Driving Industry 4.0 for major appliances

Improve connected products

Mabe manufactures home appliances, including stoves, refrigerators, washing machines, dryers, water purifiers and more. The company markets its home appliances under its own brand as well as several others, including GE Appliances, in more than 70 countries. Mabe is an early leader in developing connected products that allows its customer service personnel to monitor the health of its appliances in the field. Siemens has been at the center of Mabe’s product development process for years, and Mabe is now working with Siemens to deploy and enhance its connected products strategy.

Their challenge

Mabe gains a wealth of valuable information from its connected products, including its popular refrigerators, which improve customer satisfaction and promote repeat business. Mabe wanted to leverage historical data from the field to gain new insights into how to optimize compressor and air circulation fan operations in its refrigerators to save energy. It also wanted to automatically modulate compressor and fan operation to keep food as fresh as possible based on how often the door is opened and how long it stays open.

The Mabe team understood that consumer interaction could affect its appliances’ automated processes. The team hypothesized that adding models that predict consumer behavior might help optimize the refrigerator’s automatic routines, resulting in better performance and energy savings.

Our solution

Siemens set out to solve the problem by using the Rapidminer® software portfolio’s artificial intelligence (AI) capabilities, specifically those found in Rapidminer AI Studio. Mabe’s dataset contains over a million records, so efficient processing was important. Together, Siemens and Mabe built an AI workflow that extracts, cleans, prepares and transforms their data to predict consumer behavior based on the days of the week.

Rapidminer is part of the Siemens Xcelerator business platform of software, hardware and services.

Siemens has worked with this type of data and these types of workflows before and its team’s experience was key in helping Mabe achieve its goals.

The Siemens people leveraged their expertise in processing time series data, and Mabe provided business insight throughout the process to ensure the analysis would align with the team’s objectives.

Siemens and Mabe collaborated throughout the proof of concept (PoC) and met weekly to exchange feedback. Mabe experimented with the AI workflow using data pulled directly from the field. The solution was created from a workflow that gathers data from different tables in a database, transforms and prepares it, and outputs recommended changes to the automated functions in the product to optimize its efficiency. The team used the resulting workflow to score new data and glean insights into how consumers were using refrigerators and how the products were performing.

Mabe’s engineering team deployed the workflow using Rapidminer AI Hub software. The project took about 60 days. Given the amount of data required and the fact that Rapidminer AI Studio and Rapidminer AI Hub tools were new to the Mabe engineering team, Mabe was delighted with the quick turnaround.

Results

The new workflow helped Mabe predict consumer behavior to optimize their refrigerators’ performance and efficiency, ultimately helping customers keep their food fresher longer.

The Mabe team tested the model against historic data to compare real-world consumer behavior with its predictions. The ability to analyze, model and deploy this type of model with a large, historic dataset enables Mabe to leverage AI as part of its home appliances product development effort, giving their team a better understanding of how big data can break down barriers and unleash innovation.

“With Siemens’ help, we can expand our technology solutions and leverage AI and data analytics tools to enhance our products and improve the entire customer experience,” says Martin Ortega, design leader at Mabe.