How to Blend Standardization with Customization?

A real-life case example from Denis Gauder's presentation at ManuChem Conference 2021 in Berlin, Germany

The cloud concepts and IIoT can be particularly challenging for a specialty chemicals company with too many small sites all running their unique process.

How do you make significant optimizations in performance tracking while adapting to market prices "live"? And how can the optimization be specific when there are no one-size-fits-all solutions?

Centralizing the process data

We began by transferring the data from each site into one dedicated location via a gateway to the cloud. This centralization allowed us to run Python algorithms to find optimization solutions. Finally, visualization of the solutions was achieved by utilizing generic "do-it-yourself" tools based on BI.

The benefits of this centralization are multifold

The data management solution was extended to the entire company. This helped manage the scale while offering dedicated asset solutions, such as plug-and-play approaches to algorithms, including central governance. This, in turn, will give us a combination of standardization and customization, with enough customization to guarantee adoption by the end-users.

Not a particularly time-consuming approach

In this case, once the data management consolidation was in place, Python algorithms and user interfaces development and deployment required about four weeks. In a month, the chemicals company, which previously had too many small sites, had a good overview of their unique process in a new, standardized way.

The optimization algorithms delivered up to a 10% reduction in energy use in some plants and up to 5% increased profit margin in others by decreasing raw materials consumption.