# Graph-to-SFILES: Control structure prediction from process topologies using generative artificial intelligence

arXiv: https://arxiv.org/abs/2412.00508 · HTML: https://mlsanity.com/paper/2412.00508/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 58 references that are in the map.
- First posted on arXiv 30 Nov 2024.

## Builds on (most cited first, 22 of 22)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [Variational Graph Auto-Encoders](https://mlsanity.com/paper/1611.07308/) (2016)
- [How Attentive are Graph Attention Networks?](https://mlsanity.com/paper/2105.14491/) (2021)
- [A Generalization of Transformer Networks to Graphs](https://mlsanity.com/paper/2012.09699/) (2020)
- [Masked Label Prediction: Unified Message Passing Model for Semi-Supervised Classification](https://mlsanity.com/paper/2009.03509/) (2020)
- [GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Models](https://mlsanity.com/paper/1802.08773/) (2018)
- [Multimodal Learning with Transformers: A Survey](https://mlsanity.com/paper/2206.06488/) (2022)
- [MolGAN: An implicit generative model for small molecular graphs](https://mlsanity.com/paper/1805.11973/) (2018)
- [Molecular Transformer - A Model for Uncertainty-Calibrated Chemical Reaction Prediction](https://mlsanity.com/paper/1811.02633/) (2018)
- [Analyzing Uncertainty in Neural Machine Translation](https://mlsanity.com/paper/1803.00047/) (2018)
- [Permutation invariant graph-to-sequence model for template-free retrosynthesis and reaction prediction](https://mlsanity.com/paper/2110.09681/) (2021)
- [Graph-to-Sequence Learning using Gated Graph Neural Networks](https://mlsanity.com/paper/1806.09835/) (2018)
- [A Graph-to-Sequence Model for AMR-to-Text Generation](https://mlsanity.com/paper/1805.02473/) (2018)
- [Graph2Seq: Graph to Sequence Learning with Attention-based Neural Networks](https://mlsanity.com/paper/1804.00823/) (2018)
- [Toward automatic generation of control structures for process flow diagrams with large language models](https://mlsanity.com/paper/2211.05583/) (2022)
- [Flowsheet synthesis through hierarchical reinforcement learning and graph neural networks](https://mlsanity.com/paper/2207.12051/) (2022)
- [SFILES 2.0: An extended text-based flowsheet representation](https://mlsanity.com/paper/2208.00778/) (2022)
- [Learning from flowsheets: A generative transformer model for autocompletion of flowsheets](https://mlsanity.com/paper/2208.00859/) (2022)
- [Deep reinforcement learning for process design: Review and perspective](https://mlsanity.com/paper/2308.07822/) (2023)
- [Toward autocorrection of chemical process flowsheets using large language models](https://mlsanity.com/paper/2312.02873/) (2023)

## Built on by (most cited first, 5 of 5)

- [Multi-agent systems for chemical engineering: A review and perspective](https://mlsanity.com/paper/2508.07880/) (2025)
- [Federated Learning from Molecules to Processes: A Perspective](https://mlsanity.com/paper/2506.18525/) (2025)
- [FaultExplainer: Leveraging Large Language Models for Interpretable Fault Detection and Diagnosis](https://mlsanity.com/paper/2412.14492/) (2024)
- [GraphRAG for Engineering Diagrams: ChatP&ID Enables LLM Interaction with P&IDs](https://mlsanity.com/paper/2603.22528/) (2026)
- [LLMs in Process Diagram Engineering: From Optimal PFDs to Validated P&IDs](https://mlsanity.com/paper/2608.11220/) (2026)

## Abstract (arXiv)

Control structure design is an important but tedious step in P&ID development. Generative artificial intelligence (AI) promises to reduce P&ID development time by supporting engineers. Previous research on generative AI in chemical process design mainly represented processes by sequences. However, graphs offer a promising alternative because of their permutation invariance. We propose the Graph-to-SFILES model, a generative AI method to predict control structures from flowsheet topologies. The Graph-to-SFILES model takes the flowsheet topology as a graph input and returns a control-extended flowsheet as a sequence in the SFILES 2.0 notation. We compare four different graph encoder architectures, one of them being a graph neural network (GNN) proposed in this work. The Graph-to-SFILES model achieves a top-5 accuracy of 73.2% when trained on 10,000 flowsheet topologies. In addition, the proposed GNN performs best among the encoder architectures. Compared to a purely sequence-based approach, the Graph-to-SFILES model improves the top-5 accuracy for a relatively small training dataset of 1,000 flowsheets from 0.9% to 28.4%. However, the sequence-based approach performs better on a large-scale dataset of 100,000 flowsheets. These results highlight the potential of graph-based AI models to accelerate P&ID development in small-data regimes but their effectiveness on industry relevant case studies still needs to be investigated.