塑造未来的人工智能工厂

SHAPING THE Now in its fourth year, the Manufacturing Leadership Council’s “Future of Manufacturing Project” continuesto focus on its main goal: to enable manufacturers to envision what manufacturing might look like in thefuture. Achieving this goal entails thinking broadly about the intersection of the technology, organization and This research paper is part of the Project’s focus this year on what MLC calls “AI-Centric Factories of theFuture.” It’s a focus on how artificial intelligence will play an increasingly vital role in tomorrow’s factories.Manufacturers will leverage AI across all aspects of the factory—from the shop floor to the boardroom, from The survey report follows an MLC event held in December 2024 with the same title. The one-day conferenceexplored three key aspects of an AI-Centric Future: realization of value through AI-enabled innovationand decision-making; defining the human relationship around the relationship of AI to ecosystems, lean In the second half of 2025, the MLC will continue to focus on AI-centricity with webinars, case studies andadditional research.Please check the MLC website for updates and further information. The MLC team thanks its Future of Manufacturing Project partners—Accenture; Infor; Invisible AI; Kalypso, aRockwell Automation Business; NTT DATA and West Monroe—for their support, their invaluable feedback on EXECUTIVE SUMMARY The future of manufacturing is not waiting on AI; it is beingbuilt with it. The Future of Manufacturing Project surveyreveals that while AI adoption in manufacturing is still in theearly stages, it holds immense potential for transforming theindustry. As manufacturers look ahead, AI is poised to playa pivotal role in shaping operations, with challenges and For the manufacturers willing to commit, the payoff couldbe transformative. The data reveals a growing consensus:68% of respondents say that AI will be foundational to Even among companies still early in their AI journey, there islittle doubt about this technology’s potential. In fact, every type of AI technology is expected to see increased adoption in the next two years—with edge AI, agentic AIand physical AI showing the greatest gains in terms of expected use. As they ramp up investments, manufacturers will have to deal with many issues, including how they manageand govern the technology. Currently, 35% of manufacturers have a formalized AI governance strategy atthe enterprise level, and another 35% are in the process of developing one. This signals strong momentumin AI adoption. AI’s presence is still limited in operations, however. Only a reported 18% of manufacturers Despite this promising outlook, manufacturers face ongoing barriers to using and scaling AI. Data quality andaccessibility remain the top challenges, with 65% of manufacturers reporting they lack the right data for AIapplications and 62% citing data that is unstructured or poorly formatted. Overcoming these issues will be The imperative for AI is clear, but so are the hurdles. Manufacturers must invest in data infrastructure, talentand scalable systems to ensure they do not fall behind in this rapidly changing landscape. The path forward isn’t plug-and-play. In addition to the speed bumps already cited, skill gaps, cultural frictionand legacy systems are hurdles to adoption and usage. What will separate the leaders from the laggards is the ability to build momentum. It is less “go big or gohome” and more “crawl, walk, run.” The future belongs to manufacturers that can fail fast, scale what works „Leadership involvement in AI is still developing, with nearly one-third of respondents unaware of who isleading AI governance within their organizations. „Manufacturers are more likely to partner with outside organizations to implement AI. In fact, only 25% saythey are building their own AI solutions. „Sixty-three percent of those with defined targets for AI’s value are achieving or exceeding their targets. AI GOVERNANCE AND STRATEGY When the current age of industrial AI was at its beginning, early manufacturing use cases included predictiveequipment maintenance, optimized quality inspections and more accurate demand forecasting. As these Generative AI is serving as a catalyst for increased interest from industrial organizations, but adoption isrestrained. While many manufacturers are seeking the best place to start operational deployment, others aregetting the underpinnings in place for broader AI adoption. These preparations include strategic roadmapsfor AI usage, integrating AI strategy with a broader data strategy and establishing a governance structure for In the not-so-distant future, developing a robust AI strategy may become as fundamental to business successas having a sound financial plan or a comprehensive marketing approach. As mentioned in the ExecutiveSummary, 35% of respondents said that their organization has an enterprise-wide AI governance strategy, Accounting only for manufacturing opera