Switzerland concentrates an extraordinary amount of regulated production: pharma and biotech along the Basel corridor, medtech clusters from Solothurn to the lake of Geneva, CDMOs everywhere between. All of it runs under GMP, and most of it has had some encounter with lean. The encounters range from transformative to theatrical, and the difference is almost always the same thing: whether the lean work respected the regulatory physics of the environment or tried to work around them.
What transfers, what does not
The lean core transfers completely: flow thinking, standard work, visual management, daily management, structured problem solving. A GMP plant is still a plant; material still waits, changeovers still eat capacity, unclear standards still produce variation.
What does not transfer is the automotive reflex of rapid, informal experimentation. In regulated production, process changes can touch validated states, and "just try it" can create a deviation or worse. The practiced answer is not to abandon experimentation but to sort it: a large share of improvement work (logistics, preparation, information flow, scheduling, workplace organization) touches nothing validated and can move fast, while parameter and process changes route through change control with the quality unit at the table from day one. Practitioners who cannot make this sort quickly lose either their improvements or their inspection readiness.
Where the value concentrates
Changeover and cleaning on filling and packaging lines. Format changes and cleaning validations dominate capacity on high-mix lines. Structured SMED under GMP, including documentation and release steps, routinely recovers 30 to 50% of changeover time without touching validated parameters.
QC laboratory turnaround. Batches wait for the lab more often than for production. Lean laboratory work on sample flow, campaigning and equipment scheduling shortens release lead time with zero regulatory exposure; the methods are about queues, not chemistry.
Deviation and batch-record flow. Deviations that take 60 days to close and batch records that circulate for weeks are flow problems wearing quality clothing. Applying value stream discipline to the paper and data streams cuts release times dramatically, and a daily management rhythm around CAPA aging keeps them cut.
Tech transfer and scale-up. The most expensive waiting in the industry happens between development and routine production. Lean applied to transfer, treating the transfer itself as a value stream with standard work, shortens time to stable supply.
The practitioner profile
Sector experience is not a nice-to-have here; it is the entry ticket. A practitioner who has held responsibility under GMP knows what an auditor will ask, how Swissmedic and FDA expectations differ in practice, what a validation engineer will veto and why, and how to write improvement into the quality system rather than around it. Our network holds a dedicated pool of practitioners with exactly this history (the medtech and pharma segment accounts for 180 plus engagements), vetted for regulated-environment responsibility, matched within 24 hours of a brief.
FAQ
Does lean conflict with GMP?
No. GMP defines what must be controlled and documented; lean improves how work flows within those controls. Conflict arises only when improvement work ignores change control, and that is a competence failure, not a methodology conflict.
Where should a pharma site start with lean?
Where the waiting is: usually changeover time on constrained lines or QC turnaround. Both deliver measurable results inside a quarter, build credibility for wider work, and touch little or nothing that is validated.
How is lean different in medtech versus pharma?
The logic is identical; the regulatory frame differs (ISO 13485 and MDR versus GMP), and medtech's higher product mix makes changeover and materials flow even more dominant. The practitioner needs history in your specific regime.
Can we run kaizen events in a GMP environment?
Yes, with the quality unit in the room and scope sorted into touch-nothing-validated actions (execute now) and change-control items (structured pipeline). Events run this way produce both improvement and cleaner audits.