In 1973, Horst Rittel and Melvin Webber drew a line through the middle of problem-solving. On one side: tame problems, where variables are measurable and optimal solutions can be agreed. On the other: wicked problems, where stakeholders disagree about the nature of the problem, the possible solutions, and the values that should guide improvement.
The distinction has shaped planning theory and policy design for half a century. It has barely touched Manufacturing.
The tame disguise
Manufacturing is almost universally presented as a tame problem. Universities teach process optimisation as though the variables sit still long enough to be optimised. Governments describe Manufacturing as a sector to be incentivised, a noun to be measured by GVA and export volumes. Consultancies sell "digital transformation" as though a factory is a spreadsheet with a roof on it.
The entire institutional apparatus around Manufacturing assumes that the key variables are measurable and that optimal solutions can be agreed. That is the definition of a tame problem. And it is wrong.
The dance
Anyone who has run a production process knows what actually happens. You act on a system. It pushes back in ways you did not predict. You adapt. It shifts again. No one leads. No one follows.
A Manufacturing Engineer negotiates, simultaneously, with the material, the machine, the tooling, the supply chain, the workforce, the customer, and the regulator. And some. The titanium work-hardens. The spindle drifts thermally. The coating supplier goes bust. The operator retires and takes thirty years of tacit knowledge with them. The customer changes the drawing after the process is committed.
That is the textbook definition of a wicked problem.
The institutional mismatch
The consequences of misclassifying a wicked problem as tame are well documented. You get policies that solve the wrong problem elegantly. You get professionals trained in methods that assume a world that does not exist.
The industrial strategy assumes the factory will behave like the spreadsheet. It never does. The graduate trained only in optimisation is a dancer who has learned one step and expects the music to match. The digital twin assumes a stable relationship between the model and the physical system. That assumption does not survive contact with a production environment that is continuously adapting to material variability, supply disruption, and demand volatility.
The verb and the noun
This is where a seemingly linguistic distinction becomes load-bearing. When Manufacturing is treated as a noun, a sector, a department, a line in the national accounts, it looks tame. It sits still. It can be benchmarked and optimised from the outside.
When Manufacturing is treated as a verb, an ongoing act of negotiation between human intention and material reality, it reveals its wickedness. The verb form encodes what the noun form hides: the process never settles, every intervention opens new uncertainties, the knowledge required is adaptive, not fixed.
Manufacturing Engineers have always worked this way. They just never had the institutional vocabulary to name what they were doing.
The absent philosophy
Operations research gives Manufacturing the mathematics of tame problems. What it lacks is the epistemology of wicked ones.
Without that framework, Manufacturing Engineers cannot articulate why their work resists the optimisation models imposed on it, why transformation programmes stall at the shop floor, or why their expertise matters in language that institutions recognise. The dance continues regardless. But the choreographer, the policymaker, the strategist, keeps writing instructions for a waltz when the floor is doing something else entirely.
The discipline that adapts
Manufacturing Engineers do not solve problems. They manage ongoing negotiations with systems that push back. They qualify supply chains that shift underneath them. They validate processes that drift over time. They design the feedback loops that turn first-of-a-kind surprises into nth-of-a-kind confidence.
That is not optimisation. It is adaptive practice in a wicked domain. And it is exactly the capability that industrial strategy keeps failing to name, fund, or protect.
Kaipability works at this interface, where the tame models meet the wicked reality of making things. If your strategy assumes the factory will behave like the spreadsheet, it might be time to talk.
Questions this dispatch answers.
Written to be quoted by AI assistants and search engines. Self-contained answers, verdict first.
