Kick-off of the new Special Interest Group AI
19 november 2026
We are pleased to announce the kick-off of NAP's new Special Interest Group AI on Thursday, January 28, 2027, at KH Engineering in Schiedam.
How is AI being applied in industrial asset management? During this first meeting, KH Engineering / Smart-AIS and Tata Steel will share practical examples of AI applications in asset integrity and predictive maintenance. The meeting will also provide an opportunity to discuss and shape the SIG's programme for 2027.
Program
- 15.00h welcome and opening
- 15.15h use-case KH engineering / Smart-AIS
- 16.00h use-case TATA Steel
- 16.45h agenda setting SIG: program 2027. Chair Femke de Jager, SIG lead
- 17.30h closure.
NAP members can nominate participants for the SIG, registration takes place via the NAP bureau info@napnetwerk.nl. Participants will receive further info on address, route description the week before the SIG meeting.
Speakers
- KH Smart – AIS, Martijn van t Wout and Freek Lovink on applying AI for asset integrity solutions.
- TATA Steel, Raghunath Banerjee, on applying AI for predictive maintenance.
KH Engineering: Scaling Visual Inspection with Human-Driven AI
Industrial asset owners face a growing inspection challenge. Ageing installations require increasing attention, while experienced inspection specialists are scarce. Conventional inspection approaches are also constrained by accessibility, work at height, inspection costs and the amount of visual information specialists can realistically review.
In this case study, Smart AIS presents a practical approach to AI-assisted “schouw” inspection, developed around a large-scale Corrosion Under Insulation (CUI) inspection programme. The objective is not to replace the inspector, but to scale inspection capability while keeping specialist knowledge and engineering judgement at the centre.
The approach starts with systematic visual data capture using drones, 360° cameras and other digital imaging techniques. Inspection specialists annotate and classify observed degradation using established criteria, including ISO 4628-3:2024 and World Class Maintenance CUI management principles. Computer-vision models learn from these expert-classified examples to detect potential degradation and support severity classification across thousands of images.
Crucially, AI detection is not treated as an inspection conclusion. Inspectors validate findings, confirm or correct classifications and determine whether they become accepted inspection findings.
The practical impact is already measurable: average review time was reduced from 7.0 to 2.85 minutes per image. Across approximately 3,000 images, review effort decreased from around 350 to 136 hours.
Ultimately, the case explores a practical question: how can we scale scarce inspection expertise using AI, while maintaining standards, traceability and human control?
Martijn van ´t Wout
Martijn focuses on creating practical business value with AI, from improving and scaling industrial inspection to broader applications in asset management.
Freek Lovink
Freek focuses on guiding the practical application of AI in Asset Information Management and industrial inspection, connecting technology, asset data and domain expertise to create scalable solutions for industrial assets.
From Predictive Maintenance to Intelligent Asset Management at Tata Steel Nederland
At Tata Steel Nederland (TSN), artificial intelligence is not a recent experiment. It is the next chapter in a long journey of using data, advanced analytics and engineering expertise to make steelmaking smarter, safer and more efficient.
One of the strongest examples is Predictive Maintenance at the IJmuiden steelworks. In a complex industrial environment, equipment reliability directly influences production stability, cost and operational performance. TSN saw an opportunity: what if equipment could tell us it needed attention before it failed?
This ambition led to the Asset Monitoring & Diagnostics Centre (AMDC), combining equipment-condition data, sensor information, advanced analytics and engineering knowledge. AMDC monitors asset health and uses predictive models to identify potential failures, helping TSN move from reactive and time-based maintenance towards condition-based and predictive maintenance.
But the real success is the combination of data, AI and human expertise. Predictive models detect patterns and emerging anomalies across thousands of signals, while experienced engineers interpret these insights, determine the right intervention and provide feedback to improve the models.
TSN is now building on this foundation, scaling predictive maintenance across IJmuiden and prioritising opportunities with the greatest uptime and cost benefits.
The journey is simple: Monitor → Understand → Predict → Act → Learn.
For TSN, industrial AI creates tangible value not by replacing engineering expertise, but by augmenting it, enabling earlier detection, better decisions and more reliable assets.
Raghunath Banerjee
Raghunath is Head of Data & Analytics and AI Strategy. His role is to lead the Data Office and establish data as a trusted, governed and reusable enterprise asset, ensuring it is accessible in a secure and compliant way to drive business value. This includes creating the foundations to scale Data + AI capabilities across TSN.