Let’s continue the analysis of knowledge in a chemical plant.

In a typical chemical plant, everyone is competent, and this is ensured by continuous training, assessments, on the job training, and so on.

But knowledge and way to acquire it follow different path:

• Operators understand the process through sound, smell, timing, and deviations

• Control room engineers see trends, loops, and alarms

• Process engineers think in models, balances, and constraints

• Management sees KPIs, costs, and output

Each level is rational.

Each level is also , somehow blind to what the others see.

Because of this plant improvements are localised, leaving the plant overall optimisation still partially untapped.

The interfaces we use do not help in the case of a system base situation.  Trans-department knowledge tools remained undeveloped. Before information technology era transfer was managed through SOPs, reports, notice, meetings. Today everything is the same, only it resides in the IT domain, despite the possibility to move further on in the management of human/system interfaces.

They are designed to display local variables, not system behaviour.

As a result, we ask people to make system-level decisions using tools that describe only fragments of reality.

This creates a cognitive gap:

• Operators feel decisions are disconnected from what actually happens in the plant

• Engineers feel the plant is “not operated as designed”

• Management feels execution is slow, defensive, and resistant to change

Adding more data does not close this gap.

More dashboards do not close this gap.

More reports definitely do not close this gap.

Because of this pla

What is missing is a shared visual and conceptual layer (this is what we refer as interface) that allows different roles to look at the same plant and understand it according to their responsibility.

Before investing in more analytics,

it may be worth asking a simpler question:

What does the plant actually look like to the people running it?