The Gap Between Pumps and Sensors
Political instability in several major oil-producing regions is pushing operators to lean harder on non-OPEC reserves, and much of that oil sits in fields already in their tail-end phase. Getting the most out of ageing wells matters more than it did a few years ago, and that shifts where engineering attention goes: toward the pumps and control systems doing the daily work of keeping mature fields productive.
For years, that work was split in two. Pump manufacturers optimised the machinery. Instrumentation manufacturers optimised the sensors. The two sides rarely spoke the same design language, and the gap between them is exactly where efficiency was being lost.
AI From the Design Stage
Jorg Eitler, Head of Global Business Field Oil & Gas Upstream at NETZSCH Pumpen & Systeme, presents on closing that gap at the Oil & Gas Automation and Digitalisation Congress (Automa) 2026, taking place in Amsterdam this October. His presentation looks at how artificial intelligence is moving into pump system design from the earliest stage, simulating well behaviour against real infrastructure constraints, well geometry, casing size and fluid composition that shifts over a well’s lifetime, rather than relying on standard assumptions.
From Design to Daily Operation
The payoff shows up twice: once in the design phase, where better simulation means a system sized correctly the first time, and again in daily operation, where the pump needs to continuously reconcile its own settings against actual conditions like dynamic fluid levels. NETZSCH already builds downhole sensor systems that track pressure, temperature and vibration, feeding data that lets frequency-controlled pumps adjust in real time rather than run on fixed settings. Eitler’s presentation extends that logic further: simulation software that learns from operating history, and monitoring that catches a developing fault early enough to plan a replacement and order the right part, potentially a different one than was originally installed, before failure forces an unplanned shutdown.
The through-line is that every change made to a producing system becomes data for the next one, so a conveying system already in the field keeps getting sharper rather than staying frozen at its original design. For operators managing mature assets, that is a different proposition than buying a better pump. It’s a pump system that keeps working with the operator to extract more from a well that has already given up its easiest oil.
Automa 2026 runs on 5-6 October in Amsterdam, bringing together operators, EPCs and technology providers in a closed-door format built around structured sessions. Eitler’s presentation is one piece of a wider programme covering AI, digital twins, autonomous operations and automation across the upstream, midstream and downstream value chain.
Details on the full Automa 2026 business programme can be requested through the Congress website: https://sh.bgs.group/4xv
