In pet food manufacturing, the path from ingredient intake to a palletised bag combines two types of work: a process value stream that transforms ingredients into kibble, and a repackaging and fulfilment stream that gets the approved product ready for distribution. Recipe changes, allergen controls and sanitation checks connect the two. If a changeover is planned poorly, it can affect line availability, product quality, and delivery performance.
Value Stream Mapping (VSM) makes the full flow visible: from material and information movement to queues, changeovers and release decisions. The worked example below uses illustrative figures for one hypothetical mid-size plant. They are not industry benchmarks or product specifications. In a live project, replace them with observed plant data and confirm all food-safety controls with the site’s quality team.
Select one product family, and set the boundary
This map focuses on the plant’s dry extruded kibble family: its highest-volume product family, with the most frequent recipe changeovers and the largest share of scrap. A focused family gives the team a useful baseline without combining different equipment routes.
Set the boundary from ingredient receival and release to palletised finished goods dispatched to a distributor. Map wet pouch production separately: retort processing, cooling and pouch inspection create a different process flow that should not be blended into the dry-kibble map.
Walk the current state from intake to dispatch
Assume this plant supplies 1,200 kg of finished kibble per hour, using 20 kg bags. That equals 60 bags per hour, or one bag each minute. With 60 minutes of available production time per hour:
Takt time = 60 minutes ÷ 60 bags = 1 minute per bag.
The line needs to meet that rhythm over time while respecting batch processing, sanitation, quality checks and equipment constraints.
| Current-state step | Process time per representative batch | Queue before next step | % complete and accurate (C&A) handoff |
|---|---|---|---|
| Receival, identification and intake release | 18 min | 1.5 days | 98% |
| Grinding and milling | 22 min | 1.0 day | 99% |
| Batching and mixing | 26 min | 0.5 day | 97% |
| Preconditioning and extrusion | 34 min | 0.5 day | 96% |
| Drying | 42 min | 0.75 day | 95% |
| Fat and flavour coating | 18 min | 0.25 day | 97% |
| Cooling | 16 min | 0.25 day | 99% |
| Packaging, bagging and palletising | 22 min | 2.0 days | 98% |
| Warehousing and distributor dispatch | 14 min | 4.75 days | 99% |
| Total | 212 min | 11.5 days | – |
The process times total 212 minutes, while queues bring end-to-end lead time to about 11.5 days. In this example, that is approximately 1.3% value-added time when processing time is compared with calendar lead time. The largest queues are before intake release, packaging and dispatch, not necessarily at the equipment with the longest cycle time.
The plant runs representative 2,400 kg batches, averages 24 recipe changeovers per week, and reports 180 minutes per changeover, including a 95-minute allergen washdown. Current line OEE is 63%; first-pass yield (FPY) is 91.5%. About 4.8% of output becomes scrap or rework, with 2.0 percentage points returned as approved regrind and 2.8 points ultimately scrapped. Regrind use must follow product, traceability and food-safety rules.
Extrusion cooks and forms the conditioned mix; drying then brings the product to its approved endpoint. Coating applies fat and flavour, cooling prepares kibble for packing, and packaging controls net weight, seal integrity, coding and pallet identification. Moisture and other release criteria must be product-specific. The NC State Extension overview of extrusion processing explains the technology’s use in food and feed production. For hygiene and manufacturing practice, consult the FEDIAF Guide to Good Practice for the Manufacture of Safe Pet Foods and applicable requirements such as 21 CFR Part 507.
Find the eight wastes in the kibble stream
Use DOWNTIME to discuss waste concretely and without confusing necessary food-safety controls with avoidable delay:
- Defects: Moisture variation triggers rework, while incorrect bag coding can require product segregation.
- Overproduction: Producing slow-moving SKUs ahead of demand adds finished-goods age and storage.
- Waiting: Batches wait for laboratory results, cleaning verification, packaging availability or a dispatch slot.
- Non-utilised talent: Operators’ process knowledge is missed when they are not involved in changeover design.
- Transportation: Ingredients, WIP and finished pallets make avoidable journeys between storage and production.
