Why Engineering Knowledge Can Disappear Between Projects

Reading Time: 5 minutes

A project closes, the equipment ships and the final drawing pack is archived. By the time the next job begins, the people involved may have moved on, so the project appears complete and its files are left to speak for themselves.

The problem becomes apparent when another team faces something similar. They can see the final design, but they may have no record of why a lighter concept was rejected or why an apparently awkward detail was retained. The answer survived; much of the engineering behind it did not.

Useful knowledge usually disappears through the accumulation of small gaps in the project record. A decision remains in an informal note or somebody's memory, and the context around it gradually becomes difficult to recover. By the time anyone notices, much of the detail needed to understand the design has gone.

The drawings don’t tell the whole story.

Drawings and CAD models are good at defining the approved result. They communicate the final geometry and its manufacturing requirements, but they seldom explain the problem that led to a particular choice. A future engineer can therefore understand what was built without understanding the conditions that made it sensible.

Consider a support bracket that looks heavier than necessary. The extra stiffness may have been introduced after vibration appeared during a trial, yet that history will not be obvious from the drawing alone. Years later, somebody reviewing the machine for cost reduction might remove material, believing they are correcting an overdesigned part when they are actually reintroducing an old problem.

Revision histories offer only limited protection because they show that something changed, rather than the reasoning behind the change. The same applies to calculation files: even where the analysis is retained, the assumptions and practical observations that shaped the final decision can be hard to identify.

Rejected ideas are particularly easy to lose, although they can be among the most useful parts of a project. Knowing that an arrangement was investigated and abandoned for a sound reason prevents another team from spending time on the same route. Without that record, each project begins by rediscovering constraints that the business has already paid to understand.

Capture knowledge early

The conventional lessons-learned meeting comes at an awkward point. Once the design has been released and the immediate problems have passed, memories soften as the people closest to the detail move on. What remains is often a general observation such as “improve supplier communication”, which says little about what should change on the next job.

Knowledge is easier to preserve while the decision is still being made. A brief note after a design review or test can record why the choice was made and what conditions influenced it, while those details are clear. It doesn’t need to be a separate report. A dated entry in the project decision log (linked to the relevant drawing or model) is usually more useful than a polished document written months later.

The level of detail should reflect the consequence. Routine choices do not need an essay, whereas a decision with meaningful consequences for the machine or the people who use it deserves enough context to survive a change of personnel. A note such as “plate thickness retained after the fabrication trial because the lighter panel distorted during welding” gives the next engineer something they can assess and, if circumstances have changed, challenge intelligently.

Some lessons belong beyond the project that produced them. When a fabrication detail repeatedly causes difficulty, the lasting response may be to update a drawing template or design standard so that later work benefits automatically. Leaving the lesson in a close-out report places the burden on every future team to find it again.

Stored knowledge still needs to be found

A record has little value if nobody knows where to look. Project folders often contain several versions of the same document, while important explanations remain attached to individual inboxes or meeting minutes with vague filenames. The information technically exists, but retrieving it takes longer than solving the problem again.

The storage method can be simple, provided it is consistent. Each project needs a recognised home for decision records, and the links between those records and the released design should remain clear after handover. Somebody should also own the final record, checking that the useful material is present and that draft versions have not obscured the approved information.

Retrieval should form part of starting work, rather than depend on somebody remembering an old job by chance. A short review of comparable projects can reveal earlier assumptions and problems found during commissioning before a new concept gathers momentum. Engineers who were involved previously may still add context, but the discussion is much more productive when there is a written record to work from.

This record also helps less experienced engineers because seeing the decision behind a design teaches far more than copying its geometry. It shows how technical requirements were balanced against manufacturing and operating realities. Over time, a usable record becomes part of how the team develops judgement rather than simply a place where completed files are stored.

Let the next project start further ahead.

Engineering knowledge doesn’t necessarily need a large management system to be useful. It needs to be captured close to the event and stored where the next project will look for it, those habits protect the reasoning behind a design without adding a heavy administrative process.

The benefit appears when an existing machine is modified or a familiar problem returns under a new team. Instead of treating every inherited detail as unexplained history, the team can see which assumptions still hold and where fresh investigation is justified. That leads to better questions at the start of the work and fewer avoidable discoveries later.

Projects should leave more behind than a set of released drawings. The next team needs enough of the thinking to understand why the design took its final form and to build on it with confidence.

ATLUS provides design engineering and engineering support for industrial machinery projects. When we take on an existing design, recovering its original intent is often the first step towards improving it. Talk to ATLUS if a project would benefit from additional engineering input.