Offshore operators are increasingly looking to drones as a practical tool for safer, faster and more cost-effective inspection of platforms, flare structures, tanks, pipelines and other hard-to-access infrastructure. Yet the real promise of drone technology offshore goes well beyond replacing rope access or reducing helicopter trips. It lies in enabling high-quality inspection data to be captured in environments where pilots may be operating remotely, well beyond visual line of sight and where inspection missions must be executed with precision in demanding weather, marine and communications conditions.

This presentation examines the technical and operational challenges of offshore drone inspection, with particular focus on remote piloting out of line of sight. It will explore the issues of communications reliability, latency, situational awareness, navigation, safety case development and operational control when the pilot is not physically close to the asset. It will also discuss how drones are becoming more than flying cameras, carrying payloads such as light detection and ranging (LiDAR), thermal imaging, optical zoom, gas sensing and photogrammetry systems to produce richer inspection datasets and support digital twin development, defect analysis and better maintenance planning.

Manufacturing plants today have more and more devices than ever before. Programmable Logic Controllers (PLCs), sensors, historians, laboratory information management system (LIMS) and manufacturing execution systems (MESs) are collecting vast volumes of operational data every second. Yet for many manufacturers, the ability to actually use that data to answer basic questions about performance, quality and reliability remains frustratingly out of reach.

Pat examines why automation data connectivity continues to be one of the most persistent and underappreciated challenges in modern manufacturing. Despite decades of investment in automation technology, most plants contain incompatible protocols, isolated data silos and fragile point-to-point integrations. The result is that critical operational insight is locked away in systems that were never designed to talk to each other.

Drawing on real-world experience in manufacturing environments, this session will explore the structural and organizational challenges that create and sustain the connectivity problem, including protocol fragmentation and operational technology (OT)/information technology (IT) cultural divides, siloed data and a lack of standard data models, as well as the cost of inaction and the organizational inertia that keeps broken integrations in place long after their sell-by date.

Attendees will leave with a clear-eyed understanding of where and why connectivity breaks down, a framework for assessing the cost of inaction in their own facilities and practical principles for building a more connected, context-rich data infrastructure without ripping out the systems they already have. This session is relevant to automation engineers, operations leaders and anyone responsible for making manufacturing data work harder for their organization.

As global manufacturing transitions from reactive maintenance models to data-driven proactive strategies, traditional hardware-centric monitoring is increasingly hindered by high costs, harsh operational environments and the inherent physical limitations of sensors. This presentation explores the strategic implementation of soft sensors—software-based mathematical proxies that infer high-fidelity measurements from existing programmable logic controller (PLC) and supervisory control and data acquisition (SCADA) data streams.

Industrial automation engineers will gain deep technical insight into the engineering foundations of virtual measurement, including advanced signal noise filtering, dimensionality reduction techniques such as principal component analysis (PCA) and the application of temporal artificial intelligence (AI) logic like bi-directional long short-term memory (BiLSTM) networks. We will detail the validation metrics [root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE)] essential for ensuring mission-critical reliability in high-precision fluid control and assembly throughput.

For business leaders, the session provides a data-backed case for managerial return on investment (ROI), demonstrating documented direct hardware savings of over $110,000 per plant rollout and industrial throughput gains that reduce cycle times by up to 55%. Through real-world case studies in fluid dispensing and packaging throughput, we demonstrate how the early detection of "micro-anomalies" prevents catastrophic emergency shutdowns, stabilizing revenue streams and eliminating defective product scrap.

Attendees will leave with a three-step implementation roadmap – from asset-specific piloting to facility-wide standardization – positioning soft sensors not just as a technical upgrade, but as an essential reliability standard for the future of manufacturing intelligence.