Date: August 25, 3 — 4 pm UTC (convert to your local time)
This webinar will showcase how combined simulation and machine learning advance decision support in business and public enterprise.
Both simulation modeling and machine learning help inform and improve decision making. Simulation modeling derives its predictive capability from the causal rules embedded into the model. In contrast, machine learning derives its power from information stored in historical data. Since these two approaches are rooted in fundamentally different methods, the two have yet to find overlap in application.
With the current pandemic, having reliable predictions is as crucial as ever, both for public health scenarios and for pivotal business decisions under uncertainty. Now, advancements in simulation and machine learning combine the best of both worlds to provide quick, accurate, and exhaustive solutions.
H2O Driverless AI automates time-consuming ML tasks so that data scientists can work faster and more efficiently. Automated tasks include: model validation, model tuning, model selection, and feature engineering.
In this webinar we will demonstrate a simulation example for capacity planning and management in hospitals that are experiencing surges in patients due to COVID-19. Using this example, we showcase how to improve the predictive capability of the model by embedding an H2O Driverless AI MOJO pipeline that can predict the length-of-stay for each patient arrival based on their individual characteristics.
- Introduction to H2O Driverless AI technology
- Simulation Modeling vs Machine Learning
- Simulation Modeling plus Machine Learning
- Basics of H2O driverless AI; example of predicting patients’ length-of-stay based on their attributes
- Capacity planning in hospitals using multi-method modeling and machine learning
- Process of incorporating a trained ML model (AI MOJO pipeline) into an AnyLogic model
Dr. Arash Mahdavi is a simulation modeling expert and AI Program Lead at The AnyLogic Company in North America. He holds a PhD degree in civil engineering from Purdue University where he applied a system-of-systems approach and agent-based modeling to profitability analysis of construction companies. Dr. Mahdavi has recently authored a simulation textbook titled “The Art of Process-Centric Modeling”. He has trained hundreds of professionals and faculty members from Fortune 100 companies and elite research universities.
Dr. Niki Athanasiadou is a Customer Data Scientist at H2O AI with a passion for data-driven knowledge. Coming from a PhD on the microscopic universe of biomolecules, Niki is bringing scientific thinking to real-world big data. Niki has experience in healthcare among other sectors and loves to work in interdisciplinary teams. Her proudest moments are winning the Young Biochemist of the Year award by the British Biochemical Society and the Open Data data-science project award from the Office of the Mayor of New York City.