Generative AI and Process Systems Engineering: The Next Frontier
arXiv:2402.10977 · doi:10.1016/j.compchemeng.2024.108723
Abstract
This article explores how emerging generative artificial intelligence (GenAI) models, such as large language models (LLMs), can enhance solution methodologies within process systems engineering (PSE). These cutting-edge GenAI models, particularly foundation models (FMs), which are pre-trained on extensive, general-purpose datasets, offer versatile adaptability for a broad range of tasks, including responding to queries, image generation, and complex decision-making. Given the close relationship between advancements in PSE and developments in computing and systems technologies, exploring the synergy between GenAI and PSE is essential. We begin our discussion with a compact overview of both classic and emerging GenAI models, including FMs, and then dive into their applications within key PSE domains: synthesis and design, optimization and integration, and process monitoring and control. In each domain, we explore how GenAI models could potentially advance PSE methodologies, providing insights and prospects for each area. Furthermore, the article identifies and discusses potential challenges in fully leveraging GenAI within PSE, including multiscale modeling, data requirements, evaluation metrics and benchmarks, and trust and safety, thereby deepening the discourse on effective GenAI integration into systems analysis, design, optimization, operations, monitoring, and control. This paper provides a guide for future research focused on the applications of emerging GenAI in PSE.
References in corpus (59)
- Training language models to follow instructions with human feedback
- LLaMA: Open and Efficient Foundation Language Models
- An Introduction to Variational Autoencoders
- Sparks of Artificial General Intelligence: Early experiments with GPT-4
- Normalizing Flows: An Introduction and Review of Current Methods
- Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model
- Learning Combinatorial Optimization Algorithms over Graphs
- Surrogate Modeling for Fluid Flows Based on Physics-Constrained Deep Learning Without Simulation Data
- BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models
- Self-Referencing Embedded Strings (SELFIES): A 100% robust molecular string representation
- Metaheuristic Design of Feedforward Neural Networks: A Review of Two Decades of Research
- Deep UQ: Learning deep neural network surrogate models for high dimensional uncertainty quantification
- MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models
- MolGAN: An implicit generative model for small molecular graphs
- Optimization under Uncertainty in the Era of Big Data and Deep Learning: When Machine Learning Meets Mathematical Programming
- Scaling Autoregressive Models for Content-Rich Text-to-Image Generation
- Mol-CycleGAN - a generative model for molecular optimization
- Microstructural Materials Design via Deep Adversarial Learning Methodology
- Computer-aided molecular design: An introduction and review of tools, applications, and solution techniques
- 14 Examples of How LLMs Can Transform Materials Science and Chemistry: A Reflection on a Large Language Model Hackathon
- GraphAF: a Flow-based Autoregressive Model for Molecular Graph Generation
- GeoDiff: a Geometric Diffusion Model for Molecular Conformation Generation
- LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention
- VisionLLM: Large Language Model is also an Open-Ended Decoder for Vision-Centric Tasks
- Deep Reinforcement Learning with Shallow Controllers: An Experimental Application to PID Tuning
- LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model
- Deep Learning and Knowledge-Based Methods for Computer Aided Molecular Design -- Toward a Unified Approach: State-of-the-Art and Future Directions
- On Recurrent Neural Networks for learning-based control: recent results and ideas for future developments
- Learning the aerodynamic design of supercritical airfoils through deep reinforcement learning
- Reinforcement Learning for Integer Programming: Learning to Cut
- SE(3) diffusion model with application to protein backbone generation
- Quantum Computing Assisted Deep Learning for Fault Detection and Diagnosis in Industrial Process Systems
- Learning from flowsheets: A generative transformer model for autocompletion of flowsheets
- Physical Pooling Functions in Graph Neural Networks for Molecular Property Prediction
- SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehension
- Global Optimization of Gaussian processes
- Flowsheet synthesis through hierarchical reinforcement learning and graph neural networks
- Graph Machine Learning for Design of High-Octane Fuels
- Distributionally Robust Chance Constrained Programming with Generative Adversarial Networks (GANs)
- SFILES 2.0: An extended text-based flowsheet representation
- Scaling Autoregressive Multi-Modal Models: Pretraining and Instruction Tuning
- Multimodal Foundation Models: From Specialists to General-Purpose Assistants
- A Dynamic LLM-Powered Agent Network for Task-Oriented Agent Collaboration
- Learning To Cut By Looking Ahead: Cutting Plane Selection via Imitation Learning
- On data-driven chance constraint learning for mixed-integer optimization problems
- MUDiff: Unified Diffusion for Complete Molecule Generation
- Learning Cut Selection for Mixed-Integer Linear Programming via Hierarchical Sequence Model
- GPT-MolBERTa: GPT Molecular Features Language Model for molecular property prediction
- Prophet: Prompting Large Language Models with Complementary Answer Heuristics for Knowledge-based Visual Question Answering
- How to Bridge the Gap between Modalities: Survey on Multimodal Large Language Model
- Generative Diffusion Models on Graphs: Methods and Applications
- Harmonic Self-Conditioned Flow Matching for Multi-Ligand Docking and Binding Site Design
- Synthesis of separation processes with reinforcement learning
- Data-Driven Robust Optimization using Unsupervised Deep Learning
- Language Quantized AutoEncoders: Towards Unsupervised Text-Image Alignment
- In-Context Learning for Few-Shot Molecular Property Prediction
- Deep Reinforcement Learning for Process Synthesis
- Multimodal Deep Learning for Scientific Imaging Interpretation
- Deep reinforcement learning for process design: Review and perspective