Having a process under control, being able to predict the quality of the final product, avoid machine failures, identify before its time risky situation, execute prior maintainance, and much more are all aspects that can at last be realisticly faced.

Every process, or part of it, has a set of PLC or distributed control system (DCS) or a set os SCADA systems. Furthermore often are available ERP software that collect data regarding the logistics, the raw materials in input to the productive cycle and of the quality of the products in output from the process.

In the field of Industry 4.0 we put ourselves in the context of the Big Data Analytics, having the fundamental role to extract the relevant information from the data under analysis using all the available data sources, without preclusions or restrictions. It is pretty clear that the traditional statistics techniques (aka, Statistical Process Control), are no more sufficient to have an accurate process control. Having knowledge of suitable open source and free software tools, we use a set of multivariate statistics techniques (aka Multivariate Statistical Process Control) extremely useful in those kind of contexts.

We collaborate and we collaborated with lot of companies in this sector, having acquired the experience directly on the field in the most various contexts.

Currently available software tools allow in fact to acquire, manage and process huge amount of data in a small amount of time. We have the resources and the skills to manage the phases of:

  • data collection
  • database data model project and development to manage the data of the process to be analyzed
  • development of tools for visualization of the results of the analysis, like dashboards

Our typical approach to the process control management is divided in the following phases: first of all we collect all the data, structured or not, from the available data sources. Those sources could be sensors, PLC, SCADA, csv files, ERP, other databases. It follows a phase in which the process is analyzed, from the evaluation of the data quality to the definition of the goal of the entire work. Examples of goals of the job could be the process control, monitoring, product quality check, optimization of products or process. Is in this phase that we prepare the algorithms to clean, synchronize and prepare raw data, and the algorithms for classification/regression.

Once the entire set of tools and algorithms are prepared, it follows the final part of the system development: this can be a static report containing all the information extracted from the analysis of the process, but it can also be a web application or a library that can be used by who manages the UI/UX part.

Multivariate process control

Architecture design for data analysis tools

Process data analysis dashboard

Development of application Industry 4.0 enabler