In the realm of beverage manufacturing, a line can appear fast while the overall value stream remains slow. A filler rated at 600 bottles per minute may still miss customer demand when changeovers, micro-stops, waiting, quality holds and excessive work in process interrupt the flow.
This is where value stream mapping becomes practical rather than theoretical. The map connects syrup preparation, cleaning, filling, packing, palletising, information flow and finished-goods release in one view. It shows where customer value is created, where time accumulates and which constraint requires immediate attention.
This worked example uses a two-shift beverage plant producing 500 ml PET still water and 1.25 L carbonated products on the same line.
Scope selection: define the product family before mapping
The fundamental purpose of scope selection is to prevent a value stream map from becoming a general factory diagram. We will map one product family with a shared process route:
Syrup room batch preparation → CIP → depalletiser → bottle rinser → filler → capper → labeller → shrink-wrapper → palletiser → finished-goods release
The product family includes:
- 500 ml PET still water
- 1.25 L carbonated soft drinks
- Shared filler, labeller, shrink-wrapper and palletiser
- Format and recipe changeovers between campaigns
- Two shifts per day, each with 450 productive minutes
The Voice of the Customer is expressed as reliable delivery, correct format, intact packaging and consistent fill quality. The Voice of the Business adds throughput, energy control, asset utilisation and production cost. The Voice of the Process is the measured evidence that connects both requirements.
Takt time: the rhythm the line must support
Monthly demand is 480,000 cases. Each case contains 24 bottles.
[
480,000 \times 24 = 11,520,000 \text{ bottles per month}
]
With 26 working days:
[
11,520,000 \div 26 = 443,077 \text{ bottles per day}
]
Available production time is:
[
2 \text{ shifts} \times 450 \text{ minutes} \times 60 = 54,000 \text{ seconds per day}
]
Therefore:
[
\text{Takt time} = 54,000 \div 443,077 = 0.122 \text{ seconds per bottle}
]
That equals approximately 7.3 seconds per bottle, or 492 bottles per minute. The filler’s 600-bottle-per-minute rating provides theoretical capacity, but only if the line protects the filler from starvation, downstream blocking and repeated micro-stops.
Current-state value stream map: where the 9.4 days go
The current-state map shows a total lead time of approximately 9.4 days, while cumulative value-added processing time is only around 22 minutes. That produces a flow efficiency of:
[
22 \text{ minutes} \div (9.4 \times 24 \times 60) = 0.16%
]
Even allowing for necessary cleaning, inspection and handling, the result is comfortably under 1%. Most elapsed time is inventory, waiting, scheduling delay or release delay.
Current-state inventory and information points
| Stream location | Typical inventory or delay | Equivalent time |
|---|---|---|
| Ingredients, preforms, caps and packaging materials | 1.8 days | 1.8 days |
| Empty bottle and packaging staging | 1.6 days | 1.6 days |
| Syrup room to line queue | 0.6 days | 0.6 days |
| Line-side WIP and accumulation conveyors | 0.15 days | 0.15 days |
| Quality hold and release queue | 2.1 days | 2.1 days |
| Palletised finished goods | 2.8 days | 2.8 days |
| Dispatch scheduling queue | 0.35 days | 0.35 days |
| Total current-state lead time | 9.4 days |
Process data collected at the gemba
The following baseline uses representative campaign averages. Cycle time is shown per unit where the process is continuous and per batch or cycle where that is more meaningful.
