBA

Date & Time: Saturday, October 31 | 3:00 pm – 4:15 pm

Details: Discover the latest innovations in Gurobi Optimizer 14 and explore how optimization is evolving alongside generative AI. This interactive session will provide an overview of new capabilities, introduce the Gurobean game, and showcase how Gurobi's GAIL team is applying AI to help users solve complex decision-making challenges. Bring your questions for an open discussion with Gurobi experts.

Technology Showcase

Session: Your Optimization Copilot: Accelerating Better Decisions with Gurobi Intelligence Hub

Presenters: Caroline Weinberg, Manager of Optimization Support, Gurobi | Nicholas Parham, Technical Account Manager, Gurobi​

Location: Room 101

Date & Time: Sunday, November 1 | 1:15 pm – 1:50 pm

Details: Imagine giving every stakeholder the ability to interact with optimization through natural language and AI-guided assistance. This session showcases Gurobi Intelligence Hub and the intelligent capabilities designed to simplify model development, improve collaboration, and help organizations turn optimization expertise into business impact at scale. See how AI can help bridge the gap between complex models and confident decision-making.

Invited Sessions

Session: Brewing Better Decisions: Gamification as a Gateway to Optimization with Gurobean

Presenters: Lindsay Montanari & Larry Snyder

Location: Marriott Marquis-Sierra E (5th Floor)

Date & Time: Sunday, November 1 | 4:50 pm – 5:05 pm

Details: Gurobean is an interactive educational game designed to introduce core operations research concepts through a simulated coffee shop. Players make operational decisions involving inventory, pricing, and service capacity under stochastic demand. Progressive rounds expose learners to feasibility, nonlinear trade-offs, queueing effects, and simulation-based optimization. Developed by Gurobi Optimization and Larry Snyder, Gurobean demonstrates how gamification can support experiential learning in OR education.

Session: MIP Restrictions for nonconvex MINLPs

Presenters: Hassan Hijazi

Location: Moscone South-209

Date & Time: Monday, November 2 | 11:30 am – 11:45 am

Details: MIP restrictions are a promising tool for finding high-quality feasible solutions to hard non-convex MINLPs. The feasible region of a restriction is guaranteed to be a subset of the original problem’s feasible region, ensuring that solutions obtained during Branch-and-Bound exploration remain feasible to all nonlinear constraints. This property is particularly valuable in applications with tight time limits. In this talk, we define MIP restrictions and discuss methods for automatically generating them for generic MINLPs. We also present use cases, including models from the ARPA-E Grid Optimization Competition.

Session: Gurobi Intelligence: GenAI and Gurobi

Presenters: Caroline Wineberg

Location: Moscone South-210

Date & Time: Monday, November 2 | 11:30 am – 11:45 am

Details: This session showcases the Gurobi Intelligence Hub, our home for GenAI-powered optimization tools. We will highlight real-world use cases and key lessons learned from deploying our Gurobot, Explainer, and Modeler agents, and share our perspective on the most promising directions for integrating generative AI with mathematical optimization. Attendees will also get a preview of upcoming developments and learn how to access our agents programmatically.

Session: Recent Developments in the Gurobi MILP Solver

Presenters: Oliver Bastert

Location: Moscone South-209

Date & Time: Monday, November 2 | 11:45 am – 12:00 pm

Details: Performance improvements to the Gurobi MILP Solver will be discussed. In particular, we will focus on very hard problems that take many branch-and-bound nodes to solve.

Session: Computational Experiments with QPUs and Nested Dissection Ordering in the Barrier Algorithm for Linear Programming

Presenters: Ed Klotz

Location: Moscone South-211

Date & Time: Monday, November 2 | 4:45 pm – 5:00 pm

Details: Among the many long term scenarios for the impact of quantum computing on mathematical optimization, hybrid quantum/classical methods are promising because they can leverage the strengths of quantum computers for specific operations while leaving calculations on which they are less effective to classical computers. Probably the most common hybrid methods involve using the Quantum Processor Unit (QPU) for fast heuristics to find good solutions for Mixed Integer Programs that would otherwise take longer to find. Another common method involves solving subproblems in a decomposition algorithm more effectively than can be done classically. Both of these are primarily applied to optimization problems with discrete variables. However, other hybrid approaches exist, including ones that can help with linear programs (LPs) and other continuous optimization problems. Specifically LP algorithms rely on heuristics for the underlying, NP hard, ordering problems in order to find good orderings for the matrix factorizations they compute. This raises the question of whether QPUs can effectively provide better orderings that result in faster overall performance for the algorithm. This talk will consider such an approach for the Cholesky factorization used in the barrier algorithm for linear programs.

Session: Recent Developments in the Gurobi Optimizer

Presenters: Dan Steffy

Location: Moscone South-207

Date & Time: Monday, November 2 | 4:30 pm – 4:45 pm

Details: This talk will discuss new features and performance enhancements in the Gurobi Optimizer for LP, MILP and Global MINLP. We will also present preliminary performance results for the upcoming release.

Session: Nonlinear Modeling with gurobipy

Presenters: Robert Luce

Location: Moscone South-206

Date & Time: Tuesday, November 3 | 5:00 pm – 5:15 pm

Details: In this talk we walk through the recent additions to gurobipy for modeling nonlinear optimization problems. The new expression-based interface allows nonlinear relationships to be formulated directly using algebraic expressions, providing a natural mapping between mathematical models and their implementation. We will introduce the underlying modeling concepts and discuss vectorized operations on variables, expressions, and constraints. These capabilities enable concise model formulations and support efficient implementation patterns, particularly for applications involving large-scale collections of optimization objects. Several examples will demonstrate the resulting modeling workflow and code structure.