Inventory Optimization: EOQ, Forecasting, and ABC Analysis

Hold too little stock: stockouts, lost sales, angry customers. Hold too much stock: tied-up cash, storage costs, obsolescence, shrinkage. The sweet spot is precise — order the right amount at the right time. This is inventory optimization.

The question: How many units of Product X should I order, and how often?

Traditional answer: Guess based on experience. Result: Either too much or too little (usually both at different products).

Data-driven answer: Use three techniques: EOQ (how much), ABC analysis (which products to focus on), forecasting (when to reorder).

EOQ balances two costs:

• Ordering cost — Every order costs (shipping, admin time, handling fees). Order more frequently = higher total ordering cost.
• Holding cost — Inventory sitting in stock costs (storage space, insurance, obsolescence risk, capital tied up). Order large quantities = higher holding cost.

Formula: EOQ = √(2 × D × S / H)

Where:

• D = Annual demand (units/year)
• S = Cost per order
• H = Holding cost per unit per year

Example:

• Product: Black Hair Soap
• Annual demand: 2,400 units (200/month)
• Supplier cost per order: £50
• Holding cost per unit/year: £1 (storage, insurance, capital)

EOQ = √(2 × 2,400 × 50 / 1) = √(240,000) = ~489 units

Interpretation: Order 489 units at a time. Reorder every ~2.4 months. Result: Minimal total of ordering + holding costs.

Not all products matter equally. Some drive 80% of revenue, others 5%.

ABC Segmentation:

• A products (80% of revenue, ~20% of inventory): Your bestsellers. Focus here. Reorder frequently (lower safety stock). Monitor constantly.
• B products (15% of revenue, ~30% of inventory): Mid-tier. Standard reorder process.
• C products (5% of revenue, ~50% of inventory): Slow movers. Reorder infrequently, minimize safety stock. Consider discontinuing if margins are poor.

Pareto Principle: In most retail, 20% of products drive 80% of revenue. Focusing management effort on A products yields 80% of the benefit.

EOQ tells you how much. Forecasting tells you when. Simple approach: Reorder point = (Average daily sales × Lead time in days) + Safety stock.

Example:

• Product: Black Soap
• Average daily sales: 8 units
• Supplier lead time: 14 days
• Safety stock (buffer): 20 units

Reorder point = (8 × 14) + 20 = 132 units

When your stock reaches 132 units, place the next order. This ensures stock never runs out if demand is average.

Sophistication: Weight recent sales more heavily (they're more predictive), adjust for seasonality (sales spike around holidays), apply trend (sales growing or declining). This is what AI-powered inventory does automatically.