Series · The real breakpoints of manufacturing operations
More Systems, Yet Operations Still Get Stuck?
Most manufacturers are no longer short of systems. What is missing is one thread connecting promise, change, readiness, execution and exception - a thread that can carry accountability and warning. This series answers three questions in order: where the problem sits, how to judge it, and when AI finally becomes relevant.
The three articles in this series are currently published in Chinese. Each link below carries a language label.
Where to start
Three unrelated-looking problems that usually sit on one chain.
See which of the three is closest to your situation, then choose where to start reading.
Recommended reading order
Read in this order and the chain of reasoning holds.
Each article stands on its own, but the order matters: settle how the problem is framed, then locate the breakpoint, and only then decide about AI.
- Part 1Read in 中文 →
Why a digital project has to start from a business problem
When a project starts from a system shortlist rather than a business problem, there is no way to decide afterwards whether it worked.
Why it sits here: Answer “what is the problem” before “which system”. This piece explains why projects that start from a shortlist cannot be evaluated later.
- Part 2Read in 中文 →
The five most common breakpoints from order to delivery
Losing control of delivery is rarely a scheduling problem. It is the absence of one operating thread across order, demand, material, production and exception handling.
Why it sits here: It splits the delivery problem into promise, change, readiness, execution and exception. By the end you can say which step is failing in your own operation.
- Part 3Read in 中文 →
Four conditions to meet before AI enters real operations
The value of AI is not decided by a model demo. It depends on whether data, process, accountability and measurement can carry a recommendation into real action.
Why it sits here: Before considering AI, confirm the data, the process entry, the accountability and the measure. Miss one and AI simply produces unusable advice faster.
Tools that go with it
Run the assessment yourself first.
The two tools match the first two questions in this series and can be used directly in a management meeting.
Where the evidence comes from
The actual work behind these judgments.
Client profiles are anonymized; participation and deliverables are stated as they were. Implementation results of planning work are verified separately.
Participation: our core team members worked on this project
A globally operating manufacturer of high-performance industrial hose; approximately four months; group and several business units. It delivered assessment, target blueprint and a multi-year roadmap, and completed client acceptance. The claim in this series that breakpoints are distributed along the whole chain comes mainly from cross-unit work of this kind.
The methods, metric definitions and ownership mechanisms in the order-to-delivery and inventory programs are the direct source of the two tools in this series. Scope and attribution for the individual projects are set out on the case studies page.
Related articles
If you are already weighing AI and agents.
A separate series on governance, architecture and selection for agents in high-compliance process manufacturing.
- 01Read in 中文 →
Rules before tools: agent levels L0–L4 and a registry
- 02Read in 中文 →
A prompt is not control: the deterministic kernel and three red lines
- 03Read in 中文 →
Three questions that decide agent selection
Related services
If you want someone working on it with you.
The articles and tools are enough to support an internal judgment on their own.
Operations & digital diagnostic
When the problem is not yet clear: build a problem map and a set of priorities.
View service →Order-to-delivery & supply chain
When the problem is clearly on the delivery chain: build the metrics, process and ownership loop.
View service →Inventory & obsolescence
Align the obsolescence definition and turn clean-up into a running capability.
View service →AI scenario assessment
Once the four conditions hold, decide which task is worth introducing AI into.
View service →NEXT STEP
Bring the operating problem
you most want to change.
Delivery, inventory, operating visibility or a workflow where AI could help. Share your situation and priorities so we can identify a practical next step together.