Why Most AI Pilots Fail in Chemical Plants – Lessons from Industry Experts

Sky‑high expectations, inconsistent data and siloed pilots can prevent AI pilots from successful real‑world deployment. The solution lies in industrial intelligence that closes the loop between prediction and plant floor, writes Stephen Reynolds, Industry Principal, Chemicals at AVEVA.

Every Formula 1 car generates hundreds of gigabytes of telemetry data during a race – from tire degradation and fuel burn to brake temperature, weather changes and competitor behaviour. That data is streamed to engineers on the track and in remote labs in real time, allowing them to analyse performance, optimise strategy and gain a competitive edge in a sport where every millisecond counts. It’s a reminder that success is not about the size of your dataset or how sophisticated your model is, but whether insights are turned into timely, operational decisions.

Subscribe to PII and receive:

Our long‑established bi‑monthly digital magazine Process Industry Informer, featuring exclusive articles, news and updates across the Process Industries, plus commentary from our three industry experts:

Sean Moran

Gavin Smith

Dave Green

PLUS:

• Process Industry Update – our weekly e‑newsletter
• PII Podcast
• Special messages from our advertising partners

Chemical plants can outperform the market by treating AI like a high‑performance race car. They already generate billions of data points from sensors, labs and ERP systems. Companies are running AI pilots, testing models, and launching proof‑of‑concepts, with McKinsey reporting that 78% of organisations use AI in at least one business function.

Most AI pilots stall

Yet the vast majority of pilots stall before delivering value. A recent survey shows that 88% of AI pilots never reach production, and MIT reported in August that 95% of generative‑AI initiatives aimed at rapid revenue generation fall flat.

This “AI purgatory” isn’t a lack of imagination – it’s a lack of strategy and action. Translating insights into real‑world action, much like a perfect F1 strategy without tire changes, is essential.

AI isn’t plug and play

Teams often fall victim to shiny‑object syndrome, treating AI as a plug‑and‑play solution. When ambition outpaces infrastructure, models are applied to tasks they weren’t designed for, fed inconsistent or delayed lab and sensor data, and remain isolated from operations. The resulting insights can’t be applied to real‑world processes.

“A staggering 88% of AI pilots fail to reach production, according to a recent survey.”

Even when AI predicts fouling in a reactor or inactive catalysts, the value disappears if operators cannot act due to misaligned workflows. Impatience for instant results and the lack of continuous feedback from new feedstocks compound the problem. Fragmented technology equals fragmented results.

Achieving pole position – staying ahead of the competition – requires cross‑functional collaboration and a connected ecosystem that unifies industrial chemical systems. That means linking MES, LIMS, ERP, historians and process control systems within a single platform, creating a single source of truth that breaks down data silos and feeds intelligent insights into existing control loops at the shop floor and executive level.

“Only with these three elements – curated data, process‑aware intelligent models and inspired humans in the loop – does AI move beyond theoretical concepts to real operational and R&D outcomes, delivering lower downtime, higher yields and shorter innovation cycles.”

AI augments, humans elevate

Curating and centralising industrial data is the starting point. Models must be designed to respect upstream and downstream dependencies. Most importantly, teams must trust the intelligence they receive and be empowered to act on it. Avoiding AI purgatory should be seen as a cultural shift rather than a mere technology upgrade.

Only with curated data, process‑aware models and inspired humans in the loop does AI transition from concept to tangible operational and R&D benefits.

That’s how SCG Chemicals achieved 99% plant reliability and a ninefold return on investment in just six months. To keep one of Asia’s largest chemical supply chains humming, SCG built a digital reliability platform embedding AI across its lifecycle.

By integrating predictive analytics, centralised data and digital twin environments in one place, the platform enables teams to make process decisions on the fly, akin to arming F1 racing teams with real‑time intelligence.

With dashboards ranging from the business unit level to individual equipment, SCG’s teams access actionable information and correlate it with real‑time data within 10 seconds. Identifying critical points to avoid asset failures has closed the reliability gap, and maintenance costs are down 40%.

Scaling across the chemicals sector, AI use cases extend to improving asset uptime with predictive analytics, hybrid modelling to accelerate product innovation, and even ingredient discovery for environmentally sustainable materials.

“When chemicals companies adopt this step‑by‑step approach, they move beyond reactive troubleshooting.”

Preventing AI pilot purgatory

Success in each area requires more than treating AI pilots as technology experiments – a path that guarantees inclusion in the 95% failure statistic. As digital and analytics tools are adopted, companies need end‑to‑end approaches to convert analytics into operational improvements, as Deloitte highlighted in its recent chemical sector outlook.

Overcoming AI purgatory starts with a shift in perspective, including cultural change. First, define the KPI you want to change and quantify its workflow impact. Pilots that try to do everything achieve nothing.

Next, build a data‑first connected ecosystem, integrating historians, MES, LIMS and vendor programs. Success rests on data quality, as Arthur D. Little notes; schemas must be standardised, metadata annotated and lab protocols established.

Then select the right AI and make it observable. For example, pattern recognition can forecast equipment failures, LLMs can search compliance documents, and hybrid modelling can drive innovative applications such as formulation.

After that, productise and scale one use case at a time. Finally, cross‑functional evaluation and governance – McKinsey recommends tasking senior leaders with oversight – reduces model drift and adoption risk.

When chemicals companies adopt this step‑by‑step approach, they move beyond reactive troubleshooting. Operators can anticipate fouling, adjust reaction conditions and prevent downtime. R&D teams accelerate formulations while ensuring consistent scale‑up. The industry can finally extract real value from its AI investments, just as F1 teams convert telemetry into split‑second, race‑winning decisions.