Arcelik uses AWS to transform appliance manufacturing into a data-driven organisation - The EE

Arcelik uses AWS to transform appliance manufacturing into a data-driven organisation

Andy Isherwood of AWS EMEA

Amazon Web Services, Inc. (AWS), an company, reports that Arcelik, one of the manufacturers of household appliances, has selected AWS as its preferred cloud provider for machine learning and analytics across its entire operations.

Arcelik, which owns 12 global appliance brands including Beko and Grundig, is relying on the breadth and depth of AWS services to transform itself into a data-driven organisation and provide enhanced customer experiences, innovate new services, and lower costs across its operations in nearly 150 countries.

By adopting AWS analytics, compute, database services, Internet of Things (IoT), machine learning, and storage, Arcelik is gaining deeper visibility across its organisation, optimising processes related to production and quality control, and driving efficiency across all business segments, from customer care to manufacturing.

Arcelik is establishing a company-wide analytics and machine learning program, powered by AWS. This will aggregate and examine data from its supply chain, maintenance, manufacturing, quality control, and sales operations to achieve predictive insights for evaluating customer satisfaction ratings, forecasting growth, and managing inventory.

Arcelik began this programme by building a data lake with Amazon Simple Storage Service (Amazon S3) using AWS Lake Formation, for quickly ingesting, cataloguing, cleaning, and securing data, and AWS Glue, for preparing and loading data for analytics.

All of Arcelik’s business units have access to this data lake, which feeds into new machine learning solutions powered by Amazon SageMaker – AWS’s service that enables data scientists and developers to build, train, and deploy machine learning models quickly – to uncover patterns and identify process improvements.

The company is leveraging machine learning to examine anonymised data from millions of call centre inquiries and service requests to help its field technicians make better inventory decisions and predict the tools and parts they will need for maintenance appointments.

The company also uses Amazon SageMaker BlazingText, an algorithm providing highly optimised implementations of word2vec and text classification, to power its ‘Quality Intelligence’ solution. This service reviews notes from service technicians and identifies emerging trends in service requests, for Arcelik’s customer care and quality control teams, to help them proactively address customer needs.

Expanding its use of AWS analytics, IoT, and machine learning services to its factories and product lines, Arcelik is building on a project to develop cloud-connected appliances. The company currently collects data from more than 1 million deployed smart appliances, using AWS IoT Core to easily and securely connect these appliances to the cloud and other devices. It also uses Amazon Elasticsearch Service to gain insights from this machine-generated data, enabling Arcelik to analyse consumer usage patterns and diagnose potential performance issues remotely.

Arcelik is rolling out a number of solutions on its manufacturing lines, using machine learning models built with Amazon SageMaker, to identify process improvements and enhance quality control. For example, Arcelik’s ‘Smart Sampling for Auditing Purposes’ solution examines production and sales data to predict which finished appliances are best suited for quality control testing, enabling Arcelik to automate this step in the auditing process.

In addition, Arcelik developed a ‘Smart Component Matching’ solution on AWS, enabling the company to examine real-time and historical production line data with service data to help the factory’s automatic-production system identify the product parts that are compatible with other components, resulting in better production quality.

Arcelik is currently implementing these solutions at its award-winning washing machine factory in Romania, which was recently included in the World Economic Forum’s ‘Global Lighthouse Network’ – an initiative recognising some of the world’s most innovative and advanced manufacturing facilities. The company plans to work with AWS to roll out these solutions to additional Arcelik factories around the world.

“AWS is a strategic resource in Arcelik’s digital transformation and global growth, by keeping us agile and solidifying our position as an industry pioneer through the use of advanced analytics, IoT, and machine learning services,” said Utku Barıs Pazar, Arcelik’s chief strategy and digital officer. “We greatly benefit from the insights delivered from AWS’s vast portfolio of services to guide our business decisions, innovate quickly, and remain responsive to our customers’ needs.

As the uncertainty of the past few months makes clear, it is critical to have an integrated view of our global operations in order to adapt effectively to change. AWS is helping us navigate our business through this unprecedented time and to keep our commitments to our customers.”

“Modern corporations are awash in data that can help them create value for their business and better serve their customers, if they have the tools to harness it,” said Andy Isherwood, managing director of AWS EMEA. “Arcelik is a specialst in the manufacturing industry, leveraging the proven performance of the cloud to transform itself into a data-driven organisation, on a global scale. We look forward to continuing to support their efforts with the most comprehensive set of analytics and machine learning services available today, to drive innovation, and deliver improved customer experiences.”

Follow us and Comment on Twitter @TheEE_io

By continuing to use the site, you agree to the use of cookies. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.