How BMW Group has embraced AI for positive use cases and to improve sustainability | AWS re:Invent

Artificial intelligence (AI) is helping many distinct industries and is getting a specifically potent impression in the automotive business. Among the the most remarkable use conditions is for entirely autonomous autos, but that’s not the only area exactly where AI is possessing an effects. For case in point, Microsoft and Mercedes-Benz are working alongside one another to improve automobile manufacturing performance. 

At the AWS re:Invent cloud meeting this week, BMW Team outlined the effects that AI has had on its corporation and detailed rising use situations in which AI will produce long run constructive company outcomes.

In a session, Marco Görgmaier, GM, data transformation and synthetic intelligence, BMW Group, stated that his group had constructed up a library of 1000’s of info assets throughout the organization that can be reused for analysis and AI. Due to the fact 2019, he reported his crew has been ready to deliver additional than 800 use situations that have yielded above $1 billion in U.S. dollar value. The use conditions span analysis and growth, logistics, sales, quality and provider network.

“The eyesight and the mission of our team is to push and scale business benefit creation by means of the usage of AI across our price chain,” Görgmaier reported.

BMW driving towards a sustainable potential with some assistance from AI

An rising region the place BMW is now investing assets is in serving to to strengthen sustainability. 

Görgmaier commented that 60% of the world’s population life in towns and city places and that’s also the place 70% of greenhouse gas emissions are produced. What BMW is now making an attempt to do is support city planners in solving difficulties to support reduce emissions.

BMW is by now assisting with device studying models that are able to forecast how visitors polices can probably support to decrease both traffic and gasoline emissions. ML types are also used to aid recognize the place there isn’t nonetheless ample electrical car or truck charging infrastructure. Görgmaier said that a lack of charging infrastructure stops persons from switching to an electric powered automobile, which in change has an affect on sustainability.

There is also a BMW ML effort to enable forecast the impression of parking area availability and pricing on driving patterns. Those people patterns involve commuting routes and traffic, which also will have an influence on emissions.

Driving geospatial data with Amazon SageMaker

Görgmaier said that several of the urban sustainability troubles that BMW is trying to support solve can benefit from geospatial facts. Which is exactly where BMW is setting up to make use of new geospatial capabilities in the Amazon SageMaker ML instrument suite that were just publicly exposed this week.

One particular spot exactly where BMW is wanting to reward from geospatial ML is for encouraging to predict when an corporation with a fleet of vehicles will be in a position to transition to electric powered automobiles.

“We set up the objective to educate equipment discovering styles to understand correlations in between engine form and driving profiles,” he reported. “The rationale guiding that was if these a correlation would exist, then the model could find out to predict the affinity of certain motorists for an electric powered car based on their profiles.”

As BMW was operating with entirely anonymized information at a fleet amount, it experienced to use GPS traces and geospatial data to make the correlations.

“At the stop of the coaching, the product was able of predicting how probable it was for unique fleets to change to EV with an accuracy of far more than 80%,” Görgmaier claimed.