Automated RAM Analysis

The Same Method Was Utilised in Both Use Cases Sawmill & Hydro Power Plant.

• Unique solution that combines quick and easy history analysis methods and sophisticated modelling and simulation methods integrated to modern IoT platform
• Visibility into overall cost risks. Decision making is based on analysed facts
• Payback time for analysis can be proved through the cost savings and benefits that can be achieved
• Solution makes RAM analysis a continuous process and results more visible
• Fast algorithms, automated analysis, proven methods and making the results visible contribute to sale of analysis tools and solutions

Input(s):

• Device hierarchy and the logical structure of the system under study
• Event history (all type of events, resources, failures, repair duration, maintenance data etc.)
• Cost history (resources, spare parts, break and downtime costs etc.)
• Expert knowledge

Main feature(s):

Automated RAM Analysis Prototype:

• Data Input Interfaces
• Modelling
• Simulation
• Analysis results and reports
• Data Export Interfaces

RAMS Simulation Gateway Prototype:

• Import Interface
• Combining IoT data and Automated RAM Analysis model and simulation
• Dashboard

(RAM = Reliability, Availability and Maintainability)

Output(s):

• Comprehensive and up-to-date RAM analysis
• Visualisation of history data and simulation results
• Recognise hot spots for sensors and digital solutions
• Highlight improvement potentials
• Visibility into reliability, availability and overall cost risks
• Improvements to reliability, availability and cost risks

• Hardware requirements: a) Memory: 4GB RAM (more than 8GB RAM recommended for large models) b) Hard drive: 100MB c) Operating system: Windows (7/8/10), macOS (10.7-10.15), Linux
• The object under study: It has been shown that automatic RAM analysis can be performed in very different processes. Therefore, we do not see any technical constraints to extend the analysis to other processes as well.
• Data quality: The quality of the data collected from the object under study is important for the outcome. When the data is consistent, available, and reliable, we get correspondingly better results. However, this is also a challenge, and the data can be processed and supplemented by expert knowledge, for example. The analysis also reveals qualitative deficiencies in the data and thus efforts can be made to improve data quality.

Maintenance engineer, Reliability expert.

Licensing, many options.