Litmus Edge Provides Built-in Statistical Functions to Enhance Accuracy and Reliability of Data Analysis

There are significant advantages of using prebuilt statistical functions than writing, maintaining, and using custom scripts built overtime. So, Litmus Edge includes built-in functions that are available off-the-shelf in the platform from Day 1. In this article, we highlight the benefits and sample use cases where these statistical functions can help users do more with their data.

ICYMI: Litmus Edge documentation is now publicly available, without requiring a sign-in. Click here to learn how to use these functions.

Benefits of Using Our Built-in Statistical Functions

1. Reduced complexity: Built-in statistical functions can help to reduce the complexity of developing and maintaining IIoT applications. With a library of pre-built functions, developers can focus on the business logic of their applications, rather than having to implement statistical functions from scratch.

2. Improved performance: Built-in statistical functions can also help to improve the performance of IIoT applications. By optimizing the implementation of these functions, Litmus Edge provides users with access to high-performance statistical analysis capabilities without having to sacrifice accuracy.

3. Increased accessibility: Built-in statistical functions can also help to increase the accessibility of IIoT data analysis to a wider range of users. Litmus Edge provides a simply user interface making it easier for users with limited programming experience to perform complex statistical analysis on their data.

Examples of Litmus Edge Statistical Functions with Sample Use Cases

Here are a few sample scenarios on how our built-in statistical functions can help industrial companies.

Anomaly Detection

• Use case: Identify unusual sensor readings that could indicate a potential problem with a machine or process.

• Example: A manufacturing plant uses anomaly detection to identify machines that are producing defective products. The plant uses a variety of sensors to collect data on each machine, such as temperature, vibration, and output rate. The anomaly detection algorithm analyzes this data to identify machines that are deviating from their normal operating patterns.

Base Conversion

• Use case: Convert sensor data from one format to another.

• Example: A smart building uses base conversion to convert temperature data from Fahrenheit to Celsius. The building uses a variety of sensors to collect temperature data, and the base conversion algorithm is used to convert this data to a common format so that it can be easily analyzed.

Combination Processor

• Use case: Combine data from multiple sensors into a single output.

• Example: A power plant uses a combination processor to combine data from multiple sensors to calculate the total energy output of the plant. The power plant uses a variety of sensors to collect data on the performance of individual generators, and the combination processor is used to combine this data into a single output that can be used to manage the plant's overall energy production.

Conversion

• Use case: Convert sensor data from one unit of measurement to another.

• Example: A weather station uses a conversion algorithm to convert wind speed data from miles per hour to knots. The weather station uses a variety of sensors to collect data on the weather, and the conversion algorithm is used to convert this data to a common format so that it can be easily shared with other users.

Expressions

• Use case: Perform complex calculations on sensor data.

• Example: An oil refinery uses expressions to calculate the optimal operating conditions for its distillation process. The refinery uses a variety of sensors to collect data on the process, and the expressions are used to calculate the optimal values for temperature, pressure, and other variables.

Gaussian Filter

• Use case: Remove noise from sensor data to improve the accuracy of measurements.

  Example: A pharmaceutical manufacturing plant uses a Gaussian filter to remove noise from temperature sensor data. This helps to improve the accuracy of the plant's temperature control system, which is essential for ensuring the quality of the pharmaceutical products being manufactured.

Inputs

• Use case: Collect data from sensors and other devices.

• Example: A smart city uses inputs to collect data from traffic sensors, air quality sensors, and other devices. This data is used to monitor the city's infrastructure and make decisions about how to improve it.

Join Processor

• Use case: Combine data from multiple sensors on a machine to create a complete picture of its operation.

• Example: A manufacturing plant uses a join processor to combine data from sensors on a machine, such as temperature, vibration, and pressure. This data is used to create a unified view of the machine's operation, which can be used to identify potential problems and optimize the machine's performance.

Moving Window

• Use case: Identify trends in the quality of a product over time.

  Example: A food and beverage plant uses a moving window to identify trends in the quality of its products over time. The plant collects data on the quality of its products from a variety of sources, such as quality control tests and customer feedback. The plant then uses a moving window to calculate the average quality of its products over a period of time. This information can be used to identify trends in the quality of the products and to take corrective action if necessary.

Normalization

• Use case: Scale sensor data to a common range.

• Example: A machine learning algorithm uses normalization to scale sensor data from different types of machines to a common range. This helps the algorithm to learn from the data more effectively.

Rise and Fall

• Use case: Identify trends in sensor data.

• Example: A chemical plant can use the rise and fall function to identify trends in the flow rate of chemicals in its production process. If the flow rate of a chemical starts to decrease, this could indicate a problem with the production process. The chemical plant could then investigate the problem and take corrective action to ensure that the production process is operating correctly.

Rounding

• Use case: Reduce the precision of sensor data to improve efficiency.

• Example: A smart home thermostat manufacturer can use rounding to reduce the precision of temperature data. This helps to improve the efficiency of the device and reduce its battery consumption.

Signal Decomposition

• Use case: Identify the different components of a signal to identify potential problems in the bottling process.

  Example: A bottling plant uses a signal decomposition function to identify the different components of a signal from a pressure sensor on its bottling line. For example, if the DC component of the signal starts to decrease, this could indicate a problem with the filling machine. If the AC component of the signal starts to increase, this could indicate a problem with the bottles. If the noise component of the signal starts to increase, this could indicate a problem with the environment in which the bottling line is operating.

Statistical Prediction

• Use case: Predict future values of sensor data.

• Example: A manufacturing plant uses statistical prediction to predict the demand for its products. This information is used by the plant to plan its production schedule and inventory levels.

Try Litmus Edge Today

See for yourself how Litmus Edge can help you to take your business to the next level. Sign up for a free 15-days trial at https://central.litmus.io.

Subscribe to https://litmus.io/updates to stay on top of product updates, bug fixes, and upcoming releases.