Gains:
- Ability to calculate and interpret reorder point, safety stock and order quantity with artificial intelligence support
- Ability to analyze stockout, excess stock and dead stock scenarios and create a draft action plan
- Understanding that the order quantities suggested by artificial intelligence must be verified with lead time, shelf life and budget constraints
You guessed the demand; So when and how much will you order? Inventory management is the answer to this question. The goal is simple but challenging: to have the right product, in the right quantity, at the right time. Excess cash freezes on the shelf and forces it to be melted at a discount; Too little will miss the sale and push the customer to the competitor. In this unit, you will use artificial intelligence as an assistant that accelerates the calculations behind order decisions and makes scenarios visible.
First three basic concepts:
Reorder point (ROP): This is the threshold at which a new order must be placed when the stock drops to this level. Roughly: (average daily sales × lead time) + safety stock.
Lead time: The time from the moment you place the order to the moment the product arrives on the shelf and is ready for sale.
Safety stock: It is a buffer stock kept against uncertainty in demand and supply time. It prevents stock outs in case of unexpected demand spikes or supply delays.
Why isn't just average enough?
Let's say you sell an average of 10 units per day and the supply is 5 days. Saying "I will order when there are 50 pieces left" seems sufficient. But some days you sell 18, the supplier is one day late. Safety stock covers these surprises. How much safety stock? The more volatile your demand and the more unreliable the supply.
factor
Safety stock when increased
Why
demand volatility
increases
To meet the leaps
lead time
increases
The longer you wait, the greater the risk
Supply reliability
decreases
Supplier arriving on time reduces buffer need
Importance of the product (target product)
increases
The cost of being out of stock is high
Short shelf life
decreases
Excess buffer leads to degradation/waste
Attention: Safety stock is not set up by rote like "fixed 2 weeks for each product". The buffer of the main product that sells quickly and the product that sells once a month cannot be the same.
Order quantity: EOQ logic
EOQ (Economic Order Quantity): It is the ideal order size that balances the cost of ordering (shipping, processing) and the cost of holding (warehouse, capital, spoilage). Too often small order → high transaction cost; infrequent large order → high warehouse and cash cost. Artificial intelligence helps you calculate this balance, but you add the real constraints such as minimum order quantity, box floor, campaign.
Out of stock, overstock and dead stock
- Stockout: No product, customer arrived. It's not just that sale you're losing; Customer loyalty also suffers.
- Excess stock (overstock): Stock above demand; When cash is tied up, discount pressure arises.
- Dead stock: Stock that has not moved for months and has little hope of being sold. The warehouse takes up space and its value decreases.
Artificial intelligence can quickly extract these three situations from sales rate and stock level data and provide you with an action list.
Step by step inventory analysis
- Collect data: SKU, current stock, average daily sales, lead time, minimum order quantity, shelf life.
- Calculate ROP and safety stock: Extract the reorder point for each SKU.
- Mark the risks: Out of stock, overstock, dead stock.
- Get an action draft: Which product to order, which one to discount, which one to return/transfer.
- Verify and decide: Check with budget, supplier constraint and shelf life and confirm the order.
mini cases
Case 1 — Safety stock saves: A pharmacy store sells an average of 30 per day of a popular vitamin, 7 days supply. He was only ordering 210 units on average, but the supplier was 3 days late twice and out of stock. AI analyzes demand volatility and lag history and recommends 90 safety stocks, ROP increases to 300. There is no out of stock in the next delay; ~15,000 TL monthly lost sales are protected.
Case 2 — Dead stock cleanup: In a clothing store, 22 of 140 SKUs have not sold for 90 days, worth 180,000 TL. Artificial intelligence lists them and drafts "If not sold within 40 days, markdown, transfer to outlet at the end of the season". The manager approves the plan; warehouse space and cash are freed up.
Case 3 — Preventing overordering: A grocery store clerk is about to buy a 6-month supply of soda after falling for a supplier promotion. When the artificial intelligence draws a scenario with shelf life (4 months) and monthly sales, it shows that 30% of the stock will exceed its expiration date. The order is reduced from 6 months to 3 months; A possible loss of 40,000 TL is prevented.
Weak prompt / Strong prompt
Weak prompt:
How much should I order?
No data and no restrictions; comes the meaningless answer.
Powerful prompt:
Your role: inventory planning specialist. Data: SKU 2210. Current stock 120. Average daily sales 24, highest in last 30 days 41. Lead time 6 days, supplier is 1 day late on last 3 orders. Minimum order is 50, case multiple is 10.Task: (1) calculate reorder point,(2) suggest and justify reasonable safety stock,(3) should I order now, if so, how many cases?Rule: round up to case multiple, follow minimum order rule.Show the calculation step by step so I can verify.
Copiable prompt templates
1) ROP and safety stock calculation
Calculate the reorder point and safety stock for the following SKU. Use daily sales, volatility, lead time and lag history. Show the calculation step by step. Data: [paste]
2) Stock risk screening
Check out the stock chart below. Three lists are created: (a) those at risk of being out of stock within 15 days, (b) those with excess stock (90+ days of stock), (c) dead stock that has been inactive for 60 days. Why one row for each SKU.Data: [paste]
3) Dead stock action plan
Suggest a disposal plan for the following inactive products: markdown percentage, bundle offer, transfer or return option. Consider shelf life. Give it in a table. Data: [paste]
4) Order scenario comparison
Compare 3 order scenarios for this product: low (1 month), medium (2 months), high (4 months). Make a table of tied cash, warehouse load, out-of-stock risk and shelf life loss risk for each. Data and shelf life: [paste]
Common mistakes
- Equalizing the safety stock to the mean: Volatility and supply reliability must be taken into account.
- Ignoring supplier delay: Past delays should be added to the lead time.
- Bypassing the case floor/minimum order rule: The AI recommendation of "37 pieces" is not actually applicable.
- Forgetting the shelf life: Overordering directly means loss of short-lived products.
- Ignoring dead stock: Freezes cash and space; Scan regularly.
- Falling for a one-time promotion: Buying it because it's cheap, but if it doesn't sell, it will be expensive.
In summary
Good inventory management works with uncertainty, not averaging. Establish reorder point, safety stock, and order quantity along with demand volatility, lead time, and shelf life. AI accelerates calculation and scenarios; But you verify the parcel floor, budget and supplier and confirm the order.
Application task
Prepare an anonymous spreadsheet with current stock, daily sales, lead time, and shelf life for your 10 most critical SKUs. List three risks with the "2) Inventory risk scan" prompt, then run the "1) ROP and safety stock calculation" prompt for the riskiest SKU. Verify the resulting account manually once.
checklist
- [ ] Collected inventory/sales/supply data for critical SKUs.
- [ ] I set up the safety stock based on volatility and lag.
- [ ] I have implemented strict box and minimum order rules.
- [ ] I regularly scan out-of-stock, overstock and dead stock.
- [ ] I took shelf life into consideration in the ordering decision.
- [ ] I manually verified and confirmed the order quantity.