A smaller manufacturer can start using AI by choosing one recurring operational problem and testing an approved tool against a measurable baseline. Existing software and process data may support an initial experiment without a major platform purchase. Keep the scope focused, involve employees who understand the work, and validate the output before relying on it in production.

Live Connected Manufacturing discussions reinforced that internal ownership matters as much as technology selection. Pareshkumar Hotchandni, Global Staff Engineer, Digital Manufacturing at Smith+Nephew, explained, “Operation has the business ownership, and they are responsible for user adoption and outcome.” The distinction is useful for smaller teams: someone must own the operational result, beyond simply building or introducing the application.

Another practitioner example showed how AI-assisted development can address a specific production constraint without specialized optimization software. At a boat manufacturer, an industrial engineer described using an AI coding tool to develop scheduling calculations around floor space, staffing, and work-in-process limits. They reported more balanced workloads and improved use of production space.

Choose a task with clear inputs and a result your team can check, such as developing a proposed schedule from known capacity constraints. Set a spending limit, name an operational owner, and account for employee time, licensing, and review. Use company-approved tools and data, compare the result with current performance, and expand only after the approach demonstrates value.

Join the upcoming Connected Manufacturing Forum to learn how manufacturing practitioners turn focused AI experiments into practical improvements with the resources they have.