BI/ AI solutions

BI (Business Intelligence) solutions

BI solutions are processes based on advanced technologies, whose aim is to turn the tested data into practical insights. The data can be collected from internal and external sources. It can be made accessible to different users in the organization, from analysts to the CEO, by creating simple sections, built-in reports, and other applications upon necessity. The collected information helps the user understand trends and create consistent methods for utilizing the business.

Stages in the BI process

Analysis – the first stage in the BI process is the analysis of existing information, understanding the business processes, mapping them, and identifying weaknesses.

Data modeling – data modeling means adjusting the architecture to the solution, mapping the information sources, and building a data model that defines the information flow.

Data migration –  we consume information from a variety of sources, internal and external, including CRM and ERP systems, cloud systems, and the Internet. Data migration means receiving information from these sources and making them an integral part of the organization’s information system.

ETL & ELT – the need to analyze information from various sources and technologies requires transferring and processing data. We define the data extraction process according to the existing information sources and the client’s needs. We then load the data and store it in the designated datastore.

Data warehouse – today’s massive information requires organizations to build an organized datastore with convenient access to the analysis.

Dashboards & KPI – the last stage in creating the solution is creating a convenient, user-friendly display of the collected information.  We design reports and dashboards according to defined targets, enabling organizations to derive insights, develop new products, and improve existing work processes.

AI solutions – predicted maintenance

There are various maintenance methods for MES solutions. The popular ones are preventative maintenance and break down maintenance.

Preventative maintenance- we define for every item (machine, manufacturing line, etc.), the required maintenance intervals. Maintenance type and intervals are mostly defined according to the factory’s best practices, its familiarity with the equipment, common types of failures, manufacturer recommendations, etc.

Break down maintenance – treatment is provided only in the event of an actual failure. Intervals are defined in this case as well. However, they are significantly longer than the intervals of preventative maintenance, and the factory relies on local repairment in the event of a failure.

Predictive maintenance – over recent years, we have been witnessing a revolution in maintenance management with the use of ML and AI tools. In this method, we predict the time in which a specific machine can break down and provide treatment to prevent the predicted failure.

OEE system

Management and monitoring of the OEE indicators from the manufacturing machines. This solution is based on an infrastructural connection between the system and the factory’s control mechanism. The system reads the OEE indicators and displays them in real time. OEE indicators are crucial for ensuring the factory’s efficiency and enabling immediate response.

The calculation of the OEE indicators is as follows:

Availability = timing of a work command for a given line, minus the timing of planned breaks, divided by the duration of a line shift.

Efficiency = number of products made on the line divided by the timing of the work command, minus the time of the planned break, times the standard manufacturing pace.

Quality = number of good products made on the line divided by the number of products made on the line

OEE = availability x efficiency x quality

The OEE indicator includes the management and monitoring of failures and shut-downs at any given moment.

The system allows the operator to report the cause and location of a failure in a simple and convenient manner, thereby monitoring the most common causes and their expected timings.

These data allow us to minimize failures and manage manufacturing in an efficient, more accurate manner.