Improving Chemical Production Quality and Yield

by Minimizing Process Inefficiencies

About the company

The manufacturer, founded in 1978, has 17 production sites spread across Europe.

What started as a business that delivers construction projects for industrial manufacturers, turned into a worldwide chemical manufacturing business, employing over 3,500 employees worldwide.

Industry: Chemicals 

Employees: More than 3,500 worldwide

Yearly Turnover: Over €1.4B

 The Challenge

The manufacturer turned to Seebo to solve a number of process inefficiencies that were impacting production yield and quality. Process inefficiencies included the formation of undesired side products, and losses during separation of the desired products from the reaction mixture. They were also looking to prevent future inefficiencies from happening and affecting the production line.

The Solution

The manufacturer decided to invest in Industry 4.0 and production optimization and looked for a solution with these capabilities:

  • Combine their manufacturing expertise into data analytics and machine learning
  • Provide operational teams with simple and accurate insights regarding process inefficiencies
  • Deliver predictions on future process inefficiencies
  • Increase yield and improve quality

Results

+€850K/year

Higher sales price

+450K/year

Increased yield

+400K/year

Increased throughput

Together with the customer’s process engineers, the Seebo manufacturing excellence team modeled the customer’s production line using the Seebo  Digital Twin Modeler, together with the IT (SAP) and OT (OSI PI historian) data sources – generating a specific process-based data schema of the production line. This modeling phase was performed over the duration of 3 weeks, and the integration to the specific data sources took a further 2 weeks.

Seebo then configured the solution dashboards to present predictive insights alongside the digital twin model – providing the process engineers with the complete and visual context of the predictive insights. The customization process lasted just 2 weeks.

Access full case study