Presenters:

• Eric Bechhoefer (GPMS International Inc.)

Description:

Foresight is an edge-processing IoT system that provides aircraft health and usage monitoring. It is certified (STCd) on fourteen different aircraft. This technology demonstration would include a brief presentation and a demo of the Foresight MX that will demonstrate: 

• How Foresight MX enables connecting an onboard control unit (OBCU) with a sensor,
• How the OBCU connects/commands the sensor,
• How the sensor performs analysis, and
• How the OBCU moves that data to AWS for alerting.

Presenters:

• Frank Juarez (PHM Technology)
• Chris Stecki (PHM Technology)

Description:

The Maintenance Aware Design (MADE) Platform enables you to create a Digital Risk Twin (DRT) of your system that seamlessly integrates analyses across a range of engineering disciplines to effectively assess and manage risk at the earliest possible stage of the product life-cycle. MADE combines modelling with analyses to enable trade studies on the safety, reliability and maintainability of complex engineering systems. The PHM module of MADE, is used to design / validate the diagnostic requirements for Condition-Based Maintenance (CBM) of complex systems in an integrated analysis solution.

The demonstration will include a short presentation of the MADE Platform and tutorials on how to:

• Model a Digital Risk Twin in MADE Platform
• Generate automated RAMS Analyses
• Populate the MADE Sensor library
• Identify Sensor set ambiguities
• Conduct Trade studies on Sensor Set Designs
• Generate Diagnostic rules

Presenters:

• Chris Stecki (PHM Technology)
• Rob Marsden (PHM Technology)

Description:

Syndrome Diagnostics (SD) is a real-time FDI tool that uses Causation-based AI (Cb-AI) to detect and isolate failures in complex systems. SD leverages system context from a Digital Risk Model that identifies the sensor sets required, and the diagnostic rules the sensors will produce to indicate specific failures (syndromes).

The demonstration will include a short presentation of the SD workflow, and showcase its ability to:

• Identify failures on components that do not have a sensor
• Identify the minimum labelled data set requirements to ensure system-level coverage of failures
• Provide high-confidence FDI for systems where limited or no labelled data is available
• Ensure that the FDI capability for your system is explainable (to support certification)

Presenter:

• LaRue Brown (Altair)

Description:

The Altair RapidMiner Platform is designed to help businesses leverage data analytics, artificial intelligence, and machine learning to solve complex problems and make data-driven decisions. It offers features such as:

1. Data preparation and cleaning
2. Exploratory data analysis
3. Machine learning model building and deployment and AutoML
4. Predictive analytics
5. Visual workflow design for data science processes
6. Collaboration tools for data science teams
7. IoT (Internet of Things) and Edge Computing

Today, we will focus on presenting an automatic process control of visual check of a weld on a factory floor as an end-to-end utilization of the RapidMiner Platform.

For this technical demonstration, we will mainly focus on two topics:

1. Identifying the location of a weld between two brackets (object detection)
2. Categorizing the weld as good or bad (image classification) automatically using Machine Learning

Once the models are trained, this can be implemented on the factory floor with an edge device so that the object detection and image classification can occur in real time. This will significantly increase the productivity of the factory and mitigate the need for a human being to physically inspect each individual weld.

Presenters:

• Jitendra Solanki (Siemens Corp., Technology)
• Miao He (Siemens Corp., Technology)

Description:

Unlock the power of industrial data with cutting-edge solutions! Discover how virtual PLCs and advanced connectors can seamlessly collect OT data, transforming raw information into valuable insights using Industrial Information Hub (IIH). Learn how Siemens WinCC Unified’s built-in applications can help you harness data for predictive maintenance and effective model management. Plus, see how Siemens Mendix’s low-code platform empowers you to rapidly develop custom applications, revolutionizing the way you consume and act on industrial data. Join us to explore the future of smart manufacturing and data-driven decision-making, all in one Industrial Edge Platform from Siemens, to easily expand existing PHM solutions to similar scenarios via Siemens’ Industrial AI portfolio. The hands-on demo and presentation addresses to all those who consolidate, manage and orchestrate existing PHM solutions down to shopfloor or field.

