Process Excellence and Value Engineering for Food Manufacturing

Case study: Value Engineering for Chocolate Manufacturing

Process Excellence and Value Engineering for Food Manufacturing

We optimised production by reducing excess usage of raw material components and by controlling the production process to achieve a constant consistency of the final product.

The challenge

Producing chocolate is a surprisingly complex process. The end product should not only taste and look good, it should also adhere to a specific properties, like fineness, uniformity, humidity, rheology properties. To achieve these requirements the production line consists of multiple facets such as mixing, refining, conching, and finishing. Applying the right amount of energy, maintaining the optimal temperatures and duration at each stage are crucial.

Like many other food products, chocolate requires very expensive natural ingredients, which have varying quality depending on the sourcing location and season, suppliers, and storage conditions. Reducing the excess usage of raw material components and controlling the production process to achieve constant consistency of the final product irrespectively of the raw material quality deviations are critical goals.

The solution

Effective analysis tools like DataStories allow inspecting hundreds of product production variables using the historic data. The true value and a holistic look at production and quality deviations can be obtained by combining historian data logged from the production line, commercial data about the batch orders and shipments, the recipe data for different product recipes, as well as external weather and calendar data. When the data from various sources - SAP, ERP, sensor time series data were combined - it gave more than 350 variables and a comparable number of set points. Joined work with the process and rheology experts allowed us to better validate the data and scope the requirements on modelling targets. As a result, reliable models connecting the raw material usage, online process adjustments and output performance targets were generated and validated by process experts

"Effective analysis tools like DataStories allow inspecting hundreds of product production variables using the historic data."

The impact we created

This proof of value project had a clear ROI on the formulated targets and the project also contributed to the future process excellence initiative. It confirmed the hypotheses of process engineers on followup line improvements. It also generated new insights in optimal online recipe improvement strategies.

Interested to see how we can help you improve your manufacturing processes?

Contact us

Talk to us about how you can turn your data into a system to deliver success

Our core expertise is in business-driven applications of predictive analytics and data science to solve complex business challenges which directly impact the bottom line.