Earlier this week we tuned in for a fascinating discussion on the topic “Will AI Transform Batteries?”, hosted by Steve LeVine at The Electric, and featuring Tim Holme, CTO of QuantumScape, and Andew Ng, one of today’s leaders in the AI field.
A few points from the conversation stood out, particularly as the new field of Enterprise Battery Intelligence (EBI) is reaching maturity:
- There was broad consensus that the application of AI to the battery space has been hampered by the lack of clean, high-quality data needed to train the machine learning models that will deliver meaningful insights.
- Holme pointed out that in order to effectively apply AI to developing battery materials and improving battery performance, you need end-to-end data capturing materials inputs, build and processing parameters, and performance outputs generated through extensive, data-intensive lab testing and field evaluation. The implication is that most companies lack the infrastructure to collect and analyze this data effectively.
- Holme further offered a rule of thumb that it takes five data engineers to build the tooling and data pipelines needed to empower one data scientist tasked with performing the analysis and optimization.
- Finally, Ng presented another hurdle based on his past experience using AI for predictive maintenance of industrial systems, stating that extensive domain experience is required simply to prepare the data for analysis and extract meaningful features from it to power an AI solution.
In this one conversation, leading experts in the fields of batteries and AI clearly articulated the need for Enterprise Battery Intelligence, particularly as more companies find that they must develop a core competency around the batteries that power their products and business models. Put plainly, if you want to apply AI to batteries, you need EBI.
An EBI solution provides the full set of data pipelines and infrastructure to automatically capture data from across the battery lifecycle — from material and process inputs to detailed battery performance — providing the clean, high-quality data needed to power AI and many other applications. Moreover, a productized EBI solution can have you up and running in as little as a few weeks, versus the years it would take to design and build a robust in-house system. This also replaces the need to directly hire all those data engineers, freeing up resources to focus on the higher-value analysis work that delivers true insight and business impact. And to Ng’s point, an EBI solution has the domain expertise built in, automatically extracting the key performance indicators and other features that enable meaningful application of AI. This factor enables companies to scale their battery and battery analysis programs quickly while accessing a broader set of talent, as battery expertise is currently scarce and in high demand.
It’s precisely these types of applications that motivated us to found Voltaiq and pioneer the field of Enterprise Battery Intelligence, and that continue to inspire our team to innovate and deliver value to the battery ecosystem through our EBI platform. As the industry moves forward, we are excited to see how AI can be leveraged to enhance the full spectrum of value that EBI can deliver: accelerating product development, optimizing battery performance, improving manufacturing ramp time and yield, powering battery finance, improving the customer experience, and preventing fires and recalls.
Indeed, AI has tremendous potential in the battery field. Time is of the essence, however, as winners and losers will be determined by who can build competitive advantage through advanced data analysis and AI the fastest.
So remember — to do AI tomorrow…you need EBI today!