INVInventoryStop 9 of 11
Replenishment & deployment (DRP)
When each warehouse should be refilled and by how much, and where available stock goes when there isn't enough.
Read atChange any time
DRP (distribution requirements planning) is MRP for the distribution network. For each warehouse and each item, week by week, it works out what is needed:
forecast and customer orders + safety stock − stock on hand − deliveries already on the way = net requirement
Net requirements are rounded to sensible shipment sizes and moved earlier by the transit time, giving planned replenishment orders. Those orders become the demand on the warehouse or plant upstream. That is how a store’s forecast ripples back to the factory.
Deployment is the step that follows when supply is limited: the plant has made a certain amount, so where should it go? Say the plant has 50,000 units and three warehouses ask for 60,000 between them: Mumbai 20,000, Pune 15,000 and Delhi 25,000.
- A fair-share rule gives everyone the same fraction of what they asked for. Here that is 50/60, about 83%: Mumbai about 16,700, Pune 12,500, Delhi about 20,800.
- A priority rule serves key customers or the highest margins first.
- An optimized deployment weighs service targets, stock already in each region, transport cost and how soon more supply arrives.
Words to use
- Net requirement
- What a location still needs after counting stock on hand and deliveries already scheduled.
- Scheduled receipt
- An order already placed and on its way, counted as future supply.
- Fair share
- A deployment rule that gives each location the same percentage of its requirement when supply is short.
The DRP record
For each item at each warehouse, DRP keeps a weekly table:
- Gross requirements: forecast plus confirmed customer orders.
- Scheduled receipts: shipments already on the way.
- Projected balance: last week’s balance plus receipts minus requirements.
- Planned receipts: shipments needed to keep the balance above safety stock, rounded to the shipping lot.
- Planned orders: the same shipments moved earlier by the transit time. These become demand on the supplying warehouse or plant.
Forecast 400 a week, safety stock 300, 900 on hand, 500 units already in transit for week 2, shipments in full truckloads of 1,000, and 1 week in transit.
| Week | 1 | 2 | 3 | 4 | 5 | 6 |
|---|---|---|---|---|---|---|
| Gross requirements | 400 | 400 | 400 | 400 | 400 | 400 |
| Scheduled receipts | 500 | |||||
| Planned receipts | 1,000 | 1,000 | ||||
| Projected balance | 500 | 600 | 1,200 | 800 | 400 | 1,000 |
| Planned orders (ship from plant) | 1,000 | 1,000 |
In week 3 the balance would fall to 200, below the safety stock of 300, so a truckload must arrive that week, which means it leaves the plant in week 2. The same happens in week 6.
Add up the planned orders of every warehouse and you get the time-phased demand on the plant: the input to its master production schedule.
Deployment: sharing out what you have
When the plant can’t supply everything, deployment decides who gets what. Two common fair-share rules:
- In proportion to requests, as in the gist: everyone gets the same percentage of what they asked for.
- Equal days of cover, usually better: allocate so every warehouse ends up with the same number of days of demand in stock.
With 50,000 units to share: Mumbai has 5,000 on hand and sells 1,000 a day, Pune has 6,000 and sells 500 a day, and Delhi has 2,000 and sells 1,500 a day. Total stock after shipping is 63,000 against demand of 3,000 a day, so each can have 21 days of cover. Ship 16,000 to Mumbai, 4,500 to Pune and 29,500 to Delhi. Each warehouse would now run out at the same time, which is the fairest outcome while more supply is on its way.
Push, pull and in between
- Pull: each warehouse calculates its own needs and orders from upstream, as DRP does. Simple and local, but each site sees only its own picture.
- Push: a central planner decides what each warehouse gets, based on the whole network. Better when supply is scarce or production comes in large batches that must be spread out.
- Most companies use pull to plan and push to deploy: DRP drives production, and deployment allocates what was actually produced.
Nervousness and stability
Because DRP plans cascade upstream, a small forecast change at one warehouse can ripple into changes at the plant and its suppliers. Planners damp this with firm or frozen periods in the near term, lot sizes that absorb small changes, and by acting only on changes above a threshold. Without these, the plant is asked to reshuffle its schedule every week in response to noise.
Shortage gaming
If warehouses learn that scarce stock is shared in proportion to requests, they inflate their requests. Allocation then rewards whoever exaggerates most, and when supply recovers the inflated orders vanish, leaving excess. This rationing game is one of the four classic causes of the bullwhip effect. The defence is to allocate on actual demand, sales history or days of cover instead of requests, and to share the true supply position openly.
Optimized deployment
Deployment can be set up as a small optimization: choose shipments from plants to warehouses to minimize expected lost sales plus transport cost, within available supply, truck capacities and each site’s target stock. Useful refinements:
- Customer priorities: serve strategic customers or contractual commitments first.
- Expected supply: if a big batch lands next week, ship less now and save transport.
- Lateral moves: sometimes moving stock between warehouses beats waiting for the plant.
Who runs DRP
Retailers often let suppliers run replenishment for their warehouses under vendor-managed inventory (VMI) or continuous replenishment. The supplier sees the customer’s stock and sales and plans the shipments, which removes a layer of ordering and often reduces the bullwhip effect.
Interview kit
Practise it out loud
Answer each question yourself before opening the model answer.
DRP is time-phased replenishment planning for each warehouse: forecast plus safety stock, minus stock on hand and scheduled receipts, gives net requirements, which become planned orders offset by transit time and pushed upstream as demand. Deployment then decides how available supply is shared, by fair share, priority or an optimization of service and cost, when there isn’t enough to meet every request.
How is DRP different from MRP?
The logic is the same: net requirements, lot sizing and lead-time offset. MRP explodes a bill of materials into components for a plant. DRP works across the distribution network, from customer demand at warehouses back to the source of supply.
Supply is 50,000 units and three warehouses ask for 60,000. How would you decide?
Start with fair share, so each gets about 83% of its request. Then adjust for what matters: warehouses with key customers or low stock get priority, and I’d consider transport cost and when the next batch arrives. If the shortage repeats, the real fix is upstream in supply planning.
Connected stops
Where this idea feeds in, or gets its inputs from.
Sources and further reading
- Chopra, S. & Meindl, P. Supply Chain Management: Strategy, Planning, and Operation. Pearson.
- Silver, E. A., Pyke, D. F. & Thomas, D. J. (2017). Inventory and Production Management in Supply Chains (4th ed.). CRC Press.
- Lee, H. L., Padmanabhan, V., & Whang, S. (1997). Information distortion in a supply chain: The bullwhip effect. Management Science, 43(4), 546–558.