RSIMS Platform

A multifunctional predictive diagnostics platform

Keep track of the status of machines and devices, create reports and generate predictive models in one place. An original analytical engine allows the use of machine learning without the need to program or knowledge the theory of algorithms. We are continually responding to emerging innovations and trends, which in combination with our comprehensive scientific approach and continuous improvement of the platform keep it up to date and innovative at all times.

The biggest advantage of RSIMS platform is the analytical engine

The analytical engine is used to create predictive models and inference. It is also based on measurement, operational and production data, documentation, monitoring systems and expert knowledge. The information is collected in a dedicated database. An important part is the data pre-processing module addressing data problems (e.g. deficiencies) by transforming them, which results in higher quality. These data use innovative machine learning methods tailored by RS to maintenance prediction and prescription issues. Over 20 machine learning methods create the optimal model, which is then added to the library of analytical models with a number of models of many types of machines (e.g. turbines, engines).

Data analysis: 3-level approach

Data analysis is crucial for obtaining valuable information from available data. The method must take into account not only the limitations of the data source and available expert knowledge, but also the issues of the resources (time, money, commitment, etc.) allocated to the PdM project. Therefore, the RS data analysis method is made up of three levels - from the simplest to the most sophisticated, to increase the chances of success within the known limitations.

In order to forecast when thresholds are exceeded, you can use technological thresholds (for machines and processes) as a basis or anomaly detection to obtain baseline signal thresholds. The next step is a decision tree model that validates thresholds and builds forecasts.

Dedicated models monitor the condition of the machine (e.g. vibrations, temperatures) identifying significant changes indicating a developing failure. This approach is based on domain knowledge, operation physics and advanced statistics.

Advanced machine learning methods are used to predict specific events with a specific time horizon, and multi-dimensional anomaly detection algorithms support the operator by alerting them of an abnormal situation indicating the potential location of the deviation.

One platform - a variety of possibilities!

The RSIMS platform provides a package of functionalities necessary to implement Predictive / Prescriptive maintenance strategies.
Root cause analysis
Remaining useful life estimation
Anomaly detection
Predictive Maintenance
Prescriptive Maintenance
Process optimization

Root cause analysis

Ability to identify causation within the data

  • ability to consider full huge datasets (thousands of sensors),
  • determining and validating rules undetectable manually,
  • good insight and hints for operators or designers,
  • mixture of domain know-how with data science,
  • proprietary data pre-processing techniques,
  • rule association determination,
  • basket analysis.

Remaining useful life estimation

Long-term service planning

  • combines both pure statistics and Predictive Maintenance advantages,
  • adjusts its output by the current sensor information,
  • returns the estimated lifetime of a machine,
  • accuracy increased over time.

Anomaly detection

Full control over process stability

  • proprietary self-learning, unsupervised algorithms, including Deep Learning,
  • gives valuable information about what signals generate the failure,
  • does not require any additional information or historical data,
  • detects any new events within the monitored machinery.

Predictive Maintenance

Predict and react

  • proprietary data preprocessing and models generation modules,
  • fully scalable, automated models generation and recalibration,
  • self-learning mechanisms,
  • holistic approach – analyzing all the available data (for problems propagating through the installation),
  • predicting unwanted events for crucial machinery.

Prescriptive Maintenance

Support decision making

  • usage of PdM models for process steering to avoid unwanted events (e.g. failures),
  • holistic approach guarantees high accuracy and reliability of models,
  • decision support system,
  • closed-loop solution.


Process optimization

Increase profits, minimize expenses

  • including machinery condition, current market state, contract penalties, etc.,
  • using metamodels for simulations of process to find the optimal steering,
  • using proprietary optimization mechanisms (versions of PSO),
  • maximizing overall profits.

Why is our solution perfect for you?

Easy implementation

- use of existing sensors
- easy integration
- modular construction

Ease of use

- intuitive user panel
- a multi-level notification system
- automated recalibration of models

Easily scalable

- a library of predictive models
- modular implementation
- scalable computational architecture


- average accuracy of models 96%
- the ability to enter domain knowledge
- continuous improvement through recalibration

Architecture and licensing method

System in the local network or in the RS cloud

On Premise

  • Subscription model for fees for using the software – OPEX operating costs
  • Necessary investments in owned IT infrastructure – CAPEX investments
  • Starting access to the system requires a dedicated software installation
  • Full control over customer safety
  • Responsibility for maintaining and updating the system on the client’s side
  • IT resources and data on the client’s premises


  • Subscription model for fees for using the software – OPEX operating costs
  • No need to invest in your own infrastructure
  • Quick access to the system
  • High level of security ensured by cloud delivery
  • Responsibility for maintaining and updating the system on the ReliaSol side
  • IT resources and data off the premises of the RSIMS client

The system on a local
network or in the RS cloud

Where are our systems used?

The functionalities of our platform will let you solve many important problems of your company and optimize its operation, maximizing profits

Benefits of implementing the RSIMS platform

In addition to technical advancement, our solutions are focused on the economic benefits of cutting down costs

80 %
minimization of the number of failures
< 50 %
minimization of the number of unplanned downtime
30 %
minimization of maintenance costs
25 %
increase of production profits
> 10
return on investment