Industrial Manufacturing to Deploy Over 15 Million AI‑Enabled Devices by 2024, ABI Research Forecasts

ABI Research’s latest report projects a surge in artificial intelligence adoption within the industrial manufacturing sector, estimating that the total count of AI‑enabled devices will surpass 15.4 million by 2024—a compound annual growth rate of 64.8% from 2019 to 2024.

"AI in industrial manufacturing is a story of edge implementation," says Lian Jye Su, principal analyst at ABI Research. "Because manufacturers prefer to keep their data on premises rather than in public clouds, almost all AI training and inference workloads occur at the edge—on devices, gateways, and on‑premise servers."

To support this shift, AI chipset manufacturers and server vendors have introduced dedicated AI‑enabled servers specifically engineered for industrial environments. Consequently, an increasing share of industrial infrastructure is now equipped with AI software or specialized chipsets to perform real‑time inference.

Despite the abundance of data and available solutions, AI deployment in manufacturing has not been as seamless as anticipated. The most commercially mature applications remain predictive maintenance and equipment monitoring, with an installed base projected to reach 9.8 million and 6.7 million units respectively by 2024.

Many of these devices are multipurpose, thanks to advances in AI chipsets. Key startups such as Uptake, SparkCognition, FogHorn and Falkonry are offering both cloud‑ and edge‑based solutions that monitor overall asset performance and process flows.

Defect inspection is emerging as another high‑growth use case. The installed base for this application is expected to climb from 300,000 in 2019 to over 3.7 million by 2024, especially in electronics and semiconductor manufacturing where partners are collaborating with chipset vendors and software providers to develop AI‑driven machine vision for component‑level defect detection.

Traditional machine‑vision technology remains popular in manufacturing for its repeatability, reliability and stability. However, deep‑learning algorithms extend these capabilities, enabling the detection of unexpected abnormalities, uncovering new insights, and surpassing the limits of conventional approaches.

Manufacturers now face intense competition in building and training in‑house data‑science teams for AI implementation. Most AI professionals gravitate toward large web‑scale players or nimble startups, making talent acquisition a significant challenge for industrial firms.

"Given this landscape, the most viable path is partnership," Su notes. "Collaborating with cloud providers, pure‑play AI startups, system integrators, chipset and server manufacturers, and connectivity providers is essential to navigate the diverse AI use cases."

For more information, visit www.abiresearch.com.