Presenter:

• Kevin Klasing (GE Aerospace)

Description:

GE Aerospace is pioneering the development of Fleet Monitor X, an advanced system designed to revolutionize the monitoring and management of aircraft fleets. This initiative focuses on several critical areas, including Data Flow, Data Management, Engine Health Monitoring (EHM), Alert Management, Customer Notification Reports, Process Simplification, and ultimately, the Customer Experience.

Fleet Monitor X (FMX) employs sophisticated techniques for data collection, transmission, storage, and analysis to ensure efficient and secure handling of the extensive data generated by aircraft systems. Advanced analytics and machine learning are utilized to provide actionable insights, enhancing operational efficiency and safety.

The system offers comprehensive engine health monitoring capabilities, enabling near real-time tracking of engine performance and early detection of potential issues. This proactive approach helps in maintaining optimal engine health and reducing Unexpected Engine Removals (UER).

The FMX system generates detailed Customer Notification Reports (CNR), providing stakeholders with timely and relevant information about fleet status and performance. These reports are designed to be clear and actionable, facilitating informed decision-making.

Fleet Monitor X is designed with the customer in mind, offering near real-time monitoring, intuitive dashboards, and proactive maintenance capabilities. By providing near real-time updates and diagnostic analytics, the system enhances the customer experience by ensuring timely interventions and minimizing disruptions.

In conclusion, GE Aerospace’s Fleet Monitor X represents a significant advancement in fleet management technology, addressing key challenges in Data Flow, Engine Health Monitoring, alert management, and customer satisfaction. This innovative system underscores GE Aerospace’s commitment to safety, quality, delivery, and cost-effectiveness, ultimately setting a new standard in the aerospace industry.

Presenters:

• Rachel Johnson (Mathworks)
• Bora Eryilmaz (Mathworks)
• Sammit Jain (Mathworks)
• Shyam Joshi (Mathworks)

Description:

In this tech demo, you will learn how to use MATLAB and Predictive Maintenance Toolbox to detect anomalies in time series sensor data and design predictive maintenance algorithms. There will be plenty of time for Q&A. Highlights include:

• Organizing, analyzing, and preprocessing time series sensor data in MATLAB
• Using interactive tools extract features and develop AI models
• Exploring anomalies in sensor data using distance-based pattern matching
• One-class machine learning and deep learning approaches for algorithm development
• Using simulation to generate data for expensive or hard-to-reproduce faults
• Deploying predictive algorithms in production systems and embedded devices

Presenters:

• Rachel Johnson (Mathworks)
• Bora Eryilmaz (Mathworks)
• Sammit Jain (Mathworks)
• Shyam Joshi (Mathworks)

Description:

All necessary MATLAB licenses will be provided for the duration of this workshop. In this hands-on workshop, you will write and execute predictive maintenance examples in MATLAB® Online™ – entirely in the browser – to learn and explore how to apply principles of AI to predictive maintenance: machine learning, deep learning, feature extraction, and domain-specific data processing.

This interactive hands-on session will include the following:

• Familiarizing yourself with MATLAB Online and AI tools with an introductory example that trains a machine learning model to classify faults.
• Exploring how to extract features in the time and frequency domains, and rank them to obtain the most relevant features to train your AI model.
• Diving deep into an advanced predictive maintenance workflow that covers anomaly detection and remaining useful life estimation.

Prerequisites:

• Spots are limited – Sign up by Wednesday afternoon at the MathWorks booth!
• Bring a fully charged laptop

Presenter:

• Jubilee Rao (GTC Analytics)

Description:

Model-Based Systems Engineering (MBSE) combined with Enterprise Architecture (EA) offers a transformative approach to institutional management through digital twins. Leveraging MBSE and the Unified Architecture Framework (UAF), this technology creates a comprehensive digital model that consolidates data, processes, and capabilities into a unified source of truth. This model supports data-driven decisions, reduces risks, accelerates decision timelines, and enhances resource allocation efficiency.

By linking facilities to organizational programs, the MBSE-EA digital twin improves budget justification and uncovers new funding opportunities. This tech demo will showcase how this integration empowers decision-makers at NASA with an agile, data-rich foundation for optimizing investments and institutional management.