Batch Size Reduction is a core Lean lever for flow. Smaller batches reduce waiting and inventory, but they require stable processes, setup discipline, quality at the source, and material planning that supports frequent movement.
Definition
Batch Size Reduction is the practice of reducing the number of units processed, moved, purchased, scheduled, or released together. In Lean systems, smaller batches support faster feedback, lower work in process, shorter lead time, better problem visibility, and improved responsiveness to customer demand.
The goal is not to make every batch arbitrarily tiny. The goal is to find the smallest practical batch size that supports flow, quality, cost, capacity, setup constraints, material handling, and customer demand. In many cases, batch size reduction must be paired with setup time reduction, pull systems, and reliable processes.
History
Batch thinking has roots in traditional mass production, where large runs were used to spread setup cost over more units. Lean challenged this assumption by showing that large batches create hidden costs through inventory, waiting, defects discovered late, schedule inflexibility, and long lead times.
The Toyota Production System emphasized flow, quick changeover, Just-In-Time, and smaller lot sizes. SMED work by Shigeo Shingo helped make smaller batches practical by reducing setup time and separating internal from external setup work.
When to Use
Use batch size reduction when lead time is long, WIP is high, defects are discovered late, schedules are inflexible, queues are large, changeovers dominate planning decisions, or customers need more frequent mixed-model delivery. It is useful in production, purchasing, paperwork, approvals, testing, software releases, training, and service backlogs.
Do not reduce batches without understanding constraints. If setup time is long, material movement is unreliable, quality is unstable, or suppliers cannot support smaller releases, the change may create chaos. Reduce batch size as part of a system improvement, not as a standalone target.
Step-by-Step
- Map the current flow. Identify batch sizes, queues, changeovers, transport, waiting, inspection, rework, and release rules.
- Measure lead time and WIP. Quantify how much time is spent waiting compared with processing.
- Identify setup and transaction costs. Understand why large batches are currently used.
- Reduce setup time. Apply SMED, preparation, tooling improvements, standard work, and point-of-use materials.
- Test smaller batches. Pilot a smaller lot size in a controlled area and monitor flow, quality, capacity, and schedule impact.
- Adjust replenishment and signals. Align kanban, supermarkets, FIFO lanes, purchasing quantities, and production schedules.
- Stabilize quality at the source. Smaller batches expose defects faster, but the process must respond quickly.
- Standardize and scale. Lock in the new batch rules, update planning parameters, and expand after evidence supports the change.
Examples
- Machining cell: A plant reduces lot size from 1,000 to 200 after setup time drops from 90 minutes to 20 minutes. WIP falls and defects are detected within the same shift.
- Paint process: A team groups colors differently and improves cleaning setup, allowing smaller paint batches without excessive downtime.
- Office approvals: A finance team stops holding invoices for weekly batch review and moves to daily flow, reducing late payment risk.
- Software release: A product team shifts from quarterly bundled releases to smaller validated releases, reducing deployment risk and feedback delay.
- Training deployment: A training team replaces one large annual class with smaller frequent sessions closer to the time of need.
Common Pitfalls
- Reducing batch size without reducing setup. More changeovers can overwhelm capacity if setup is not improved.
- Ignoring planning rules. ERP lot sizes, reorder quantities, and supplier minimums can silently force large batches back into the system.
- Local optimization. One department may prefer large batches while the whole value stream suffers longer lead time.
- No quality response. Smaller batches reveal problems faster; teams need reaction plans to act on them.
- Confusing transfer batch and process batch. A process may run a certain lot size but move smaller quantities downstream more frequently.
- Skipping material handling design. More frequent movement requires clear locations, routes, signals, and ownership.