Cutting Machine Production Energy Consumption with IoT Data and AI

Approach Why optimize existing machinery? Production machines are often not adapted to each application by the manufacturer. Therefore, there is a high potential for reducing energy consumption and ongoing production costs: Adastra’s solution allows to have more energy-efficient machines without costly hardware upgrades.

Moreover, while production machine optimization is usually carried out manually by domain experts, this data-driven solution allows deeper insights into the processes to be obtained and made accessible to a larger audience as well as to automate the optimization process.

Activities Start small, grow big:

• Single out the production machine with the highest optimization potential
• Collect the machine and quality data
• Set up the architecture to extract the data and push them in AWS

Run a discovery analysis on the machine data:

• Identify the machine’s process responsible for the highest energy consumption
• Look for correlations between the machine and product’s quality data

Develop the automated optimization strategy:

• Build the optimization engine at the core of the solution
• Implement robust statistical checks for quality monitoring

Develop ad hoc dashboards:

• For product quality and machine status monitoring
• Showcase the results of the optimization engine and propose the settings to be used
• Interactive: the settings proposed by the solution can be adjusted by domain experts before they are automatically used

Deliverables Automated and scalable solution fully deployed in AWS

• Integration of IoT data in AWS

A robust strategy to adjust production machine’s settings to reduce energy consumption

• Guarantee the safety of the machine
• Provide optimized parameters for further energy savings

Dashboards with KPIs for managers and technical information for specialists

• Easily monitor the data acquisition process
• Detect product quality issues

Outcomes The iterative approach enables the company to lower the production costs during series production without compromising the throughput, guarantying the safety of the machines and in compliance with production targets. This proposed solution is adaptable and easily applied to multiple machines and production sites across the client's operations.