Online Monitoring of Cyclical Processes for Quality Assurance and Machine Diagnosis

ibaInCycle provides specialized functionality to monitor process data online in real time, focusing on cyclical and repeating machine operations. By aligning measurement data to cycles instead of pure time, the software allows a detailed and precise evaluation of machines and processes that follow repeating patterns.

Cycle-based monitoring gives users immediate insight into deviations and anomalies that may otherwise remain hidden in time-based views. With configurable profiles and rules, the system continuously checks process signals against predefined conditions, enabling early detection of irregularities and supporting preventive maintenance strategies.

Key capabilities

• Monitor process data cycle by cycle for precise evaluation of repetitive operations.
• Define profiles and monitoring rules to capture process-specific conditions.
• Detect deviations, alarms, or limit violations directly during ongoing operation.
• Visualize cycles in real time with overlay and comparison functions.
• Support preventive maintenance by identifying irregular behavior early.

This online monitoring approach makes ibaInCycle an essential tool for plants where reliability and consistency in cyclical processes are critical.

Typical use cases

• Monitoring rotating machinery such as turbines or compressors.
• Detecting irregularities in cyclical production processes like presses, rolling mills, or extruders.
• Early identification of wear or imbalance in mechanical components.
• Supporting predictive maintenance through continuous condition monitoring.

By transforming real-time signal data into cycle-based insights, ibaInCycle enables engineers to keep their processes stable, reliable, and efficient.

ibaInCycle simplifies configuration by enabling the automatic learning of monitoring limits from real production data. Instead of defining thresholds manually, the software analyzes recorded cycles and derives characteristic values that serve as dynamic reference points for future monitoring.

This learning functionality makes it easier and faster to set up reliable monitoring profiles. By basing limits on actual machine behavior, the system ensures that thresholds are realistic, robust, and tailored to the specific process. Engineers can then validate and adjust these values as needed, striking the right balance between sensitivity and stability.

Key capabilities

• Automatically determine limit values from real process data.
• Accelerate setup of monitoring profiles by reducing manual configuration effort.
• Adapt thresholds dynamically to match actual machine behavior.
• Ensure realistic monitoring conditions based on production-relevant cycles.
• Support continuous improvement by refining limits as more data becomes available.

By learning limits from the process itself, ibaInCycle provides a data-driven approach to monitoring configuration, improving both efficiency and reliability.

Typical use cases

• Quick commissioning of cycle monitoring for new machines.
• Deriving reference values from production data during test runs.
• Improving accuracy of alarm thresholds by basing them on real behavior.
• Adapting monitoring profiles to process changes over time.

With its automatic limit learning, ibaInCycle turns historical cycles into smart baselines for stable and precise monitoring.

ibaInCycle extends beyond rule-based monitoring by enabling the integration and deployment of machine learning (ML) models directly into cyclical process monitoring. This makes it possible to capture complex patterns, correlations, and anomalies that traditional threshold-based methods may overlook.

Machine learning models can be trained on historical cycle data and then embedded into ibaInCycle for online evaluation. Once deployed, they operate in real time, continuously classifying cycles, predicting deviations, or detecting unusual behaviors. By combining cycle-based monitoring with data-driven intelligence, ibaInCycle provides a powerful framework for advanced process supervision.

Key capabilities

• Deploy trained machine learning models directly within the monitoring environment.
• Use historical cycle data as a basis for model training and refinement.
• Detect complex patterns and anomalies beyond simple threshold checks.
• Run real-time evaluations of each new cycle using ML insights.
• Combine ML-based monitoring with rule-based approaches for maximum reliability.

This functionality allows engineers to move from reactive to predictive process monitoring, reducing downtime and improving product quality.

Typical use cases

• Classifying cycles into normal and abnormal patterns automatically.
• Detecting subtle changes in machine behavior before failures occur.
• Predicting process deviations that may affect product quality.
• Complementing rule-based monitoring with AI-driven evaluations.

By embedding machine learning models into its workflow, ibaInCycle transforms cyclical monitoring into a proactive, intelligent, and future-ready solution.