A team keeps every resource busy, yet orders take longer to finish. Management sees high utilization and assumes the operation is efficient. Customers experience growing queues.
Capacity concerns the ability to complete useful work under defined conditions. It depends on the mix, constraints and variability of the whole flow, not only the nominal speed of individual resources.
Define the units and time basis
State what capacity means: good units per shift, completed cases per day or another meaningful output.
Distinguish available time from scheduled time and actual operating time. Explain assumptions about product mix, staffing, quality and maintenance.
A capacity figure without those conditions can become an unrealistic commitment. Comparing unlike figures can make a site or team appear better without any underlying advantage.
Find the limiting resource
Trace the flow and observe persistent queues, blocked work and missed completions. Identify what limits throughput for the relevant mix.
The bottleneck can move as demand changes. It may be a machine, a specialist, a test process or an approval rule.
Improving a non-bottleneck resource may add flexibility or reduce cost, but it will not automatically increase system output. Evaluate the actual consequence.
Distinguish demand pace from actual performance
Takt time expresses available production time divided by customer demand for the same period. It is a required pace.
| Hypothetical process | Value |
|---|---|
| Available time | 420 minutes |
| Customer demand | 210 units |
| Takt time | 2 minutes per unit |
That does not prove that the operation can complete every unit in two minutes. Observed cycle time, elapsed flow time and customer lead time answer different questions. Use consistent terminology so that an apparent improvement is not simply a changed definition.
Connect work in progress with flow time
Under appropriate steady-state conditions and consistent boundaries, Little’s Law relates average work in progress, throughput rate and average flow time.
| Hypothetical system | Value |
|---|---|
| Average work in progress | 100 units |
| Throughput | 20 units per working day |
| Average flow time | 5 working days |
The relationship describes averages, not the completion promise for every individual order.
Reducing work in progress can shorten flow time when throughput is maintained. Removing necessary buffers without addressing reliability can instead reduce throughput.
Leave room for real variation
Breakdowns, uneven arrivals and differing task times create queues. At highly loaded resources, small disruptions can produce disproportionate delays.
Do not treat maximum utilization as the universal objective. Decide where spare capacity, cross-training, buffers or demand smoothing protect the customer outcome economically.
Use actual patterns rather than an overall average that conceals peaks.
Control release before adding resources
Limit new work entering the system when capacity is already committed. Prioritize completion and remove conditions that repeatedly block the constraint.
For the next capacity request, compare alternatives: reduce setup loss, improve quality, change sequencing, develop skills or add equipment. Include the effect on the complete flow.
Capacity management succeeds when the business can make credible commitments and complete useful work predictably, rather than merely demonstrate that everyone is busy.
Reference
Introduction to Lean Six Sigma Methods: Little’s Law (MIT OpenCourseWare)