Honeywell Technologies’ Uygar Doyuran discusses the cybersecurity, data and AI foundations needed to advance industrial autonomy while protecting safety and uptime.
Industrial autonomy brings new opportunities to improve operations, alongside cybersecurity challenges as connected systems and AI take on greater responsibility. Uygar Doyuran, Vice President and General Manager for Honeywell Technologies Process Automation across the Middle East, Turkey and Africa, explains why progress depends on secure operational technology, connected data and AI grounded in industrial expertise. He also outlines how organisations can build on existing infrastructure through a phased approach that keeps resilience, safety and business outcomes central.
Interview excerpts
Why does greater industrial autonomy create cybersecurity challenges beyond those associated with traditional automation?
Greater autonomy means more assets, systems and data are connected, while AI is increasingly supporting or taking operational decisions. That expands the cyber-attack surface and increases the potential consequences of a breach. In an industrial environment, a cyber incident can affect more than information. It can disrupt production, equipment, safety or uptime. Honeywell Technologies therefore sees cybersecurity as an integral part of the path to autonomy, particularly as operational technology becomes more connected with enterprise systems and cloud platforms. This is why we are investing in OT-specific technologies such as Honeywell Technologies Cyber Proactive Defense, which uses AI and process-domain knowledge to correlate cyber anomalies with operational activity and help identify risks before they affect operations.
Can organisations realistically deploy AI-driven autonomous systems across aging or fragmented operational technology infrastructure?
Industrial organisations rarely operate entirely new infrastructure, so the transition to autonomy has to work across legacy, third-party and newer systems rather than depend on replacing everything. The first step is understanding what assets, data and connectivity already exist, then identifying where AI can solve a clear operational problem. Honeywell Technologies Forge is designed to connect fragmented data, domain knowledge and control systems across existing operations. Globally, Forge connects more than 32,000 customers, 324,000 sites and approximately 6.4 million assets. We are also seeing organisations take a more pragmatic approach to autonomy by building on the infrastructure they already have.
The priority is to understand existing systems, digital maturity and operational gaps, then connect the relevant data and introduce AI where it can support clear business outcomes.
How can industrial organisations protect connected environments while enabling data and AI to move securely across operations?
Organisations need to understand what assets are connected, how systems communicate and where vulnerabilities exist before they can secure data movement effectively. From there, organisations need OT-specific cybersecurity covering network segmentation, continuous monitoring, controlled data flows and threat detection. The aim is not to stop operational data moving, because that data is essential for AI and autonomous operations. It is to ensure that it moves through secure and trusted pathways. As operations become more connected and data-intensive, organisations need cybersecurity built into their digital strategy from the outset. Secure infrastructure, controlled access and continuous monitoring can help organisations use operational data more effectively while maintaining the resilience required in mission-critical environments.
Why is deterministic, domain-specific AI critical in industrial environments where incorrect decisions could affect safety or uptime?
Industrial AI operates in environments where decisions can directly affect physical processes. In a refinery, manufacturing plant or critical infrastructure environment, an incorrect action can have consequences for production, safety and equipment. Honeywell Technologies’ approach therefore combines Data, Domain Knowledge and Deterministic AI. The data tells the system what is happening, domain knowledge provides the context of how the process should operate and deterministic AI applies that intelligence within defined physical and operational constraints. Our solutions are built around this principle, combining deterministic models with real-world operational constraints. Honeywell Technologies Experion Cognition takes this further by using AI-enabled agents to identify and address process abnormalities. In pilots, its Operations Assistant predicted alarm incidents an average of five to 10 minutes before they occurred.
What should Middle East industrial organisations prioritise to progress securely from automated to autonomous operations?
Organisations should start with a defined operational challenge, then assess whether their existing data, systems and cybersecurity can support greater autonomy. From there, they can build a phased roadmap around connected data, domain expertise and deterministic AI. Cybersecurity needs to be built into that roadmap from the outset. That includes visibility across OT assets, segmentation, secure data movement, continuous monitoring and clear controls over where autonomous systems can act. The opportunity is to build greater intelligence into existing operations rather than treating autonomy as a complete technology replacement. Across the Middle East, this is particularly relevant as organisations modernise existing industrial infrastructure while also investing in new digital capabilities.
Source: Tahawul Tech
