Increasing complexity and process speed pose significant challenges for technology-intensive manufacturing companies. Not only do these factors make processes more difficult to understand, analyze, maintain and optimize, but they also complicate smooth cooperation between various departments. Different departments — such as maintenance, production, technology and quality management — often follow different work habits and procedures, leading to untapped synergy potential. These synergies can only be realized when employees establish a common ground for communication. A proven approach is the recording and analysis of process data. However, this step also requires the introduction and implementation of a unified and holistic digitalization strategy.

Maximizing the use of measurement data

Measurement data should not belong to individual departments or be recorded for a single purpose and then hidden in departmental data silos. Instead, they must cover the entire process and be utilized across departments. Only then can the full potential of the measurement data be unlocked.

Figure 1: Measurement data should not belong to individual departments or be recorded for a single purpose and then hidden in departmental data silos. Source: iba America

High-resolution and comprehensive data collection

A key prerequisite is the method of data collection: Only when data is recorded as high-resolution raw values, rather than pre-aggregated, can key performance indicators (KPIs) be flexibly calculated according to user and task-specific aspects and then related to each other. To analyze the causes of unusual behavior patterns without losing information, and thus improve process understanding, high-resolution raw data must remain easily and quickly accessible from the KPIs via drill-down. This creates a company-wide and unified data basis, simplifying discussions and enhancing cross-departmental collaboration.

To ensure that all departments can collaborate effectively when implementing the digitalization strategy, it is essential to capture the entire process. Individual measurements, such as vibration or energy data, must not be recorded separately but instead made available in a unified data set along with all other process data. This approach allows for the identification of correlations between production, processes and any wear-related effects, which is crucial for future optimization efforts.

Process connectivity as a crucial element

The heterogeneity of automation components within manufacturing plants presents a significant challenge for measurement systems. The more complex the process, the greater the number of interacting signal sources and automation systems. In addition to programmable logic controllers (PLCs), sensors and bus monitors, cameras can also be used as data sources. Especially when no sensors are available for critical process variables, or they cannot be installed due to cost or space constraints, time-synchronized video data provides valuable information. Comprehensive process connectivity is thus a crucial element of a powerful measurement system: All data, regardless of manufacturer, device generation or data format, can be captured isochronously and with high resolution, offering users a global and non-intrusive view of the technical process.

ibaPDA: The scalable core software in measuring value acquisition

ibaPDA (process data acquisition) provides the basis for the unification of high-speed process data across complex and distributed automation systems and is the core product of the iba system. ibaPDA is an extremely powerful, PC-based software for acquiring and recording different process data in automated technical environments.

One special feature of ibaPDA is the exceptionally broad connectivity for acquiring different data types with different acquisition methods in heterogeneous systems. This includes analog and digital I/O signals, signals from field and drive buses, data from the control system, production data, product characteristic values, energy data, vibration data and descriptive additional information. This broad connectivity is crucial in enabling a continuous consistent acquisition of data from an entire plant. ibaPDA is also scalable and is suitable both for individual test stands as well as for factory-wide plants in which several thousand signals are acquired. This method of centralized data collection and time stamping makes it possible for engineers and technicians to identify causal relationships and easily determine sequence of events within high-speed automation systems.

Figure 2: Increasing complexity and process speed pose significant challenges for technology-intensive manufacturing companies. Source: iba America

Scalable sampling rates

The rate at which process data is sampled plays a key role in the usability of the acquired data. This is especially important in high-speed industrial processes or when troubleshooting systems with extraordinarily fast clock speeds. These fast collection speeds are crucial for identifying sequence of events when troubleshooting. With ibaPDA, synchronous data collection can be achieved at rates up to 500 kHz.

Figure 3: ibaPDA is an extremely powerful, PC-based software for acquiring and recording different process data in automated technical environments. Source: iba America

Long-term data recording and analysis with ibaHD-Server

To analyze measurement data over a long period and thus identify trends or demonstrate compliance with quality requirements, the data must be efficiently stored. This is achieved using a historical data server (ibaHD-Server), with continuous recording or recording triggered by events, depending on the need. Event-triggered data recording is ideal when only signals related to a specific workpiece should be captured in a file. The start and stop conditions can be customized and derived directly from the recorded signals, making product-specific evaluations easier and improving the comparability of similar products.

Figure 4: Large datasets can be interactively searched, analyzed and utilized for key figure development and report generation. Source: iba America

Large datasets can be interactively searched, analyzed and utilized for key figure development and report generation. The actual increase in value comes from the analysis of the measurement data. When applying the described methods and techniques for data acquisition and recording, it is ensured that the data can be used by various user groups, such as maintenance teams, process engineers, production and quality managers, as well as data analysts.