- Inventory: Excess ingredients, WIP and finished goods conceal imbalances between extrusion and packaging.
- Motion: Teams search for cleaning tools, change parts or approved labels during a recipe transition.
- Extra processing: Repeated cleaning steps, unnecessary handling or avoidable kibble regrind consume capacity without adding customer value.
An allergen washdown is not automatically waste: it may be essential. The improvement opportunity is to plan and standardise the work, reduce avoidable movement, and validate any revised method before adopting it.
Build a future state around safe flow and pull
The future state should preserve product safety while reducing queues and stabilising the path through the line.
- Create a finished-goods supermarket. Set a controlled replenishment quantity for high-runner SKUs, replenished from actual distributor demand. Use clear visual signals, FIFO rules and defined stock limits rather than producing slow SKUs simply to keep equipment busy.
- Group recipes and level the mix. Sequence products using an approved allergen and recipe matrix. Group compatible runs to reduce unnecessary transitions, while maintaining customer requirements and validated sanitation standards.
- Apply SMED to changeovers. Separate external preparation (such as staging tools, parts and documentation) from work that must occur while the line is stopped. Establish a target average of 105 minutes per changeover and reduce frequency from 24 to 15 per week through better scheduling. A proposed washdown-time reduction, from 95 to 60 minutes, is a validation target, not permission to shorten a required clean.
- Control critical process inputs. Add in-line moisture and bulk-density monitoring at suitable process points, with defined response limits and confirmation methods. Track trends by recipe and dryer condition; route out-of-spec product through an approved hold-and-disposition process.
- Manage regrind as a controlled loop. Define eligible material, maximum inclusion, identification, traceability and quality checks. Do not return material to a recipe unless it is authorised by the site’s specifications and food-safety programme.
Compare the current and target state
These future-state figures are 90-day planning targets, not guaranteed results. Validate them through trials and confirm that quality and food-safety performance remain acceptable.
| Measure | Current state | 90-day target |
|---|---|---|
| End-to-end lead time | 11.5 days | 4.5 days |
| Touch/process time | 212 min | 198 min |
| Value-added time as share of lead time | 1.3% | 3.1% |
| First-pass yield | 91.5% | 97.0% |
| OEE | 63% | 76% |
| Average changeover duration | 180 min | 105 min |
| Scrap after approved regrind recovery | 2.8% | 1.5% |
| Inventory turns | 15/year | 24/year |
| Internal WIP waiting days* | 5.25 days | 1.8 days |
*Internal WIP days exclude the 1.5-day intake queue and finished-goods warehousing/dispatch queue. The future lead-time target assumes those queues also fall through planned release and dispatch routines.
Sequence the first 90 days of kaizen
Days 1–30: Establish the baseline. The operations manager leads a cross-functional team with quality, maintenance, planning, warehouse and operators. Validate the map at the floor, confirm how each metric is defined, and stratify waiting, moisture rework, changeover time and scrap by SKU. Expected gain: agreed data and a ranked set of causes, not a promised financial saving.
Days 31–60: Improve changeover readiness. The production supervisor and quality lead pilot external staging, standard work and a recipe-sequencing plan. Quality approves cleaning and verification requirements before trials. Expected gain: progress toward the 105-minute average changeover target and fewer schedule-driven transitions.
Days 61–90: Connect flow to demand. Planning, warehouse and production introduce supermarket rules for selected high-volume SKUs, test WIP limits, and pilot in-line moisture and bulk-density response plans. The plant manager reviews the results against FPY, OEE, inventory and dispatch performance. Expected gain: movement toward the 4.5-day lead-time and 97% FPY targets, subject to verified results.
For practical support in mapping, measuring and improving a complex value stream, explore Lean Six Sigma Hub’s self-paced online Green Belt and Black Belt courses. Both are CSSC-accredited and include practical learning such as real-world simulations and end-to-end DMAIC case studies. Build the capability to lead your next improvement project, pursue Lean Six Sigma certification.
Kaizen. Kai-Care. Kai-Done. Lean Six Sigma
