Tier 3 Pharmaceutical Supplier
Improving Production Processes with Real-Time Data
One challenge was that a lot of batch corrections needed to be made. The current state resulted in wasted machine time, labor, and materials. There was no tracking of the processes or machines. The recipes were programmed into the PLCs. They needed batch cycle times, the number of corrections made, ingredient selection tracking, and results of the batches (pH levels).
The second challenge was no Enterprise and Plant visibility into production, waste metrics, active labor participation, or material usage. Stakeholders at the Enterprise level also wanted to have real-time data on plant performance.
The third challenge was equipment utilization. There was an existing OEE downtime application, but it only provided 60% of the data needed. The users had to manually enter data to calculate OEE and machine downtime.
The final challenge was developing a digital roadmap. There was a smart manufacturing initiative in place. However, the company stakeholders were conflicted regarding where to start. The IT department wanted an off-the-shelf MES system, which would require 18 months to implement and force the manufacturer to compromise important needed capabilities. The Operations and Engineering departments wanted a non-proprietary, non-monolithic solution. Essentially, they wanted data and improvements faster than 18 months.
A DTMA was performed to review plant processes, departments, data available and data used. We uncovered the biggest areas of opportunity for solving short and long-term challenges. A digital roadmap was defined for the next several projects to get the architecture in place and begin solving challenges. A nonproprietary, scalable, flexible architecture was put in place that can be replicated and scaled affordably.
We used the Unified Namespace architecture. The technologies in the architecture included HiveMQ and HighByte on the OT network with Ignition Edge and KEPServerEX to get machine data.
The Pilot Project was selected to connect to brownfield devices to get machine utilization data, batch data, and store machine data. We connected to the OEE/downtime tool to subscribe to the UNS to get data directly from the machines. Next, we connected Canary Labs and Flow Software for historian data storage and data analysis.
The second project was scheduling manufacturing work orders and tracking the data against those work orders. We replaced the existing OEE/downtime tool with the SepaSoft OEE tool which includes downtime and scheduling, which is then connected to the UNS. This process analyzes production times, start/stop plan vs actual plan, and downtime Pareto charts with reason codes.
Additional projects included importing manufacturing work orders, adding recipe management, batch records, track and trace, and scaling to other lines. Also, we connected the existing CMMS to the UNS to generate maintenance work orders based on the machines’ conditions.
The first two projects were completed and provided valuable data before IT selected an off-the-shelf MES system. The batch cycle times and waste issues were identified and resolved. Our client was able to reduce the annual waste by 14% and reduced machine downtime by 45%.
The solution implemented was flexible, and scalable, with an open technology architecture in place and already modeled and ready for consumption by AI cloud tools or a deeper analysis and ML forecasting.
Check out some of our other recent projects!
Estimated Life Remaining Calculation Reduces Expensive Downtime
Company is a wood products manufacturer creating OSB board with a very large press.
Let’s Start Something new
Get in touch with us. We’re happy to answer all of your questions and point you in the right direction.