In addition to interactive offline analysis and the calculation of key metrics for abstracting, evaluating and documenting the process, the role of automated real-time analysis is becoming increasingly important for the early detection of process deviations, especially in the context of proactive maintenance.

ibaHD-Server at a glance

• Continuous recording of measured data and events over a long time periodFigure 5: To analyze measurement data over a long period and thus identify trends or demonstrate compliance with quality requirements, the data must be efficiently stored. Source: iba America
• Simultaneous recording from several ibaPDA systems and import of measurement files
• Direct access to historical data with intuitive use for visualization, such as scrolling or jumping to a date
• Quick zoom function from the annual, monthly or weekly overview down to the range of milliseconds
• Display and filtering of historical events and joint visualization with measurement data
• Storage of measurement data and additional information in defined time periods enables the fast analysis of shifts and process steps
• Long-term analysis of historical data with ibaAnalyzer and ibaDaVIS/InSightBI
• Automatic calculation of KPIs and automatic reporting with ibaDatCoordinator and ibaAnalyzer
• API access for third party data analysis applications

Time-synchronous recording and analysis of videos and process data

With the increasing availability and affordability of cameras, video data has become an important tool for many users within an industrial setting. Especially when no sensors are available for critical process variables, or they cannot be installed due to cost or space constraints, time-synchronized video data provides valuable information to operators and engineers alike. ibaCapture records video and HMI images time-synchronously with ibaPDA — either continuously or triggered by events. Important events can be automatically stored as still images. The exact relation and simultaneous display of recorded process data and visual information offers completely new and robust capabilities for process analysis.

Figure 6: With the increasing availability and affordability of cameras, video data has become an important tool for many users within an industrial setting. Source: iba America

ibaCapture highlights

• Synchronous recording of video images and measurement data with ibaPDAFigure 7: Automatic online evaluation allows real-time proactive monitoring of a process to detect and avoid process errors early or to improve product quality. Source: iba America
• Continuous and event-triggered recording
• Integration of HMI images and images from ibaVision
• Protected storage areas for important sequences
• Capturing up to 64 cameras (analog, IP, GigE or HMI) and online visualization
• View and analyze video sequences and measurement data with ibaAnalyzer
• Live image display as replacement for a CCTV system
• Event-triggered switching of display-layouts (Scenario Player)

Automatic online evaluation

Automatic online evaluation allows real-time proactive monitoring of a process to detect and avoid process errors early or to improve product quality. This type of evaluation is also known as edge analytics, as the measurement data are processed directly at the edge, between the production and IT networks — exactly where the data are generated.

Examples of edge analytics available within the iba system:

1. Threshold monitoring with ibaPDA: Signal values can be monitored in the acquisition system for specific threshold limits. If these limits are exceeded or not met, an alarm is triggered. In addition to raw signals, the signals can also be aggregated, linked or averaged, and the threshold check can be conditioned by specific parameters. For example, a threshold check can be performed only when the machine is in automatic mode.
2. Machine vision applications using ibaCapture/ibaVision: Machine vision (MV) applications allow numerical values, text or classified information such as "material present/not present" to be extracted from video data. Since video data is captured synchronously with process data and the MV information is available in real-time, the results of the MV application can be treated like sensor values and recorded in the measurement system. With MV, products can be measured, material tracking can be realized or the process can be monitored more effectively.
3. Online frequency analysis for vibration monitoring using ibaInSpectra: If the vibration behavior of a process is recorded using IEPE sensors, the necessary frequency domain analysis can be automatically performed directly on the edge device. Frequency bands to be monitored can be defined either as fixed or dependent on process values and checked for threshold violations. Process vibrations are thus detected in real-time, and an alarm is triggered in case of a fault.
4. Monitoring using TSA and auto-adapting with ibaInCycle: The time synchronous averaging (TSA) method can monitor a process based on already captured measurement signals, such as temperature, pressure or motor current. Based on the signal progression, both slowly emerging process deviations, for example due to wear, and sporadic anomalies can be detected early, and their effects on product quality and machine condition can be reliably predicted. In the time domain, wear indicators are calculated online, and alarms are triggered in real-time when thresholds are exceeded. This allows users to make timely adjustments, preventing machine or system malfunctions, negative impacts on product quality and unplanned downtime. The process behavior can be automatically learned for different conditions using the auto-adapting method.

The key to increasing value through digitalization lies in the multiple and shared use of measurement data. Only when the digitalization strategy is implemented collaboratively across all departments can cross-departmental synergies be realized.

To achieve this, appropriate tool support is essential, such as the solutions provided by the iba system. This ensures that access to the data is simple and intuitive, allowing each user group to independently carry out analyses tailored to their specific needs.

Figure 8: In the time domain, wear indicators are calculated online, and alarms are triggered in real-time when thresholds are exceeded. Source: iba America