| Process step | Cycle time | Format changeover | Uptime | Scrap |
|---|---|---|---|---|
| Syrup room batch preparation | 18 min/batch | 40 min | 91% | 0.2% |
| CIP | 42 min/cycle | Included in cleaning plan | 88% adherence | Not applicable |
| Depalletiser | 0.8 sec/bottle equivalent | 12 min | 95% | 0.1% |
| Bottle rinser | 0.3 sec/bottle | 10 min | 96% | 0.1% |
| Filler | 0.10 sec/bottle at rated speed | 55 min | 78% | 0.7% |
| Capper | 0.10 sec/bottle | 55 min | 89% | 0.15% |
| Labeller | 0.20 sec/bottle | 55 min | 90% | 0.3% |
| Shrink-wrapper | 0.6 sec/case | 35 min | 92% | 0.4% |
| Palletiser | 0.6 sec/case | 30 min | 94% | 0.1% |
The current filler OEE is approximately:
[
\text{OEE} = 78% \text{ Availability} \times 81% \text{ Performance} \times 98.5% \text{ Quality}
]
[
\text{OEE} \approx \mathbf{62.2%}
]
The filler is the practical bottleneck. Its performance loss is not caused by one dramatic event. It is accumulated through short stops, speed reductions, capper interruptions, carbonation variation and long format changes.
The filler downtime Pareto: prioritise the vital few
The monthly downtime Pareto below converts observations into an improvement agenda.
| Downtime category | Minutes/month | Share |
|---|---|---|
| Filler micro-stops and speed losses | 1,920 | 43% |
| Format changeovers | 960 | 21% |
| Capper and downstream jams | 620 | 14% |
| CIP overruns | 420 | 9% |
| Material starvation or waiting | 300 | 7% |
| Quality holds and start-up checks | 260 | 6% |
| Total | 4,480 | 100% |
The first two categories account for 64% of recorded downtime. That is the logical starting point for the Improve phase of DMAIC.
Micro-stops should be captured automatically wherever possible. A stop lasting six seconds may appear insignificant on an individual event log, yet hundreds of such events can remove a substantial portion of available throughput. Use reason codes that distinguish filler faults from the true origin of a stop, such as capper jams, low product supply or downstream blocking.
Changeover analysis: convert internal work into external work
The current 55-minute format change breaks down as follows:
| Changeover element | Current time | SMED direction |
|---|---|---|
| Stop line, confirm schedule and isolate equipment | 8 min | Standardise and shorten to 3 min |
| Drain, rinse and secure product path | 7 min | Use prepared connections and visual checks |
| Remove format parts | 12 min | Externalise tool staging; use quick-release fittings |
| Install new format parts | 14 min | Parallel work by two trained operators |
| Set label, film, cap and recipe parameters | 6 min | Pre-load and verify settings before the stop |
| Start-up checks and QA approval | 8 min | Create first-off standard and defined approval limits |
| Total | 55 min | Target: 15 min |
Approval is essential for food safety and governance, but an unclear approval boundary can create a bottleneck. The solution is not to bypass QA. It is to define the standard, prepare the evidence and establish a rapid first-off decision.
Using SMED, the target is to reduce the changeover to 15 minutes by separating internal and external tasks, staging parts before the stop, using parallel operators and standardising recipe verification.
Eight wastes visible in this value stream
The DOWNTIME framework makes the current map actionable:
- Defects: Fill-level, cap, label and shrink-pack defects create scrap and inspection demand.
- Overproduction: Long campaigns produce pallets before the next customer order is ready.
- Waiting: Operators, syrup, QA release and dispatch schedules wait at different points.
- Non-utilised talent: Operators spend time searching, resetting and escalating recurring faults rather than improving standards.
- Transportation: Pallets and packaging materials travel between staging areas without improving the product.
- Inventory: Materials, WIP and finished goods conceal flow problems.
- Motion: Changeover teams search for parts, tools and specifications.
- Extra-processing: Repeated checks, manual reconciliation and rework compensate for unstable process conditions.
The map also highlights Y = f(x) in practical terms: output quality and throughput are functions of inputs such as carbonation pressure, cap torque, filler speed, label settings, material availability and changeover discipline.
Future-state design: make the filler the pacemaker
The future state should not simply remove inventory. It should create controlled flow around the constraint.
Recommended future-state design:
- Schedule the filler as the pacemaker at approximately 520 bottles per minute, above takt but below a speed that creates instability.
- Sequence products to reduce the number of format and recipe changes.
