Reviewed on 9 October 2026. Supply chain optimisation means meeting customer demand at the lowest total cost and working capital, without letting service levels slip. Every improvement involves a trade-off: holding more stock improves availability but ties up cash, and faster freight cuts lead times but costs more. This guide covers the main levers, the metrics that show whether they work, the Indian policy context, and a worked inventory example you can rebuild in a spreadsheet.
The Indian context in three facts
- National Logistics Policy. The Prime Minister launched the National Logistics Policy on 17 September 2022 (PIB release). At the launch, he set the goal of bringing logistics cost from 13 percent to a single-digit number.
- A new cost estimate. In September 2025, the government released an NCAER assessment prepared for DPIIT. It estimated India's logistics costs at about 7.97% of GDP and noted that the commonly cited 13–14% figures came from external studies or partial datasets (PIB release, 20 September 2025).
- International ranking. India ranked 38th of 139 countries in the World Bank's Logistics Performance Index 2023, up six places (PIB release, 26 April 2023).
For professionals, these show where public investment and policy attention are going: multimodal transport, data-driven planning and lower logistics costs.
Keep preparing
Continue with Sarkari Resume Templates₹299 — coaching के एक महीने से काफ़ी सस्ता / far cheaper than a month of coachingThe main optimisation levers
| Lever | What you change | Typical trade-off |
|---|---|---|
| Demand forecasting | Statistical forecasts, sales inputs, promotion calendars | Better accuracy needs clean data and discipline |
| Inventory policy | Order quantities, safety stock, reorder points by item | Higher service levels need more stock |
| Network design | Number and location of warehouses and plants | Fewer sites cut fixed costs but lengthen deliveries |
| Transport | Mode (road, rail, coastal), load consolidation, route planning | Cheaper modes are often slower or less flexible |
| Supplier management | Lead times, reliability, multiple suppliers | Dual sourcing raises resilience and admin effort |
| Warehouse operations | Layout, picking methods, slotting | Automation saves labour but needs capital |
Metrics that tell you if it's working
| Metric | Formula | What it shows |
|---|---|---|
| Inventory turnover | Cost of goods sold ÷ average inventory | How quickly stock converts to sales |
| Days of inventory | 365 ÷ inventory turnover | How many days of stock you hold |
| Fill rate | Units shipped from stock ÷ units ordered | Availability for customers |
| On time in full (OTIF) | Orders delivered on time and complete ÷ total orders | Reliability as customers experience it |
| Forecast error | Average absolute error ÷ average actual demand | How much to trust the plan |
| Freight cost per unit | Total freight cost ÷ units shipped | Transport efficiency |
Track these monthly and per product family. Averages across all items hide the slow movers that tie up cash.
Worked example: order quantity, safety stock and reorder point
The item and numbers below are illustrative.
A distributor sells a spare part with steady demand of 12,000 units a year, over 300 working days, which is 40 units a day. Placing and receiving an order costs ₹1,500, and holding one unit for a year costs ₹40 (capital, storage and handling). The supplier's lead time is 9 working days, and daily demand varies, with a standard deviation of 10 units.
1. Economic order quantity (EOQ). EOQ = √(2 × annual demand × order cost ÷ holding cost) = √(2 × 12,000 × 1,500 ÷ 40) = √9,00,000 ≈ 949 units. That's about 12.6 orders a year.
2. Check the costs. At the exact EOQ of about 948.7 units per order:
- Ordering cost ≈ (12,000 ÷ 948.7) × 1,500 ≈ ₹18,974
- Holding cost ≈ (948.7 ÷ 2) × 40 ≈ ₹18,974
- Total ≈ ₹37,947 a year
At the EOQ, ordering and holding costs are equal. Rounding to a convenient 1,000 units changes the total only slightly, to ₹38,000 (₹18,000 ordering + ₹20,000 holding). Ordering 500 units at a time would cost ₹46,000. The cost curve is flat near the optimum, so practical rounding is fine; big deviations aren't.
3. Safety stock for about 95% cycle service. Using z ≈ 1.65 for a 95% service level, safety stock = z × daily standard deviation × √(lead time) = 1.65 × 10 × √9 = 1.65 × 10 × 3 = 49.5, rounded up to 50 units.
4. Reorder point. Reorder when stock falls to (average daily demand × lead time) + safety stock = 40 × 9 + 50 = 410 units.
This simple model assumes steady demand and a fixed lead time. If your supplier's lead time also varies, safety stock needs to cover that too. Check the result against what actually happened over the last few months before adopting it.
A 30-day improvement plan for one product family
- Week 1: Pull 12 months of demand, stock and freight data. Calculate turnover, fill rate and OTIF.
- Week 2: Sort items by value and by demand variability. Find the ten items with the most excess stock and the ten with the most stockouts.
- Week 3: Recalculate order quantities and reorder points for those twenty items, using the method above.
- Week 4: Agree the new parameters with purchasing and sales. Set a review date and record the baseline metrics.
Building a career in supply chain
Entry roles include demand planner, inventory analyst, procurement executive and logistics coordinator. Employers look for strong spreadsheet skills, comfort with ERP data, and the ability to explain trade-offs to sales and finance colleagues. A small project like the worked example, done on real (and permitted) data with before-and-after metrics, makes stronger interview evidence than a list of tools. For the numerical habits behind good supply chain analysis, see our guide to analytical skills.