Production Learning Log¶
Capture register — instantiates Experience Curve Cost Reduction
A record kept at the workbench where each discovery, snag, and trick is written down the moment it surfaces, before it evaporates into tacit memory.
The friction that a curve is built on is discovered at the work, in the moment, by the person doing it — and it disappears just as fast, folded into the operator's hands as something they now "just know." Production Learning Log is the artifact that intercepts it. It is a low-friction record living at the point of work, where an operator writes down each discovery the instant it happens: the fixture that chattered, the sequence that jammed, the small trick that shaved a step. Its defining property is immediacy and proximity — it captures the perishable, half-articulated detail while it is still fresh and still specific, before it hardens into tacit memory or vanishes at shift change. It is the first mile of turning what one person learned into something the whole system can keep; it captures and first-drafts the explicit note, and nothing more.
Example¶
A precision machine shop runs short, varied CNC jobs, and every new part teaches the machinists something the program didn't. At each station sits a running log — a tablet, a card, a shared sheet — and the rule is one line per surprise, entered before the machinist walks away. During a run of a thin aluminum bracket, a machinist finds the part rings and leaves a poor finish until she drops the feed rate on the final pass and adds a support tab; she logs it against the part number in twenty seconds: "thin-wall bracket — chatter on finish pass, fix: feed to ~60%, temp support tab."
By itself that entry saved nobody but her — until it is there in writing, tagged to the part, when the job runs again next quarter on a different machinist's shift. Across a month the log accumulates a few dozen such lines, each a small piece of hard-won friction made explicit and specific enough to act on. What the log does not do is decide which of them become standard practice; it just makes sure none of them are lost first.
How it works¶
- Capture at the point of work. The log lives where the work happens and is written by the person doing it, so tacit detail is caught before the walk back to a desk erases it.
- Keep friction near zero. An entry is a line, not a report — low enough cost that people actually log the small stuff, which is where most learning hides.
- Structure just enough. A light template (unit or part, symptom, fix) makes entries specific and later-codifiable without turning capture into paperwork.
- Tag for triage, then hand off. Entries are flagged as candidate-for-standard or one-off and passed to the revision cadence; the log gathers raw ore, it does not smelt it.
Tuning parameters¶
- Capture friction — how much effort an entry costs. Lower friction yields more entries but noisier ones; higher structure yields cleaner entries but fewer of them.
- Structure vs. freeform — fixed fields versus open notes. Structure aids downstream codification; freeform lowers the barrier to logging at all.
- Proximity — how physically and temporally close to the work the log sits. Closer captures perishable tacit detail; a log filled out later captures a sanitized memory.
- Triage tagging — whether entries are marked for the revision cadence as they land. Tagging speeds codification; untagged logs become an undifferentiated pile.
- Retention and pruning — how long raw entries live before they are codified or discarded, so the log stays a live queue rather than an archive.
When it helps, and when it misleads¶
Its strength is that it catches the cheapest, most perishable, most valuable input to a learning curve — the tacit floor knowledge that never makes it into a report because nobody thinks to write it down. Polanyi's point that we know more than we can put into words is precisely why the log has to sit at the work: the detail is half-articulated and it decays, so the only way to keep it is to capture it there and then.[n1]
Its failure mode is the write-only log: a register that fills up and is never read, capture with no codification behind it, so the entries accumulate and rot. Vague entries ("machine acting up") are nearly as useless as none. The classic misuse is treating the log as the deliverable — "we logged it" standing in for actually folding the lesson into standard work — capture theater that produces the paperwork of learning without the learning. The discipline that guards against it is to pair the log with a downstream cadence that genuinely consumes it, to insist entries are specific enough to act on, and to keep it a queue that is worked, not an archive that is filled.
How it implements the components¶
Production Learning Log fills the capture-side components of the archetype — the ones an at-the-work register operates:
learning_capture_loop— it is the loop's front end: the routine that catches a discovery at the point of work the moment it surfaces.tacit_to_explicit_codification_path— it drafts the first explicit form of tacit floor knowledge, turning "something I now just do" into a written, specific line others can read.
It captures and first-drafts but does not fold the entry into official standard work or push it to every site (standard_work_update_path, transfer_and_training_channel — Playbook Revision Cadence); the episodic, post-event counterpart to its continuous capture is After-Action Review.
Related¶
- Instantiates: Experience Curve Cost Reduction — it is the capture front end that feeds everything downstream on the curve.
- Sibling mechanisms: Playbook Revision Cadence · Standard Work Revision · After-Action Review · Yield and Defect Pareto Review · Time-and-Motion Study · Setup Reduction Workshop · Simulation Drill Ladder · Cumulative Volume Cohort Analysis · Experience Curve Model · Learning Rate Dashboard
Editorial Notes¶
Form Classification¶
Form family: Record, Log & Register
Rationale: Production Learning Log operates as a persistent ledger, log, register, or case record that preserves history and traceability because it a record kept at the workbench where each discovery, snag, and trick is written down the moment it surfaces, before it evaporates into tacit memory.
Independent corroboration: The frozen evidence defines Production Learning Log as 'A record kept at the workbench where each discovery, snag, and trick is written down the moment it surfaces, before it evaporates into tacit memory', so its operative form is Record, Log & Register.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Organizational & Management Science
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Production Learning Log is most plausibly rooted in the organizational_management tradition because its characteristic form depends on the coordination, governance, learning, and redesign of organized work. The assignment tracks that formative lineage, not the many settings in which the mechanism can now be applied.
Related originating lineages:
- Education & Pedagogy — The education_pedagogy tradition materially shaped Production Learning Log through its own practice of instruction, learning progression, assessment, and scaffolded practice.
- Engineering & Design — The engineering_design tradition materially shaped Production Learning Log through its own practice of physical-system design, process control, reliability, and safety engineering.
Review resolution: Both blind reviewers agree that organizational management is the primary origin. Explicit reconciliation resolves alternate origin disagreement, origin mode disagreement, encyclopedia synthesis disagreement. Formative alternate lineages are retained as education_pedagogy, engineering_design; later breadth of use is recorded separately as domain_reach=multi_domain, while origin_mode=cross_disciplinary_synthesis describes the relationship among origin lineages.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
A log with nothing downstream to consume it is a graveyard. Its value is entirely realized by the mechanism that reads it — Playbook Revision Cadence — so a shop that starts the log without also standing up a revision beat will find, months later, a full register and an unchanged practice.
[n1] Tacit knowledge — Michael Polanyi's observation that much of what skilled people know is held as feel and habit they cannot fully articulate. Because that knowledge is both perishable and hard to state, the log's job is to catch its first explicit fragment at the moment and place it is least lost. ↩