AI-based machine condition monitoring software
We created an AI-based solution to manage the health of the car fleet and convert vehicle data into valuable insights to predict and prevent failures, showing a condition monitoring software case study. It allows an estimate of the Remaining Useful Life (RUL) of components and provides early detection of malfunctions and performance degradation.
Modern vehicles are costly to fix and maintain. For example, warranty claims, maintenance, and vehicle downtime costs affect the car fleet and influence customer satisfaction. To identify and settle these costs, vehicle health management systems, or VHM systems, use sensors and software to monitor vehicle performance, detect and predict problems, and recommend needed maintenance.
Today, most VHM systems rely on diagnostic trouble codes. These codes identify an issue only after it has already occurred. Besides, they provide a piece of limited information only. It adds extra work for technicians to identify the cause of the problem.
Modern vehicles use telematics to send diagnostic trouble codes and additional information to the cloud. Still, telematics data is too general and limited. As an effective alternative to the current solutions, our client created an AI-powered solution to monitor the health of vehicles effectively, processing thousands of different signals in real-time.
One of the project’s main goals was to create a possibility for users to compare and evaluate different anomaly detection models on unsupervised data.
- Duration: March 2019 - Ongoing
- Location: Israel
- Industry: Automotive
- Data science analysis, Product support
Modern vehicles are complex machines that are costly to fix and maintain. For instance, warranty claims, recalls, maintenance, and vehicle downtime costs reduce commercial fleet and customer satisfaction. To tackle these challenges, vehicle health management systems (VHM) utilize sensors and machine condition monitoring software to monitor vehicle performance, detect and predict issues, and suggest necessary maintenance.
Today, most machine condition monitoring solutions rely on diagnostic trouble codes. These codes identify an issue only after it has already occurred. Besides, they provide a piece of limited information. It adds extra work for technicians to identify the cause of the problem.
Modern vehicles employ telematics and machine health monitoring to transmit diagnostic trouble codes and additional data to the cloud, enhancing vehicle health assessment. Still, telematics data is too general and limited. As an effective alternative to the current software, we developed AI-powered machine condition monitoring solutions to effectively assess the health of vehicles, processing thousands of different signals in real-time.
– One of the project’s goals was to create a possibility for users to compare and evaluate different anomaly detection models on unsupervised data.
– The critical goal was to enable faster selection of an unsupervised anomaly detection model for every new client dataset, even those with no or very few labeled anomalies, using a machine health monitoring system.
– Another goal is to develop a robust data augmentation framework adjustable for different client datasets and a machine protection system to help recognize unknown types of anomalies.
– The client needed to strengthen the in-house development team with skilled ML developers and Data Scientists.
– Our client was looking for a development team with good communication skills ready to start fast.
- Select the anomaly. Tailoring unsupervised anomaly detection models for specific client datasets, with a particular emphasis on their application in AI vehicle control systems
- Anomaly streaming. Offline and online streaming anomaly detection
- Detection and compression. Changepoint detection and data compression
- Data augmentation. Automatic data augmentation
- Samples generating. Generating synthetic abnormal samples for each client dataset having no ground truth anomalous data available
Using Deep Learning technologies, the client’s solution can determine the vehicle’s health conditions and then flag any deviations from the expected behavior.
This equipment monitoring system can also predict when a vehicle’s health condition could lead to failure and help identify the cause of the problem.
– Unlike other similar machine monitoring systems that can only operate in the cloud, this one can operate onboard vehicles in real-time, assisting fleet owners to predict vehicle health issues.
– CHI Software Data Scientist is currently working on implementing AI software for vehicles and testing Deep Learning models for unsupervised anomaly detection and auto-ml solutions based on synthetically generated anomalous data along with research on related state-of-the-art models and novel ideas in this field.
– The tasks of the CHI Software’s Machine Learning Engineer include obtaining models from Data Scientists, running them on Big Data, and parallelizing calculations on the cloud.
Our technology stack
- Microsoft Azure
- Building precise and powerful anomaly detection tools for offline and online streaming anomaly detection in time-series automotive data.
- Developing a robust framework for data augmentation applicable to various kinds of data.
- Recognition of previously unseen anomalies.
As you know, Data Scientists analyze a massive amount of data for different actionable insights on a machine monitoring project. Working on the machinery monitoring system, I have tasks to identify the data-analytics problems that offer the best opportunities, collect large sets of structured and unstructured data from disparate sources, analyze and interpret the data to discover solutions and options. What motivates me is working on a valuable product and using my skills and expertise to improve the client's initial solution.