Station 4 of 6 Return and track
Setting Up a Bread Tray Inventory Management System
Contents: 7 sections
Start with counts you can trust, then add technology where the counts show it will pay. Record trays sent out and trays returned for every route and store, set your fleet size from how many days a tray spends in the loop, and flag trays that stay out too long. Barcode scanning adds a record at each handoff; RFID adds automatic counts at fixed read points. Move to each one when your manual counts show where the losses are.
What should a tray inventory system track?
Four numbers, kept current:
- Trays owned. Everything you have bought and not retired with a record.
- Trays available. Clean, inspected and ready to load today.
- Trays out, by location. On trucks, at each store, at each depot or partner site. The store-level figure is a running balance: trays delivered minus trays collected.
- Trays out of service. Waiting for washing, inspection or repair, with the reason.
Behind those numbers sit three events you record every time: a tray sent out, a tray returned, and a tray retired with a reason. If owned minus available, out and out of service does not come to zero, the difference is trays you cannot account for.
How many trays does your fleet need?
Enough to cover every day a tray is away from the loading line, plus your busiest stretch, plus what you expect to lose before new trays arrive, plus a buffer for the days that go worse than average. The first three steps below are a minimum planning model; the fourth shows one conservative way to size a buffer. The formulas are this guide’s heuristics, built for planning, not an established inventory model:
| Step | Formula | Example (illustration) |
|---|---|---|
| 1. Trays in the loop | Trays loaded per day × days in the loop | 500 × 4 = 2,000 |
| 2. Minimum fleet level for your busiest stretch | Highest total of trays loaded over any run of consecutive days as long as the loop | 500 + 650 + 600 + 550 = 2,300 |
| 3. Minimum reorder point | Minimum fleet level + (trays lost per day × supplier lead time in days) | 2,300 + (10 × 28) = 2,580 |
| 4. One conservative buffer example | (Longest loop in your records − average loop, in days) × busiest day's loading + trays usually in quarantine + (longest lead time − usual lead time) × trays lost per day | (5 − 4) × 650 + 100 + (42 − 28) × 10 = 890 |
Sources for the table: formulas are this guide’s; all numbers are illustrations, not benchmarks. With the buffer, the example’s reorder point is 2,580 + 890 = 3,470.
Days in the loop is the average time from loading to being back in clean storage, measured on your own routes: delivery, time at the store, the ride back, washing and inspection. Step 2 works because the trays loaded over the last loop’s worth of days are, roughly, the ones still out; take the highest such total from your loading records for the past year. If you have no records yet, your busiest day’s loading times the days in the loop (650 × 4 = 2,600 in the example) gives a higher, safer figure. Trays lost per day comes from your own loss rate, not from an industry figure.
The minimum model assumes every loop takes the average time and every order arrives on time. Real loops stretch over holiday weekends, at stores that hold trays, when the washer is down or when damaged trays wait in quarantine, and lead times slip. The buffer in step 4 stacks the worst loop, the usual quarantine and the worst lead time on top of each other, which makes it deliberately generous; it is one example of a planning allowance, not a statistical safety stock. If you have daily data, the better method is to calculate safety stock from the observed variability in demand, loop time and lead time at the service level you choose. Add growth you have planned for before a season, not after it.
The levels do different jobs. The fleet level in step 2 is the least number of usable trays you need. The reorder point is when to buy: when your usable trays (owned, minus those out of service) plus the trays already on order fall to that number, place an order, because losses keep coming until the new trays arrive and will bring you down to the fleet level about when they do. Order at least the trays you expect to lose between this order and the next one you plan to place. Recalculate before seasonal peaks, since the busiest day and the loop time both change.
Barcode or RFID?
They differ in how a tray gets counted:
| Method | How a tray is counted | What it needs | Where it fits |
|---|---|---|---|
| Manual count | Someone counts trays and writes the number on the route or stop record | A count sheet or a field on the driver's stop record | Every operation, as the baseline |
| Barcode | Each tray label is scanned, one at a time, in view of the scanner | A durable label on each tray and handheld scanners | Handoffs where someone already handles each tray, such as store stops |
| RFID | Tags are read by radio as trays pass a reader, without scanning each one | A tag on each tray and fixed or handheld readers | High-volume points such as dock doors and wash lines |
Sources for the table: RFID and barcode differences as described by GS1 US, quoted below; the rest is ours.
GS1 US describes the difference this way: “Unlike barcode systems that require line-of-sight scanning and manual user action, RAIN RFID enables automatic, high-speed data capture without direct scanning,” and “Tags can be read even when they are not visible, such as inside containers or stacked items” (GS1 US, RFID). For trays, that means a stack can be counted as it passes a reader instead of tray by tray.
Neither technology works if the label or tag does not survive your washer, your freezer and the handling on your routes. Before a fleet order, tag or label a sample of trays, run them through your real wash cycle and routes for several weeks, and check the read rate at each point you plan to use. Some tray makers offer tracking as an option; ORBIS, for example, lists “Asset tracking systems available” for its NPL660B tray (ORBIS, NPL660B). Ask for quoted prices for tags, readers and software at your fleet size rather than planning on published price ranges.
Which reports find the losses?
Four, each run by route and by store:
- Store balance. Trays delivered minus trays collected, running. A balance that only grows is a store keeping your trays.
- Days since last seen. For each tray with an ID, the days since it last passed a read point. Set the flag at a multiple of your own loop time; a tray well past it is sitting somewhere unread, lost, or damaged and unlogged.
- Loss by route and store. Your loss rate for each, with the same definition every period; the definition, and how to read it with the lag between delivery and pickup, is on the page on tray loss and theft.
- Retirements by reason. Trays taken out of service, and why. A cluster of the same reason on one route or from one purchase lot is a cause to fix.
How do you make the case for tracking?
With your own numbers. The cost of doing nothing is the trays you lose in a year times the price of a replacement tray, plus what shortages cost you: late trucks, substitute packaging, time spent settling disputed counts. Set that against quoted prices for the system you are considering, including labels or tags, readers, software and the staff time to run it. If your manual counts do not yet give you a loss figure you trust, that is the first thing to fix; a business case built on an industry average is a guess.
In what order should you set it up?
- Manual counts at dispatch and return, per route. This is your baseline and costs nothing but discipline.
- Store balances on the driver’s stop record. Trays delivered and trays collected at each stop, recorded separately from product. How pickups, returns and rotation are organized so those counts happen is set out in running tray pickups and returns route by route.
- Barcodes at the handoffs where people already handle each tray.
- RFID readers at the high-volume points, such as dock doors and the wash line, once the counts show where automatic reads would save the most time or catch the most loss.
- Per-tray history. Once every tray has an ID, you can record how many trips it has made and why each tray is retired. Over time that shows whether retirements cluster at a trip count or on a route, which lets you plan replacements instead of reacting to breakages.
If several locations share one tray fleet, every location has to count the same way and read the same IDs; the setup for shared fleets is covered in sharing one tray fleet across locations.
What should you set up this month?
Count trays out and back on every route, and record trays delivered and collected at every store stop. Within a few weeks you will have store balances and a loss figure per route, which tell you where the trays go and whether barcodes or RFID would pay for themselves. Buying tracking technology before you have those numbers means you cannot measure what it fixed.