Data analytics

Data analytics is the process of exploring, cleaning, transforming and modeling data in order to extract useful information, make decisions and predictions. In the context of the “Global Secure IoT Platform”, analytics plays a key role in processing large volumes of information from various devices.

Analytic types

Descriptive Analytics

• Purpose: To describe the current state of the system based on historical data.
• Examples: Gas consumption reports, graphs of pressure, temperature and other parameters.
• Tools: SQL queries, data visualization (graphs, charts).

Diagnostic Analytics

• Purpose: Understanding the causes of the events that occurred.
• Examples: Analyzing the causes of accidents, identifying anomalies in the data.
• Tools: Correlation analysis, finding dependencies.

Predictive Analytics

• Purpose: Predicting future events based on historical data.
• Examples: Forecasting gas consumption, predicting equipment failures.
• Tools: Machine learning, regression analysis.

Prescriptive Analytics

• Purpose: Recommendations for process optimization based on data analysis.
• Examples: Optimization of session scheduling, equipment maintenance recommendations.
• Tools: Optimization algorithms, decision support systems.

Analytics mechanisms

Data Collection

• Data Sources: IoT devices, servers, cloud storage.
• Data formats: JSON, XML, CSV, binary data.
• Collection Frequency: Real-time, periodic polling.

Data cleaning and preprocessing.

• Noise Removal: Filtering out incorrect data.
• Normalization: Bringing data to a consistent format.
• Aggregation: Combining data to simplify analysis.

Data Storage

• Databases: PostgreSQL, MongoDB .
• Data Warehouses: Self-hosted solutions or cloud providers.
• Caching: Redis.

Data Visualization

• Tools: ApexCharts, Chartkick, Mermaid.
• Heatmaps: Data Distribution Visualization.
• Graphs and Charts: Line graphs, histograms, pie charts.