Buffer Stock Math for a Site 180 km From the Supplier

Reordering at two pallets left works when delivery takes two days. It fails when the site is 180 km from the depot and the road closes for a week after a typhoon, which is exactly when leachate volume spikes and the trucks cannot get in.

Buffer stock is a calculation, not a feeling. Get the four inputs right and the answer is defensible to a finance manager who would rather not hold inventory at all.

The four inputs

Average consumption per week, in bags. Take the last twelve weeks and use the median rather than the mean, because a single emergency campaign can distort the average by 30 percent.

Consumption variability, expressed as the ratio of the busiest week to the median. On a site with a wet season this can be 2.0 or higher. A site with a covered cell and steady inflow might be 1.2.

Replenishment lead time, in weeks, including the ordering delay. The supplier’s dispatch time is not the lead time; the lead time runs from the moment someone notices the stock is low to the moment bags are on the rack.

Lead time variability. A two-day nominal delivery that is sometimes five days has a variability of three days, and that is what the buffer has to cover.

A worked example

Median consumption 40 bags a week. Busiest week ratio 1.8, so peak demand is 72 bags. Lead time 1.5 weeks nominal, worst case 3 weeks. Safety stock should cover the worst-case extra lead time at peak demand: (3 – 1.5) x 72 = 108 bags.

Add cycle stock for the normal ordering period, say one week at median: 40 bags. Reorder point is safety stock plus cycle stock, 148 bags, and the maximum stock level is that plus one order quantity.

That is a lot of bags for a site used to holding 50. The number is uncomfortable, which is why sites under-hold and then pay for emergency freight, which costs more than the carrying cost of the stock by a wide margin. Emergency freight on two pallets can add 40 percent to the material cost and still arrive late.

Where to hold it, given that it absorbs water

Superabsorbent polymer does not expire, but it does take up moisture, and a pallet that has sat through two humid seasons in an unsealed container will feed badly and cake in the bag. Buffer stock and storage conditions are the same problem.

Hold stock in a sealed container or a lined, ventilated shed, on pallets, off the floor, with the oldest bags at the front. Rotate: use buffer stock first and replenish behind it, so nothing sits for two years. Write the receipt date on each bag with a marker; it takes seconds and prevents the situation where the emergency stock is the unusable stock.

If the site has no dry storage, hold less and pay for faster delivery. That is a legitimate trade, but it should be a decision rather than an accident of shed availability.

The failure that looks like a supply problem

A site in a coastal province ran out during a rain event and concluded the supplier had let them down. The delivery had been dispatched on time; the truck could not reach the gate because the access road was flooded. The real failure was that the reorder point had been set on consumption, not on the combination of consumption and access risk.

The fix was not more stock in the main shed, which flooded anyway. It was a satellite pallet of 20 bags in a raised container at the weighbridge, above the flood line, plus a written rule that it is only opened when the main store is unreachable. That is cheap resilience, and it is invisible until the day it matters.

Review the numbers once a year, and after any month where consumption exceeded the median by more than 50 percent. Where a site is choosing between two grades at similar cost, the one with better moisture tolerance in storage reduces the practical buffer requirement, and the storage and shelf-life notes cover what each grade tolerates.