Hello readers!

This week we’re looking the potential for IoT applications in the shipbuilding industry, best practices for using LLMs with the IIoT, drones as an IoT-enabled construction inspection tool, and more!

Maritime shipping carries more than 80% of global trade, yet the industry building those vessels has been remarkably slow to modernize. A new review paper from researchers at Shanghai Polytechnic University argues that IoT-driven digital twins represent the clearest path forward. The core idea is straightforward: embed sensors throughout a shipbuilding workshop, stream that data in real time into a live virtual replica of the facility, run simulations against it, and send corrective instructions back to the physical equipment. It's a closed loop — sense, model, decide, act — and IoT is the connective tissue that makes it work.

The paper uses a steel-cutting workshop as its case study. Hall-effect sensors measure motor current, infrared thermometers track cutting head temperature, laser rangefinders gauge plate thickness, and particulate detectors monitor air quality. That data moves through a hybrid protocol stack — MQTT for fast, lightweight sensor bursts at the edge, OPC UA for structured machine-to-machine communication — into tiered storage, and ultimately into a physics-capable virtual model built in Unity 3D or Unreal Engine. When the simulation diverges from what sensors are reporting, a decision module generates corrective commands and pushes them back to the workshop floor.

One of the paper's most instructive examples is plasma cutting nozzle lifecycle management. Rather than replacing nozzles on a fixed schedule or waiting for cut quality to visibly degrade, the system tracks orifice diameter over time, monitors gas pressure variance, and feeds these signals into an LSTM neural network that predicts wear progression across future cutting cycles. When predicted expansion crosses a threshold, the system flags the nozzle for replacement automatically — before quality is affected, and not a moment sooner than necessary. It's a useful illustration of what genuinely predictive maintenance looks like when built on a well-designed IoT foundation.

The paper is candid about what's still in the way. Protocol fragmentation is the foundational problem: legacy equipment from dozens of manufacturers uses incompatible standards and proprietary data formats, making unified integration deeply difficult. Security concerns grow as IT and OT systems converge around sensitive shipbuilding IP. Real-time performance is harder to guarantee than architects typically assume, requiring 5G, Time-Sensitive Networking, and careful edge-cloud task partitioning. And cross-workshop coordination — getting cutting, welding, assembly, and painting to share data coherently — remains as much an organizational challenge as a technical one.

Several leading shipbuilders are already executing pieces of this. Hyundai Heavy Industries runs a virtual sea trial system that simulates LNG carrier conditions digitally, cutting trial duration and cost by up to 30%. Meyer Werft operates a digital shipyard where engineers design hull, piping, and electrical systems in VR before fabrication begins. The gap between these examples and industry-wide adoption remains wide, but the framework is clear, the enabling technologies are mature enough to deploy, and the economic case — in an industry where a single vessel represents hundreds of millions of dollars and years of production time — is hard to argue against.

The narrative in the boardroom has shifted. Six months ago, the conversation was about "Generative AI" and how it could summarize emails or write code. Today, the buzzword is "Agentic AI." We are no longer asking models to just talk; we are asking them to act.

Read more

Gmail is free, but you pay with your data. Proton Mail is different.

We don’t scan your messages. We don’t sell your behavior. We don’t follow you across the internet.

Proton Mail gives you full-featured, private email without surveillance or creepy profiling. It’s email that respects your time, your attention, and your boundaries.

Email doesn’t have to cost your privacy.

Ditch the Gmail data grab

In this episode of the IoT For All Podcast, Kevin Dewald, Founder at The California Open Source Company, joins Ryan Chacon to discuss the current state of Bluetooth. The conversation covers why Bluetooth in IoT is so mobile-centric, the promise of desktop as a Bluetooth platform, why dongles pose challenges, the economics of open source, the complexities of the Bluetooth protocol, and the need for a shift in how IoT products are developed.

YouTube • Spotify • Apple • YT Music • Amazon

Interested in becoming an IoT For All Partner? Reach out here!