As of 8 November 2026, Woolworths has confirmed that its Sydney Chilled and Fresh Distribution Centre in Western Sydney will open in the new year, following an AUD 900 million investment. The multi-level hub is described as the largest refrigerated distribution centre in the Southern Hemisphere and is sized to move 3000 tonnes of produce a day, with capacity to expand over the next 20 years and 800 new roles attached to the site. For teams that manage shelf availability, the question worth asking is not how big the building is but which constraint it actually removes.
The constraint the site removes
Fresh food rarely runs out because the network has no stock. It runs out because stock is in the wrong place at the wrong moment, and a chilled site that can process 3000 tonnes a day shortens the distance between the distribution layer and the shelf. That only converts into fewer gaps if store ordering and delivery windows are rebuilt around the new rhythm; retailers that add capacity without touching the order cycle usually see service levels plateau within two quarters.
Woolworths frames the project in reliability terms. Chief executive Amanda Bardwell said the facility was built to improve the availability of fresh fruit and vegetables in stores and to keep pace with a growing population over the next two decades. That is a supply-side promise, and the commercial version of it is a fill rate stores can plan promotions against, which is the number an availability team should be asking for in writing.
Three levers between capacity and the shelf
Order cadence at store level
Fresh orders usually follow a fixed daily cycle with manual top-ups. When transit time compresses, the ordering cycle itself becomes the binding constraint. Retailers that shorten order windows capture more of the benefit than those that simply add trucks, because replenishment is then driven by the last day's sales rather than by a forecast written three days earlier.
Range discipline inside the chilled cabinet
Every extra line in a chilled range consumes capacity, cabinet space and labour, and slow movers sit beside fast ones. A site designed for two decades of growth invites category teams to widen assortment, which is precisely the risk. SKU coverage should be signed off against sales density before extensions are approved, not reviewed after the cabinet starts showing holes.
Acting on exceptions rather than watching them
Logistics operators now argue that knowing where a shipment sits has become a commodity, and that the value lies in acting on exceptions across a dozen trading partners. The same shift applies inside a grocery network: the useful signal is not an on-time percentage on a dashboard, but an alert that a named store will run out of a named line before the next delivery arrives.
Reading the numbers together
| Measure | What it captures | Where it misleads |
|---|---|---|
| Case fill rate | Share of ordered cases delivered | A good average can hide store-level imbalance |
| On-shelf availability | Share of time a line is buyable | Ignores phantom stock held out the back |
| Waste as a share of sales | Cost of over-ordering | Rewards under-ordering that creates gaps |
These three measures pull against each other, which is why single-metric targets get gamed. A store judged only on waste will under-order; a store judged only on availability will over-order and let the surplus expire. The workable approach is to publish them as a set, for the same category in the same week, so that a decision improving one at the expense of another becomes visible straight away.
Market context sharpens the stakes. Australian retail spending grew 4.9 per cent in the year to September while food retailing rose 3.3 per cent against food inflation of 3.2 per cent, which leaves volume growth close to flat. When baskets stop expanding, service level becomes the main route to share, and a distribution investment has to earn its return through sales that were not lost rather than through price.
A four-week availability review
The pattern from comparable projects is consistent. Capacity is a precondition, not an outcome. Availability moves when ordering, range and exception handling are redesigned in the same quarter a site opens, and it stalls when the building is treated as the answer instead of the input.
FAQ
Does a bigger distribution centre automatically cut stockouts?
No. It removes a throughput limit, but ordering cadence and range density still decide whether stock reaches the shelf.
Which metric should a fresh team lead with?
On-shelf availability paired with waste, because either one alone can be improved by making the other worse.
Why does phantom stock matter?
Inventory recorded in the system but missing from the shelf suppresses replenishment, so the store looks stocked while shoppers see a gap.
How soon do results show up?
Fill rate usually moves within a quarter, while waste and range productivity take two to three quarters to settle.
