Batch processing is the practice of producing a group of identical items together in one run, then passing the whole batch to the next step. It has been the default logic of manufacturing for centuries, and it is one of the largest sources of waste that lean tries to remove.
Instead of making one part and passing it downstream immediately, a batch producer accumulates a run, 50, 500, 5,000 units, before moving anything. The logic is scale: setups are expensive, so amortize them across as many parts as possible. Machines are dedicated to long runs of one product before changing over to the next.
Long batches create long lead times: a part made first in a batch of 500 still waits for all 500 to finish before the batch moves on. Defects are detected late, by the time station three catches a problem, station one has already produced hundreds of defective parts. Inventory piles up between stations. Response to demand changes is slow because the system is committed to whatever is already in production.
Lean prefers small batches, ideally one-piece flow. The trade-off is more setups, so lean invests heavily in reducing setup time (SMED) until it becomes cheap enough to change over frequently. Small batches shorten lead time, expose defects immediately, cut inventory, and let the factory respond to demand as it changes instead of days or weeks later.
Some processes have physical batch constraints, a heat treatment oven, a chemical bath, a printing plate. In those cases the lean answer is not to force one-piece flow but to make the batch size as small as physically possible and to sequence batches to match downstream pull. The goal is always the shortest path from raw material to customer.
Batch processing is the traditional default; flow is the lean alternative. Moving from one to the other is one of the largest transformations a factory can undertake, and one of the most rewarding, shorter lead times, less inventory, faster feedback, happier customers.
