LotusEye

LotusEye is a sophisticated AI-powered service specializing in anomaly detection for time-series sensor data. It provides a user-friendly, no-code platform for businesses to monitor the health of their equipment and systems. By analyzing data from various sensors, LotusEye's AI can learn the normal operational patterns and accurately identify deviations that may indicate an impending failure or an unusual event. This proactive approach helps in scheduling predictive maintenance, preventing unexpected downtime, and ensuring quality control in manufacturing processes.

The platform is designed for accessibility, allowing users without deep expertise in artificial intelligence to create, train, and deploy powerful anomaly detection models. It supports flexible data ingestion through both manual CSV file uploads and a REST API for automated, real-time data streams. The intuitive dashboard provides clear visualizations of anomaly scores and sensor data, making it easy to interpret the AI's findings and take action.

How to use LotusEye

Using LotusEye involves a straightforward, multi-step process designed for ease of use:

1. Registration & Setup: Begin by creating an account. During the initial setup, you select a pricing plan (including a free option) and create your first AI model.
2. Model Creation & Training: To create a model, you upload a CSV file containing sensor data from a period of normal equipment operation. This 'training data' teaches the AI what to consider as a baseline. The platform supports both wide-format and long-format CSV files.
3. Data Upload for Analysis: Once the model is trained, you can upload new sensor data (test data) to be analyzed for anomalies. This can be done manually by uploading a file or automatically by integrating with your systems via the REST API (available on paid plans).
4. Monitoring & Visualization: The main dashboard displays a time-series graph of the 'Anomaly Score'. A score exceeding a predefined threshold indicates a potential anomaly. You can also view graphs of individual sensor values to investigate further.
5. Model Refinement: LotusEye allows for continuous improvement. You can retrain the model with more recent data, roll back to a previous model version if a new one performs poorly, and activate a 'False Positive Reduction Mode' to enhance accuracy.
6. Analysis & Reporting: For in-depth analysis, the Business Plan offers a reporting feature that identifies which sensors contributed most to a high anomaly score, helping to pinpoint the root cause of an issue.

Core Features of LotusEye

• AI-Powered Anomaly Detection: Leverages machine learning to learn normal behavior from sensor data and detect subtle deviations.
• Flexible Data Ingestion: Supports manual upload of wide or long-format CSV files and automated data streaming via a REST API.
• Interactive Dashboard: Provides clear visualizations of anomaly scores, thresholds, and raw sensor data over time. Users can add comments to specific data points for collaborative analysis.
• Advanced Model Management: Users can train models with new data, roll back to previous versions, and initialize models.
• False Positive Reduction Mode: A special mode for model training that creates a more robust model less prone to false alarms, ideal for fluctuating sensor data.
• Detailed Analysis Reports: (Business Plan) Generates reports that break down high-anomaly periods, showing the contribution of each sensor to the score.
• Customizable Alerts: Set thresholds for anomaly scores and configure email alert notifications to be informed of critical events immediately.
• Collaborative Workspace: Invite team members with different permission levels (Owner, Operator, Inspector) to collaborate on monitoring and analysis.

Use Cases for LotusEye

LotusEye is ideal for a wide range of industrial and IoT applications:

• Predictive Maintenance: Monitor industrial machinery in factories to detect early signs of wear and tear, preventing catastrophic failures and reducing maintenance costs.
• IoT Device Fleet Monitoring: Track the operational health of a large number of IoT devices in the field to ensure reliability and service continuity.
• Infrastructure Health Monitoring: Analyze sensor data from structures like bridges, buildings, or power grids to detect structural or operational anomalies.
• Manufacturing Quality Control: Monitor sensors on a production line to identify process deviations that could impact product quality.

Advantages of LotusEye

LotusEye offers significant advantages for businesses looking to implement AI-driven monitoring:

• Proactive Problem Solving: Shifts maintenance from a reactive to a predictive model, saving time and money.
• No-Code AI Accessibility: Empowers domain experts, not just data scientists, to build and manage sophisticated AI models.
• Scalability: Offers a range of plans from a free tier for initial testing to comprehensive business plans for enterprise-level deployment.
• Automation and Integration: The REST API allows for seamless integration into existing data pipelines and automated monitoring workflows.
• Improved Accuracy Over Time: Features like model retraining and false positive reduction ensure the system adapts to changing conditions and becomes more accurate.

Pricing and Plans

LotusEye operates on a freemium model, providing several tiers to suit different needs:

• Free Plan: A great starting point to test the core functionality with limitations on features and data volume. Only one Free plan model is allowed per account.
• Light Plan: An entry-level paid plan that unlocks key features like the REST API, training from graphs, model rollback, and the false positive reduction mode.
• Standard Plan: Offers increased capacity and more advanced features over the Light plan.
• Business Plan: The top-tier plan designed for enterprise use, including all features such as the advanced Analysis Report.

Payments for paid plans are handled securely through Stripe. Users can upgrade, downgrade, or cancel their subscriptions through the model management settings.