Supply Chain & Inventory Analytics

Inventory turnover rate by SKU is one of the cleanest tests of whether your catalog is converting working capital into gross margin at the right pace. A SKU that sells quickly in one region can sit in another warehouse, and a single blended turn ratio at brand level hides the SKU-location pairs that are quietly tying up cash.

FireAI builds d2c supply chain analytics views that combine outbound velocity from your OMS with stock, open POs, and landed cost so you can rank SKUs by turns, days on hand, and cash trapped. SKU-level turnover rollups compare the same product across channels and warehouses without manual joins.

What FireAI delivers for SKU-level turnover:

• Turns and days-on-hand by SKU, warehouse, and channel, using consistent cost and quantity definitions
• Slow-turn and fast-turn outliers within each category so merchandising and ops see the same priorities
• Correlation between promotional calendar and turn changes, so you can separate demand lift from margin-funded overstock
• Recommended review lists: SKUs whose turns dropped more than a threshold versus trailing periods

Real example: An apparel D2C brand with three fulfillment nodes found that 23 SKUs accounted for 31% of inventory value but only 9% of trailing-90-day revenue. Moving clearance and bundle rules for 14 of those SKUs and reallocating inbound POs for 6 others improved blended inventory turns from 3.1x to 3.8x in one quarter without a site-wide sale.

FireAI natural language queries:

• "Rank active SKUs by inventory turns in the last 90 days, split by warehouse"
• "Which SKUs have days on hand above 120 but below category median velocity?"
• "Show SKUs where turns dropped more than 20% quarter over quarter"

Dead stock flagging for D2C brands must connect sales silence to cash and shelf life, not just a static "no sales in N days" report. Liquidation only works when alerts arrive with enough context: stranded quantity by node, landed value, margin left after markdown, and channel rules that affect cleanup.

FireAI flags dead and near-dead SKUs using velocity, returns, and inbound timing together. Merch sees dashboards with severity bands and rupee exposure; finance sees the same numbers for write-off risk and working capital release.

What FireAI includes in dead stock workflows:

• Configurable zero-velocity windows and minimum on-hand thresholds by category or seasonality
• Liquidation paths: site sale, bundle, marketplace clearance, B2B jobber, or donate with tax treatment flags where your data supports it
• Alerts when inbound POs would add to already dead positions
• Batch or expiry overlays when your catalog includes constrained shelf life

Real example: A home and kitchen brand had ₹26 lakh in dead stock spread across four nodes. FireAI grouped SKUs into liquidation tiers: immediate flash for high-holding-cost lines, bundle-with-hero for mid-value slow movers, and transfer-first for SKUs with demand in alternate regions. Executed over 55 days, net recovery was 58% of carrying value versus a historical 40% on ad hoc markdowns.

FireAI natural language queries:

• "List SKUs with zero sales in 75 days and stock value above ₹2 lakh"
• "Which dead SKUs have inbound POs arriving in the next 45 days?"
• "Show liquidation priority by cash recovery potential this month"

Supplier lead time variance is often the hidden driver behind stockouts, emergency transfers, and inflated safety stock. If your planning system assumes a static 21-day lead time but one vendor cluster averages 34 days in peak season, your turns and fill rates will drift together until a fire drill forces expensive fixes.

FireAI tracks supplier lead time from PO issue to inbound GRN, split by vendor, SKU family, and season. Rolling averages, percentile bands, and late-shipment rates give procurement and planning a shared baseline for MOQ and buffer decisions.

What FireAI captures for inbound reliability:

• Acknowledged vs actual lead time by vendor and lane
• Volatility indices so safety stock can scale with observed variance, not guesswork
• Late arrivals correlated with fulfillment SLA or split-shipment incidents downstream
• Vendor scorecards that combine lead time, quality holds, and fill rate in one view

Real example: A nutrition brand saw elevated stockouts on two hero SKUs. FireAI showed that one contract manufacturer's confirmed-to-receipt lane had stretched from 19 to 31 median days over six weeks while other vendors stayed stable. The team split the next PO across two vendors and raised a temporary buffer at the primary node until the slow vendor recovered, avoiding a projected ₹9.4 lakh revenue-at-risk window.

FireAI natural language queries:

• "Which vendors exceeded promised lead time by more than 5 days last month?"
• "Show median supplier lead time by SKU family vs the same quarter last year"
• "Correlate late GRNs to marketplace stockout incidents in the following week"

Multi-warehouse optimization for D2C is the problem of placing the next unit of inventory where it minimizes stockouts, shipping cost, and split shipments at the same time. Static allocation rules break when demand shifts by region, courier lanes change rates, or marketplace share moves week to week.

FireAI models demand by node, lead time, and outbound cost signals you connect, then surfaces allocation and transfer recommendations your team can accept or tune. Optimization outputs are expressed as rules you can implement in the OMS or as exception queues for the ops lead.

What FireAI optimizes across nodes:

• Forward placement of inbound PO receipts by expected regional demand and marginal shipping saving
• Transfer triggers when imbalance crosses a service or cash threshold
• Split-order probability estimates so you see when duplicate shipments will hit margin
• Simulation of safety stock by node under alternate vendor lead time scenarios

Real example: A beauty brand with Mumbai and Bengaluru nodes ran FireAI allocation recommendations for eight weeks. Split orders fell from 14.2% to 9.1% of D2C shipments while stockouts on top 40 SKUs dropped 22%. Incremental freight savings were approximately ₹3.6 lakh in the period, net of inter-node transfer cost.

FireAI natural language queries:

• "Recommend forward allocation for next week's inbound ASN by SKU and node"
• "Which regional imbalances should trigger a transfer this week?"
• "Estimate split-shipment rate if we close virtual inventory at Node B for two SKUs"