This study presents an eXplainable Artificial Intelligence (XAI) framework to predict the performance of marine scrubbers used for sulfur dioxide ( SO 2 ) removal from marine diesel engine flue gases. Using an aggregated dataset from a roll-on/roll-off (Ro-Ro) cargo ship equipped with an open-loop scrubber, combined with satellite data, the study constructs and evaluates multiple artificial intelligence models, including ensemble models, which were benchmarked against each other using standard regression metrics such as the coefficient of determination (R 2 ), mean absolute error (MAE), and mean squared error (MSE). Results achieve high accuracy R 2 > 0 . 92 and offer insights for optimizing scrubber operations. Nevertheless, artificial intelligence models lack transparency. To overcome this problem, this research integrates post-hoc explainability techniques to elucidate the contributions of various features to model predictions, thereby enhancing interpretability and reliability. The integration of SHapley Additive exPlanations (SHAP) and Explain Like I’m 5 (ELI5) not only confirmed the consistency of feature importance rankings (e.g. seawater acidity level, SO 2 inlet concentration, outlet temperature) but also aligned with the physical-chemical principles of SO 2 absorption. Quantitative comparisons with theoretical expectations demonstrated the reliability of the XAI insights, enhancing both model transparency and interpretability. This can improve the current capability of designing scrubber units by defining more efficient and less expensive options for environmental regulation compliance.

Monitoring of non-linear phenomena, such as pollution dynamics, which is the result of several combined factors and the evolution of environmental conditions, greatly benefits by AI tools; a larger benefit derives by the application of explainable solutions, which are capable of providing elements to understand those dynamics for better informed decisions. In this paper we discuss a case with real data in which a posteriori explanations have been produced after the application of ensemble models.

Modern environmental regulations require rigorous optimization of operations in process engineering to reduce waste, pollution, and risks while maximizing efficiency. However, the nature of chemical plants, which include components with non-linear behavior, challenges the use of consolidated tuning and control techniques. Instead, ad-hoc, self-adapting, and time-variant controls, with a balanced tuning of parameters at both the subsystem and system level, may be necessary. Needed computing processes may require significant resources and high performance systems, if managed by means of traditional approaches and with exact solution methods. In this regard, domain experts suggest instead the use of integrated techniques based on Artificial Intelligence (AI), which include Explainable AI (XAI) and Trustworthy AI (TAI), which are unique in this industry and still in the early stages of development. To pave the way for a real-time, cost-effective solution for this problem, this paper proposes an AI-based approach to model the performance of a real chemical plant, i.e. a marine scrubber installed on a Ro-Ro ship. The study aims to investigate Machine Learning (ML) techniques which can be used to model such processes. Notably, this analysis is the first of its kind, at the best of the authors’ knowledge. Overall, the study highlights the potential of using ML-based techniques, to optimize environmental compliance in the shipping industry.

Composite science workflows are gaining traction to manage the combined effects of (1) extreme hardware heterogeneity in new High Performance Computing (HPC) systems and (2) growing software complexity – effects necessitated by the convergence of traditional HPC with data sciences. Composing, analyzing, and optimizing a composite workflow remains highly challenging as the component technologies are generally developed in isolation and often feature widely varying levels of performance, scalability, and interoperability. In this paper, we propose novel workflow composition and analysis techniques to create and optimize a scalable and effective composite workflow for heterogeneous HPC centers, and define the performance space of variables that impact composite workflow performance. We present PerfFlowAspect, an Aspect Oriented Programming (AOP)-based tool to perform cross-cutting performance analysis of composite workflows and better understand the impact of key performance variables on workflows. Our solution directly addresses AOP concerns that can affect workflow performance and covers the full software lifecycle, ranging from the workflow's initial composition through performance analysis and optimization. We use our science workflow composition techniques to implement the American Heart Association Molecule Screening (AHA MoleS) workflow. Through experimentation, we demonstrate that tuning a single performance variable can improve AHA MoleS workflow performance by a factor of up to 2.45x. Our evaluation suggests that our techniques can significantly enhance the ability of a multi-disciplinary research and development team to create a high performance composite workflow.

Scientific research and development campaigns are materialized by workflows of applications executing on high-performance computing (HPC) systems. These applications con-sist of tasks that can have inter- or intra-application flows of data to achieve the research goals successfully. These dataflows create dependencies among the tasks and cause resource con-tention on shared storage systems, thus limiting the aggregated I/O bandwidth achieved by the workflow. However, these I/O performance issues are often solved by tedious and manual efforts that demand holistic knowledge about the data dependencies in the workflow and the information about the infrastructure being utilized. Taking this into consideration, we design DFMan, a graph-based dataflow management and optimization framework for maximizing I/O bandwidth by leveraging the powerful storage stack on HPC systems to manage data sharing optimally among the tasks in the workflows. In particular, we devise a graph-based optimization algorithm that can leverage an intuitive graph representation of dataflow- and system-related information, and automatically carry out co-scheduling of task and data placement. According to our experiments, DFMan optimizes a wide variety of scientific workflows such as Hurricane 3D on Cloud Model 1 (CM1), Montage Carina Nebula (NGC3372), and an emulated dataflow kernel of the Multiscale Machine-learned Modeling Infrastructure (MuMMI I/O) on the Lassen supercomputer, and improves their aggregated I/O bandwidth by up to 5.42 x, 2.12 x and 1.29 x, respectively, compared to the baseline bandwidth.

The advancement of machine learning techniques and the heterogeneous architectures of most current supercomputers are propelling the demand for large multiscale simulations that can automatically and autonomously couple diverse components and map them to relevant resources to solve complex problems at multiple scales. Nevertheless, despite the recent progress in workflow technologies, current capabilities are limited to coupling two scales. In the first-ever demonstration of using three scales of resolution, we present a scalable and generalizable framework that couples pairs of models using machine learning and in situ feedback. We expand upon the massively parallel Multiscale Machine-Learned Modeling Infrastructure (MuMMI), a recent, award-winning workflow, and generalize the framework beyond its original design. We discuss the challenges and learnings in executing a massive multiscale simulation campaign that utilized over 600,000 node hours on Summit and achieved more than 98% GPU occupancy for more than 83% of the time. We present innovations to enable several orders of magnitude scaling, including simultaneously coordinating 24,000 jobs, and managing several TBs of new data per day and over a billion files in total. Finally, we describe the generalizability of our framework and, with an upcoming open-source release, discuss how the presented framework may be used for new applications.

Computational models can define the functional dynamics of complex systems in exceptional detail. However, many modeling studies face seemingly incommensurate requirements: to gain meaningful insights into some phenomena requires models with high resolution (microscopic) detail that must nevertheless evolve over large (macroscopic) length- and time-scales. Multiscale modeling has become increasingly important to bridge this gap. Executing complex multiscale models on current petascale computers with high levels of parallelism and heterogeneous architectures is challenging. Many distinct types of resources need to be simultaneously managed, such as GPUs and CPUs, memory size and latencies, communication bottlenecks, and filesystem bandwidth. In addition, robustness to failure of compute nodes, network, and filesystems is critical.