Choosing a Planning Bucket: Day, Week, or Period in EDGEBIC

EDGEBIC plans demand and supply in buckets, and it offers three sizes: a day, a week, or a 28-day period. EDGEBIC by User Solutions lets you pick the granularity when you view the inventory calendar or the master production schedule, and every forecast, order, and receipt lands in whichever bucket contains its date. The choice is not cosmetic. A bucket that is too coarse hides short stockouts; a bucket that is too fine buries a long-lead item in noise. This post explains how each size behaves and how to pick the right one for a product.

Bucket size is one of the quiet decisions that shapes how useful a plan feels. Get it right and the calendar shows exactly the timing precision the item needs; get it wrong and you either miss a dip or drown in rows. For the wider planning layer, see the inventory and planning pillar; for the generic concept, see what a master production schedule bucket is.

The Three Bucket Sizes

EDGEBIC's three granularities are simple spans.

Each row in the inventory calendar or MPS grid is one bucket for one product. The projected available balance is computed once per bucket: the previous bucket's balance, plus receipts landing in the bucket, minus the bucket's gross requirements. That single fact drives the whole trade-off, because a balance computed once per 28-day period cannot show a trough that happens on day 10 of that period.

What Finer Buckets Buy You

Day buckets give the most timing precision. Every day is its own row with its own projected balance and its own suggestion, so a stockout that happens on a Wednesday shows up as a Wednesday problem. For an item whose stockouts matter within a few days, that precision is the whole point. The cost is row count: a 30-day horizon in Day buckets is 30 rows, which is a lot to scan for a slow item.

Week buckets are the common middle ground. They smooth daily noise into a weekly total while still catching timing within a week, which is precise enough for most make-to-stock products. A quarter of horizon fits in about 13 rows, readable at a glance.

What Coarser Buckets Buy You

Period buckets give the longest readable horizon in the fewest rows. A 28-day span per row means a full year fits in 13 rows. For a long-lead or slow-moving item, where demand is lumpy and the exact day of a stockout matters less than the fact that one is coming in the next month or two, Period buckets keep the whole horizon on one screen.

The cost is timing precision. Because the projected balance is computed once for the whole 28-day span, a dip that happens mid-period never appears as its own row. If an item can stock out within the period, a Period bucket will average right over the trough and show a healthy balance while the real stock briefly hits zero. That is the classic failure mode of coarse bucketing, and it is why fast items should not use Period buckets.

The Trade-Off in One View

The whole decision reduces to a single trade between timing precision and horizon length.

A Worked Example: The Same Product Two Ways

Take a consumable with 100 units on hand and demand of about 15 units a day for the next four weeks, roughly 420 units. Suppose a large one-time order pulls 90 units on day 8.

In Period buckets, the whole 28 days is one row. The projected balance is computed once: 100 opening, minus 420 total demand, showing a deep negative for the period. That tells you the month is short, but it hides that stock actually crosses zero around day 7, before the big order even lands. You would size a build for the month but might release it too late.

In Day buckets, day 7 shows a projected balance dipping toward zero and day 8 shows it going sharply negative as the 90-unit order pulls. The calendar flags a stockout on day 7, days ahead of where the Period view would have alarmed. You release a build in time. Same product, same demand, two very different pictures, decided entirely by bucket size.

Choosing the Bucket for a Product

Ask one question: how precisely do this item's stockouts need to be timed?

• If a stockout that hurts can happen within a few days, use Day or Week buckets so the dip shows as its own row. Fast consumables, short-lead parts, and anything with spiky demand belong here.
• If demand is steady and lead times are long, use Week or Period buckets to keep a long horizon readable. Long-lead castings, slow catalog items, and stable base-load products belong here.

Because you choose the granularity at view time, you can plan the same catalog at different resolutions: Week for the everyday items, Day for the two SKUs that keep surprising you, Period for the long-lead specials. The change is a view setting, not a data change.

Bucket Size and the Suggestion Quantity

Bucket size does not only change what you see; it changes the size of the builds the system suggests, because a suggestion is computed per bucket. In Day buckets, a shortfall is calculated for a single day, so suggestions tend to be smaller and more frequent. In Period buckets, a shortfall spans 28 days of demand, so a single suggestion tends to be larger and covers a longer stretch. Neither is wrong, but they suit different replenishment styles.

If your shop prefers frequent small builds, perhaps because a work center changes over quickly and short runs are cheap, finer buckets align the suggestions with that style. If your shop prefers fewer, larger builds, perhaps because setup is expensive and you want to run a month of demand at once, a Period bucket naturally batches the demand into one larger suggested quantity. The lot-size rule interacts with this: a period-order-quantity rule is only meaningful on a period-based calendar, where it batches a whole period into one order. Choosing the bucket is therefore partly a choice about how often you want to build.

The one caution is to keep the bucket consistent with the item's economics and its stockout timing at the same time. A fast item with expensive setup is a genuine tension: fine buckets catch its stockouts but suggest small frequent builds, while coarse buckets batch the builds but hide the stockouts. For those items, Week buckets are often the compromise that keeps timing visible while still batching demand into runs worth setting up for.

What the Bucket Change Does Not Do

One behavior to keep straight: changing the calendar's granularity does not re-bucket your stored forecast entries. Each forecast row keeps its own granularity and date, and the projection places it in whichever displayed bucket contains that date. So you can hold daily forecast rows and weekly forecast rows for the same product at once, and both are included in the calculation for the bucket their dates fall in. Switching the view from Week to Day does not lose or duplicate any forecast; it just displays the same underlying demand at a finer resolution.

One nuance for Period buckets: a Period always spans a fixed 28-day window anchored to a start, so a date entered mid-period is mapped to that period's range and the displayed start may differ from the exact day you typed. The quantity still counts in the right period; only the shown start reflects the bucket boundary. If you need a demand pinned to a specific day, enter it in a Day bucket. To change a product's bucket on the grid, see how to change the MPS bucket size.

The Payoff

Bucket size is a small lever with a real effect on how a plan reads. Match the granularity to how tightly an item's stockouts need to be timed, and the calendar shows exactly the precision you need with no wasted rows. Mismatch it, and you either miss a dip or scroll through noise. Because it is a view setting, the cost of getting it right is one dropdown, and the payoff is a plan you can trust to alarm at the right moment.

To see the balance that bucket size shapes, read reading projected available balance across the horizon. For how the horizon length interacts with bucket count, see how to set a planning horizon.