- Create a small FIFO lane between syrup preparation and the filler, with a defined maximum and minimum.
- Apply SMED to reduce changeover time from 55 to 15 minutes.
- Install automatic downtime capture with standard reason codes.
- Use Andon signalling for filler, capper, material and quality conditions requiring immediate support.
- Establish a controlled finished-goods supermarket linked to dispatch demand.
- Use standard work, visual checks and daily review of OEE, yield, waiting and energy.
Current versus future performance
| Metric | Current state | 90-day future state |
|---|---|---|
| Lead time | 9.4 days | 2.8 days |
| Filler OEE | 62% | 78% |
| Total scrap | 1.9% | 0.8% |
| Format changeover | 55 min | 15 min |
| WIP and finished-goods inventory | 1,120 pallet equivalents | 380 pallet equivalents |
| On-time in-full delivery | 91% | 98% |
| Energy per 1,000 bottles | 18.5 kWh | 16.2 kWh |
| Flow efficiency | 0.16% | Approximately 0.5–0.7% |
The future state does not make value-added time dramatically larger. It reduces the time surrounding the work, which is the central value of value stream mapping.
90-day Kaizen sequencing and weekly cadence
Days 1–14: Define and measure
- Owner: Operations Manager and Lean Six Sigma Black Belt
- Confirm product family, customer demand, takt time and project charter.
- Validate OEE definitions, scrap counts, WIP quantities and lead-time assumptions.
- Capture filler micro-stops at one-second resolution where practical.
Days 15–28: Analyse the constraint
- Owner: Black Belt, Maintenance Engineer and QA Lead
- Complete the downtime Pareto.
- Use a process capability review for fill volume, cap torque and label position.
- Review variation, common causes and special causes by SKU and shift.
- Confirm whether the filler remains the primary Theory of Constraints constraint.
Days 29–49: Execute SMED
- Owner: Production Supervisor and Changeover Team
- Film and time the current changeover.
- Separate internal from external tasks.
- Pre-stage parts, tools, recipes and QA forms.
- Run three controlled trials and verify the 15-minute target.
Days 50–70: Stabilise micro-stops
- Owner: Maintenance Engineer and Line Team
- Attack the top three micro-stop causes.
- Tune guide rails, cap delivery, back-pressure, sensors and buffer logic.
- Introduce Andon escalation and a daily repeat-fault review.
- Track first-pass yield and filler speed against takt.
Days 71–84: Build the future-state flow
- Owner: Planning Manager, Warehouse Lead and Operations Manager
- Set FIFO limits and supermarket quantities.
- Reduce finished-goods staging and release queues.
- Align production scheduling with customer demand and product sequence.
- Validate the future-state value stream map at the gemba.
Days 85–90: Control and sustain
- Owner: Black Belt and Site Leadership
- Lock standard work, visual boards and escalation rules.
- Audit changeover adherence, OEE, scrap, energy and OTIF weekly.
- Transfer ownership to the line team and schedule a 30-day control review.
The weekly cadence should remain simple: a Monday 20-minute tier review, a Wednesday gemba walk and a Friday data review. This rhythm keeps improvement connected to production rather than isolated in a project folder.
Turn bottling-line data into improvement capability
A well-built value stream map is more than a drawing. It is a decision system that connects customer demand, process data, bottleneck management, waste reduction and governance. To fully appreciate its impact, the team must learn how to measure the stream, test root causes and control the gains after the workshop ends.
Lean 6 Sigma Hub provides CSSC-accredited, self-paced online training from White Belt through Black Belt, with practical simulations, worked examples, charts, templates and end-to-end DMAIC case studies. Explore the Lean Six Sigma online training programmes, review the DMAIC learning pathway, and build the capability to lead measurable improvement in manufacturing, logistics, healthcare, finance or IT.
Start your Lean Six Sigma certification journey today and learn to convert process data into reliable customer value.
Kaizen. Kai-Care. Kai-Done. ( Lean Six Sigma)
