Slot every SKU where velocity, physics, and pick-paths agree.

Four engineering tracks

Each track is a self-contained guide with overview pages, deep-dive subsections, and operational walkthroughs. Open any section to read the full architecture brief, or jump directly into the pages you need.

Start here — hands-on walkthroughs

The fastest way in is a concrete build. These step-by-step guides each take one real slotting problem from raw data to working, vectorised Python you can lift straight into production.

• Slotting Architecture How to Map Warehouse Aisles to Logical Zones A deterministic Python method for translating physical aisle coordinates into velocity-driven logical zones — tolerance buffers, contiguity checks, and a WMS-ready mapping payload. Read the walkthrough
• Slotting Architecture Building a Pick Path Model from Scratch Build a velocity-weighted, constraint-aware pick path model in Python from raw WMS data — graph construction, composite edge weights, an A* solver, and shadow-mode validation. Read the walkthrough
• Slotting Architecture How to Implement Role-Based Access Control for Slotting Configs A deterministic RBAC decorator for slotting configuration writes — role-to-operation matrix, zone-scope checks, TTL-bound overrides, optimistic locking, and pytest validation before anything reaches the solver. Read the walkthrough
• Slotting Architecture How to Implement Exponential Decay Velocity Weighting Weight pick history by recency with half-life exponential decay so a fresh pick outranks a stale one — the w = 0.5**(age/half_life) score, config, verification, and pitfalls. Read the walkthrough
• Slotting Architecture How to Classify SKUs by Inventory Velocity You have a rolling feed of pick transactions and a master SKU list, and you need one deterministic job that turns them into an A/B/C/D/Z velocity tier per SKU — r… Read the walkthrough
• Data & Sync Pipelines How to Configure Async Batch Recalculation for 1M+ SKUs Recalculate velocity for a million-plus SKUs on a nightly budget: chunking, asyncio.Semaphore concurrency bounds, resumable checkpoints, and backpressure — with config and pitfalls. Read the walkthrough
• Data & Sync Pipelines Python Async Batch Jobs for SKU Velocity Tracking A single asyncio job that pulls SKU movement from a WMS under bounded concurrency, validates it, and pushes clean velocity records to the slotting engine — with backoff, semaphores, and a dead-letter path. Read the walkthrough
• Data & Sync Pipelines How to Detect Anomalies in Inventory Feeds Flag anomalous inventory feed runs — row-count drops, null-rate spikes, out-of-range values — with a typed median/IQR and EWMA detector, config, verification, and pitfalls. Read the walkthrough
• Data & Sync Pipelines How to Stream Pick Events with Kafka Consumers Build a typed confluent-kafka consumer that reads pick events, updates an idempotent per-SKU rolling velocity counter, and commits offsets manually with safe rebalance handling. Read the walkthrough
• Data & Sync Pipelines Transforming Legacy ERP Sales Logs for Velocity Targeting Turn fixed-width legacy ERP sales dumps — composite item keys, free-text UOM, mixed encodings — into canonical base-unit velocity facts, with a runnable Python transform and pytest checks. Read the walkthrough
• Data & Sync Pipelines Validating JSON Schemas for Inventory Updates Enforce a Draft 2020-12 JSON Schema on raw inventory-update payloads with Python's jsonschema — a versioned contract, a precompiled validator with RFC 3339 date-time checking, and structured errors that route cleanly to a dead-letter queue. Read the walkthrough
• Data & Sync Pipelines How to Implement Watermark Cursors for Delta Sync Pull only changed rows since the last sync with a high-watermark cursor on updated_at, a safety overlap window for late updates, and durable persist-after-commit ordering. Read the walkthrough
• Data & Sync Pipelines How to Poll a WMS API With No Published Rate Limit Build a self-regulating adaptive poller for WMS/ERP APIs that advertise no rate limit — MIAD interval control from observed latency, request dedup, schema gating, and pytest checks. Read the walkthrough
• Assignment & ABC ABC vs ABCD Velocity Tiers: When to Use Four Bands Choosing between 3-band ABC and 4-band ABCD velocity tiers: what the D band buys you, when four bands pay off, and a config-driven classifier for either. Read the walkthrough
• Assignment & ABC How to Calculate Optimal ABC Thresholds for Seasonal SKUs A config-driven Python method for setting ABC tier thresholds on seasonal SKUs — seasonal-index adjustment, capacity-anchored cutoffs, and hysteresis that stops slotting thrash. Read the walkthrough
• Assignment & ABC How to Group Complementary Products into Adjacent Pick Slots Take known high-lift complementary pairs and greedily place them into adjacent forward-zone bins so multi-SKU orders collapse into a single reach-and-step. Read the walkthrough
• Assignment & ABC How to Mine Order Affinity with Market-Basket Analysis Find SKU pairs frequently ordered together with support, confidence, and lift, then turn high-lift pairs into co-location groups — with a typed Python miner and pytest checks. Read the walkthrough
• Assignment & ABC How to Calculate Re-Slotting ROI and Break-Even Model a SKU move as labor cost versus daily travel savings, compute payback days and net ROI, and emit it only when it pays back inside a threshold. Read the walkthrough
• Assignment & ABC How to Schedule Re-Slotting Windows to Minimize Disruption Bin approved SKU moves into low-activity windows under a per-window budget, order every move vacate-before-fill, and cap churn in congested aisles. Read the walkthrough
• Assignment & ABC How to Solve Slot Assignment with Linear Programming Solve velocity-weighted slot assignment exactly with scipy linear_sum_assignment: build the cost matrix, run the Hungarian solver, decode and validate slots. Read the walkthrough
• Assignment & ABC How to Assign Heavy SKUs to Ground-Level Slots Route heavy SKUs to low rack levels for safety and stacking, enforce a weight-by-level ceiling, and reconcile with velocity — with a typed Python slot ranker and pytest checks. Read the walkthrough
• Assignment & ABC How to Model Golden-Zone Ergonomic Limits Encode the knee-to-shoulder golden zone as slotting constraints so high-velocity and heavy SKUs land in low-strain bins — with a typed Python ergonomic scorer and pytest checks. Read the walkthrough
• Pick-Path & Space How to Detect Aisle Congestion from Pick-Density Heatmaps Bin pick events by aisle and time interval into a density heatmap in Python, then flag congested cells on a picks-per-interval threshold with a picker floor. Read the walkthrough
• Pick-Path & Space How to Score Cube Utilization Per Location Compute a location's cube utilization in Python as occupied cube over usable cube with a packing-efficiency factor, plus config, verification, and pitfalls. Read the walkthrough
• Pick-Path & Space How to Sequence Pick Waves to Balance Labor Assign proximity batches to time-boxed pick waves with a greedy longest-processing-time algorithm that levels picker-hours per wave under hard cutoff times. Read the walkthrough
• Pick-Path & Space How to Build a Travel-Time Distance Matrix for a Warehouse Precompute an all-pairs travel-time matrix over a warehouse aisle graph with networkx and floyd_warshall, cache it to disk, and use it to score pick routes as a cheap lookup. Read the walkthrough

Focus areas

SKU velocity tracking · storage location assignment · pick path optimisation · space utilisation scoring · WMS integration · batch recalculation · reporting.