3 Ways to Execute a Successful Industrial AI Strategy
Today’s industrial marketplace depends on industrial AI, which is being embraced by IT and operations decision makers in process engineering companies as a significant driver in their digital transformation strategies.
But getting the most out of industrial AI depends on successfully scaling an industrial AI strategy across a business organization. Here are critical steps to help companies reach these goals.
- Build an Enterprise-Level Data Management Strategy
For too long, process engineering businesses have been stuck in a mindset of mass data collection. The notion of “more data is better data” has been the default approach to data management strategies, but it is erroneous. Because of this mindset, the industrial sector has been aggregating massive volumes of data for years that sit unused, unoptimized, unstructured and functionally useless.
Creating a strategy that drives the most value out of industrial AI means undertaking a fundamental shift in an organization’s data strategy – from mass data collection to strategic data management. That means building an enterprise-wide strategy that focuses on managing, integrating and mobilizing disparate, unstructured data sets, then making that data actionable across the organization will finally allow all teams to draw from and leverage that large pool of data to feed into their industrial AI applications.
- Reduce Friction Between Functional, Data and Technology Silos
Part of building an enterprise-level data strategy is also reducing or even eliminating the friction caused by teams that remain divided through the presence of data siloes. When teams use and keep their data, domain expertise and storage technologies separately, it adds layers of friction that are worsened by decades of mass data collection. This, unfortunately, heaves industrial data trapped in silos and data swamps.
Those problems only compound when:
- Data sets that may be relevant to multiple teams, but live in a single team’s database, offering little to no visibility for the rest of the organization. This also forces other teams to either tediously chase down the relevant information from different corners of the business, or redundantly double up collection of the same data for their own silos.
- Data lakes that are meant to be transient rest stops for data passing through the organization become permanent data swamps, where information lives in an unstructured format that is difficult, if not impossible, to search for relevant queries.
- Data can be left in multiple formatting and security stages, so that no one person in the organization can ever access parts of the data that is stored in different parts of the business.
One optimal way to reduce or eliminate this friction is with the deployment of a next-generation data historian. Data historians help democratize data access and insights by putting all industrial data into a universal, standardized and secure formatting stage. Rather than data format and structure being decided by individual team and technology silos, all data across the organization is stored in the same formats so that all users have equal access to it – and an equal ability to leverage that data for creating new value. This universal formatting is a core component of an industrial AI strategy, effectively eliminating data silos and ensuring that industrial data access is not so dependent on individual technologies or expertise.
- Make industrial AI skills a focus of recruiting and employee education
Businesses of all stripes are feeling the pinch of the “Great Resignation,” but this labor shortage is having a particularly major impact on the industrial sector. Our industry has been in the middle of a generational change even before COVID-19 hit, with veteran employees who have worked at the same plants for decades retiring and being replaced with younger recruits who do not have that same level of operational or institutional expertise.
Industrial organizations can stop this brain drain by providing an industrial AI infrastructure for employees to leverage. This has two unique benefits for employee retention and training:
- It ensures that users are given the tools they need to succeed in their jobs. Even if they do not have years of experience just yet, industrial AI can democratize historical data access and insights well enough to fill that gap, empowering younger employees to tackle their roles just as well as their predecessors.
- Making industrial AI a centerpiece of the user experience in turn also works as a recruiting tool. When employees can see the tools they will be given to set them up for success, that has greater power to draw people to work for an organization compared to one that throws new recruits into the deep end with yesterday’s technologies.
To survive and thrive in today’s market, industrial organizations need industrial AI at the core of their operations and workflows; it needs to drive their digital transformations. Creating a strategy uniquely attuned to executing fit-for-purpose industrial AI applications is the only way to maximize the value that industrial AI has to offer.
That may be easier said than done, but building an enterprise-level data management strategy, reducing friction between technology, people and silos, and making industrial AI skills a core part of the employee user experience are all critical ways for reducing that learning curve and making an industrial AI strategy a reality.
About the Author
Bill Scudder is a senior vice president and general manager of AIoT solutions at industrial optimization software vendor AspenTech. Scudder joined AspenTech in 2015, where he previously served as the senior vice president and CIO and remains responsible for the company’s IT organization. He previously worked as a vice president and CIO for Sonus Networks, where in 2011 he was named CIO of the Year by the Massachusetts Technology Leadership Council. Scudder holds a bachelor’s degree in mechanical engineering from Rensselaer Polytechnic Institute, and an MBA from Boston University.