MECO’2026 and CPSIoT’2026 Keynote Speakers

Conference Keynote Speakers will be announced soon.

Keynote speaker #1

University of Antwerp, Department of Computer Science, Belgium

Title: CPSIoT – The Twinning Paradigm

Abstract: The Cyber-Physical Systems (of Systems) we design, maintain and above all, evolve, over increasingly long periods of time, are ever increasing in complexity. In the meantime, demands on quality, maintainability, sustainability, etc. become more and more stringent. Using “digital twins” have been shown to enable a host of desirable extra-functional system features such as condition monitoring, fault diagnosis, predictive maintenance.
There has been a proliferation of terminology, examples and standards for digital twins. In this presentation, I will try to unify and structure these under the “twinning” umbrella, building on many existing techniques, architectures and standards from real-time simulation and execution, co-simulation, systems and control theory, IoT, knowledge management, machine learning, surrogate modelling, etc.
In the twinning paradigm, a twinning architecture (a system in its own right) is used whereby a virtual instance of a System under Study (SuS) or “asset” is continually updated with the SuS’s health, performance, maintenance, etc. status information, and this throughout the asset’s life-cycle (requirements analysis, design, production, assembly, operation and optimization, maintenance, re-purposing, disposal).
I will take a product family/line approach to structure desirable system features, the different workflows and conceptual twinning architectures needed to realize these, and finally, the different means to deploy/realize these architectures.

Keynote speaker #2

Prof. dr. Milan Prokin

University of Belgrade – School of Electrical Engineering

Title: (Near) Future of Vehicles

Abstract: During recent decades, most vehicle manufacturers decided to start design and manufacturing of fully electric (battery only powered) vehicles (cars, trucks, airplanes, helicopters, etc.) that were initially enthusiastically accepted by buyers, but the situation is completely opposite nowadays due to many reasons that will be described in this keynote. Spoiler alert, future is still bright, but without enforcing a single vehicle solution that won’t attract most buyers.

Keynote speaker #3

Dr. Akbar Sheikh-Akbari

Leeds Beckett University, School Of Built Environment, Engineering And Computing, Leeds, United Kingdom

Title: Advancing Non‑Destructive Spectral Intelligence for Aflatoxin Risk Mitigation in Pistachio Supply Chains

Abstract: Aflatoxin contamination in pistachios remains a major constraint for global trade and food safety, particularly under increasingly stringent regulatory frameworks such as the European Union’s 8 µg/kg limit for aflatoxin B₁. Conventional analytical methods, including HPLC‑FLD and ELISA, provide high‑fidelity quantification but are destructive, labour‑intensive, and unsuitable for high‑throughput industrial screening or for deployment in resource‑limited production contexts. These limitations have accelerated interest in hyperspectral imaging (HSI) as a non‑destructive modality capable of capturing chemically informative spectral–spatial signatures at scale.

Recent advances in deep spectral modelling have significantly improved the feasibility of HSI‑based aflatoxin detection. Lightweight architectures such as AS‑HybridSN introduce learnable asymmetric spectral compression within 3D convolutional layers, reducing hundreds of spectral bands to compact, task‑specific representations while preserving spatial fidelity. When evaluated on the chemically validated HyperPistachio dataset, such models achieve near‑perfect classification accuracy with millisecond‑level inference, demonstrating clear potential for real‑time industrial integration. Complementary developments in spectral attention, particularly the incorporation of squeeze‑and‑excitation mechanisms into 3D CNN, Inception, and ResNet backbones, further enhance inter‑band feature weighting, yielding consistent performance gains across multiple architectures.

This keynote will synthesize current progress in spectral–spatial modelling, attention‑based feature refinement, and dataset curation, while outlining a research agenda aimed at standardisation, cost reduction, and equitable deployment. By bridging algorithmic innovation with practical implementation pathways, it aims to advance non‑destructive aflatoxin surveillance as a robust, scalable, and globally accessible technology for the pistachio industry.