Senior Vice President and General Manager, Artificial Intelligence of Things (AIoT) Solutions at AspenTech.  

getty

What is your industrial AI readiness?

That’s a question that’s top-of-mind for many industrial executives lately — and simultaneously one that has not taken on enough importance for many others. While AI, machine learning and other means of automation have swept through industries, including the industrial sector, in recent years, AI still too often gets treated as an add-on technology. But AI isn’t something to be tacked onto an existing framework; it has to be treated as the strategy itself. This is especially true for industrial AI. With its capabilities for driving autonomous production, operational agility and remote collaboration, industrial AI can unlock tremendous value for industrial organizations — but only if it’s properly interwoven into how the organization operates.

Getting started with industrial AI isn’t about the “how,” but rather the “where.” Where do you start when building a business and technical framework around it?

It starts with these five elements of industrial AI readiness:

1. An Industrial AI Action Plan

A recent Accenture study found that approximately 69% of industrial organization executives know how to kick-start an AI strategy but struggle with how to scale it across their enterprise. The real starting point, then, has to be identifying business problems, corporative objectives and strategic goals — and then working backward on how industrial AI addresses, enhances and meets them.

The only way to effectively integrate industrial AI as part of a larger digital transformation is to map the AI solution’s capabilities to selective, fit-for-purpose applications that are geared at driving specific business value and meeting KPIs. More targeted applications are how you can start this industrial AI journey, but these applications can’t be treated as one-offs — rather, they need to work together as a part of a holistic AI strategy.

2. Building An Enterprise-Wide Industrial Data Strategy

Just as AI needs data, industrial AI needs industrial data. That’s a no-brainer. But the quantity and quality of data is a different matter entirely. How much data is too much (or too little)? How clean does that data need to be? How frequently and quickly does that data need to be fed into and processed by the AI? These are all questions to be answered by an enterprise-wide strategy for managing, integrating and mobilizing industrial data.

The main challenge facing industrial organizations today isn’t a lack of data; it’s a lack of actionable data that can be put to real, value-driving use. Making data useful and valuable relies on leveraging AI solutions to process and mobilize raw data relevant to specific use cases, rather than forcing users to wade through unstructured data swamps to find the information they need. An enterprise-wide industrial strategy ensures this approach is scaled across the whole organization so that all teams are getting consistent results out of both their aggregated data and the AI solutions deployed to make that data actionable.

3. Future-Proofed Industrial AI Infrastructure

Eliminating silos across industrial data environments and fostering collaboration across all parts of the organization — from tools, workflows, developers and data scientists to CloudOps, DevOps and MLOps — is essential to building a future-proofed, industrial AI infrastructure.

What does that look like? Faster time to market, flexible and scalable AI applications and harmonized AI model life cycles across all use cases — all amounting to an across-the-board futureproofing so that performance and results are consistent no matter the application or user. More than that, modernizing a company’s data infrastructure around AI offers a new kind of competitive advantage — something noted by a plurality of companies in Deloitte’s “State of AI in the Enterprise” report.

4. An AI-Centric Mindset — With The Skills And Training To Enable It

This need was reflected in a McKinsey survey, which found that over 80% of companies plan to retrain some employees in the next three years because of AI-related changes. That will require investing in AI-geared business and technical training programs, supporting continuous education and ensuring that workers have a seat at the table when developing industrial AI process strategies.

5. Transparency And Ethics

Industrial AI readiness isn’t solely about a singular process or hardware and software — trust and transparency have to be baked into the DNA of any industrial AI strategy for it to truly succeed. No one is comfortable with lock-in, black-box AI strategies. AI in any industry must be used safely, responsibly and ethically. That means transparency and clear, open channels of communication across all workers and stakeholders — to ensure that everyone is aligned on the goals, uses and value of industrial AI.

This readiness checklist should be taken as more of a guidepost, rather than a one-size-fits-all prescription for making your organization AI-ready on day one. But it should hopefully get any industrial organization off on the right foot on how to best approach the technical, organizational and cultural requirements for integrating industrial AI into your everyday workflows and processes — and setting yourself up for success with it.


Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?